The Vulnerability Principle

In one of the most under appreciated books in moral philosophy to come out in the past few decades, Protecting the Vulnerable: A Reanalysis of Our Social Responsibilities (Chicago: University of Chicago Press, 1985) , Robert Goodin argued that moral responsibilities, though diverse in many ways, all derive from a common underlying moral principle, which he called the Vulnerability Principle (VP):

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

According to Goodin, when we analyze many commonsense moral intuitions about our moral responsibilities towards others we recognize that what is crucial to them, "is that others are depending on us. They are particularly vulnerable to our actions and choices. That, I argue, is the true source of all the standard special responsibilities that we so readily acknowledge. The same considerations of vulnerability that make our obligations to our families, friends, clients, and compatriots especially strong can also give rise to similar responsibilities toward a much larger group of people who stand in none of the standard relationships to us" (Goodin 1985, 11). He says that this will use the VP to "ratchet up" from our intuitions about special role-related responsibilities to argue that what we normally think of as general moral duties "derive from fundamentally the same sorts of moral considerations" (11). Before summarizing key aspects of Goodin's argument, it might be helpful to define what is meant by vulnerability.

The concept of vulnerability is, essentially, the state of affairs in which a moral patient is in some way susceptible to injury or harm. The most vulnerable people in the world are, for example, refugees who have lost everything; they are without food, shelter. or clean water; children who have lost their parents and are without schools or caregivers; those stricken with natural disasters such as hurricanes, earthquakes, and floods; those who are sick without access to medical care; those who are captives and are at the mercy of others, and in general, anyone who lacks the ability to protect their own most basic interests. The vulnerability principle (VP), calls upon competent and capable moral agents to act so as to avoid placing vulnerable people at risk, and to prevent harm or injury from befalling those who are at risk or are specially vulnerable in some way.

A quote from Goodin serves to clarify this idea further: "It makes perfectly good sense to speak of someone's being vulnerable either to manmade threats or natural ones. Likewise, it makes perfectly good sense to speak of someone's being vulnerable either to harms that come about through others' omissions or neglect or to harms that come about through others' positive actions" (110). His notion of vulnerability is further explained the same page: "This point emerges particularly in relation to such cognate notions as 'helplessness' and 'dependence.' The former is defined as the state of being 'unable to help oneself'; the latter as 'depending upon, being conditioned or subordinate or subject; living at another's cost; reliance, confident trust.' In both these situations, the vulnerability in view is to harms that come about through other people's inactions rather than their actions" (110, note 3). Vulnerability is a dispositional property of things. To be vulnerable is to be susceptible to being harmed. But harmed in what way, and by whom, and under what circumstances?

Philosophers who have analyzed the concept of a disposition have distinguished dispositions that are intrinsic to things from those that extrinsic. An example of an extrinsic disposition is the property of my front door key to unlock my front door. My key has the dispositional property of unlocking only the lock on my door; it does not have the disposition to unlock other doors. Similarly, other dispositional properties such as weight, visibility, recognizability, solubility, and many others are relational in that a complete description requires at least two variables, usually more than two.

We can understand vulnerability, in a general sense, as susceptibility to being harmed. To be harmed is to be made worse off than one was at an earlier time. Moral patients can be harmed either by the direct acts of another or by the omissions of others who fail to intervene so as to protect them from threats that they themselves do not create but which they can prevent or thwart. In both cases, a moral agent who possesses some capacity to affect a vulnerable other's well being acts or refrains from acting so as to bring it about that the vulnerable moral patient who is the object of his moral responsibilities is not made worse off because of the agent's acts or omissions. Moral responsibilities to protect the vulnerable, then, are moral obligations that require moral agents to avoid causing harm and to act so as to prevent harm from coming to moral patients whose well-being they have to power to affect. It is important to see that the idea of vulnerability that Goodin is using is a relational one: "Vulnerability implies that there is some agent (actual or metaphorical) capable of exercising some effective choice...over whether to cause or to avert threatened harm" (112). Similarly for the notion of dependency, "one depends upon someone for something." Goodin explain this as follows:

References to vulnerability imply two other references. One is to what the persons or things are vulnerable. Where do their weaknesses lie? What mechanisms are capable of inflicting harm on them? The other is to whom the persons or things are vulnerable. Who can inflict harms on me? Who can protect me against them? One is alway vulnerable to particular agents with respect to particular sorts of threats....Like the notions of power and freedom, that of vulnerability is inherent object and agent relative. (112)

Rather than a three-place relation, I think it is preferable to think of vulnerability as a four place relationship. The vulnerability relation can be presented in general as having the following four-variable form:

The Vulnerability Relation: A is vulnerable to B because of C with respect to D.

In this formula (A) stands for a moral patient who is the object of a moral agent's (B) moral responsibility. (B) is the subject or bearer of a moral responsibility towards (A). (C) represents some aspect of A's good, well-being, or interest that is at risk or is threatened by B's acts or omissions. (D) stands for some power or capacity that B possesses that allows B to affect A's good, well-being, or interest. (C) is the condition or circumstance that makes A specially vulnerable, and (D) refers to a feature of B's power, capacity, or ability to affect C. I will clarify what is meant by special vulnerability at a later time (See Special Vulnerability).

In thinking of vulnerability as a dispositional and relational notion, Goodin theory resembles the feminist ethics of care developed by philosophers such as Virginia Held who notes that, "It is characteristic of the ethics of care to view persons as relational and as interdependent" (The Ethics of Care: Personal, Political, Global. Oxford: Oxford University Press, 2006, p. 46). Both the ethics of care and the ethics of vulnerability differ from traditional deontological and consequentialist ethical theories which regard moral agents as independent and equal autonomous individuals, and which sees them as competing with other independent individuals for resources and advantages. In contrast, the ethics of care, "conceptualizes persons as deeply affected by, and involved in, relations with others;…it does not assume that relations relevant for morality have been entered into voluntarily by free and equal individuals, as do dominant moral theories. It appreciates as well the values of care between persons of unequal power in unchosen relations such as those between parents and children and between members of social groups of various kinds" (Held, 46).

Goodin's vulnerability principles and Held's ethic of care share more than just this basic similarity, and are in fact, I shall argue, complementary accounts of the kinds of moral responsibilities that arise as the result of relationships characterized by vulnerability and dependence. The apprehension of the vulnerability of others induces the moral response of care in socially responsible moral agents. Held tends to see the vulnerability relationship from the point of view of the caregiver who responds to the vulnerability of others, while Goodin tends sees it from the point of view of the vulnerable others who deserves to be cared for. But it is possible, and indeed necessary, to see it both ways.
I think it possible to combine Goodin's VP and Held's ethics of care into a general approach to normative ethical theory that I will sometimes refer to as the Ethics of Vulnerability and Care. At a later stage in my argument I shall suggest some important modifications in the way Goodin's Vulnerability Principle (VP) is framed and combine it with some insights Held and others into a single general ethical principle, what I will call the Vulnerability-Care Principle (VCP).

Goodin's version of the VP has certain theoretical limitations; it is designed to explain what are called "special obligations" or "special responsibilities", but I want to use it as the basis of a general theory of moral responsibilities and as a fundamental principle of a global ethics. In order to extend the VP in this way I will need to clarify what is meant by the notion of moral responsibility. I will also need to define and explain the concept of moral status, which will be used to specify what sorts of things can count as moral agents and moral patients within the vulnerability relationship.

But before turning to these tasks I need to say more about the notions of care and vulnerability.

Special Vulnerability

It is necessary now to clarify what is meant by the notion of special vulnerability. Moral patients are, in general, always vulnerable in the sense that they can be harmed. However, there are certain kinds of circumstances or conditions in which moral patients are actually under threat of being harmed; in these cases we can say that they are specially vulnerable.

Many of the standard ethical thought experiments that philosophers use to elicit strong moral intuitions are ones that employ some kind of threat that makes the moral patient seems specially vulnerable. For instance, in the ever popular trolley examples, the scenario always involves some people who are tied to trolley tracks and are in danger of being run over by a trolley. Individuals who are tied to trolley tracks are unable to flee to avoid the on-rushing trolley and so are specially vulnerable to being harmed or killed. It is the present danger that makes this kind of case so compelling as an example for the duty to rescue. One has a completely different response if the stipulation that they are tied to the tracks is omitted. Suppose that there are five people who are merely standing in the path of the trolley, but who are perfectly capable of moving off the tracks as the trolley approaches. Suppose on the other track there is a person who is lying unconscious across the track. In this case, I would not suppose that many people would opt for directing the trolley towards the one who is specially vulnerable -- the unconscious one. It would be assumed that the five others can protect themselves from the trolley by merely stepping to one side of the tracks, and so the moral agent should send the trolley toward them on the assumption that it would be preferable to risk hitting five people who can avoid being hit than it would be to surely kill the one specially vulnerable individual. The special vulnerability of the people on the tracks is a feature of this familiar case that is rarely remarked upon by commentators.

Or consider Peter Singer's famous case of the small child drowning in a shallow lake. The child in this scenario is specially vulnerable in the way in which, for instance, another child walking quietly on the shore beside the lake is not. One would not, I expect, be inclined to think that one has a special responsibility to rush over to the later child and warn him not to go into the lake where he might drown. He might well be equally vulnerable to drowning, in the general sense, but because he is not immediately threatened with drowning the special responsibility to protect the vulnerable is not triggered where it would be in the case of the child who is actually floundering helplessly in the water. Again, the scenario works to evoke the moral intuition that bystanders have a special responsibility to rescue the floundering child precisely because she is in a circumstance which makes her specially vulnerable.

There are other famous philosophical arguments that trade on the special vulnerability of a moral patient. Hobbes, for instance, in describing the "state of nature" makes it abundantly clear that persons in the state of nature are vulnerable to a great many harms, such as being killed in their sleep. But in discussing the right to life, he stipulates that it comes into play in cases where one's own life is actually being threatened. In such cases, he argued, we have a right of self-defense that allows us to act so as to protect our own lives, even if this means killing an aggressor. Indeed, for Hobbes, the right to life is the most fundamental and the only natural right we have, and we do not lose it even when we enter into civil society and live under the rule of a sovereign monarch; we would still in his view have the natural right to defend our own lives when mortally threatened by the King.

As Goodin explains it:

According to the Oxford English Dictionary, something is "vulnerable" if it "may be wounded," either literally or figuratively; it is "susceptible of injury, not proof against weapon, criticism, etc." Essentially, then the principle of protecting the vulnerable amounts to an injunction to prevent harms from befalling people. Conceptually, "vulnerability" is essentially a matter of being under threat of harm; therefore, protecting the vulnerable is primarily a matter of forestalling threatened harms. (Goodin, 1985, p. 110, italics added)

In the Vulnerability Relation, the C term, is the condition or circumstance that constitutes the threat which makes a moral patient specially vulnerable. Goodin mentions infants and young children as classes of moral patients who are specially vulnerable (p .33). Also the mentally and physically handicapped, the poor, the aged and infirm (p. 34), terminally-ill cancer patients (38), American Indian tribes (40), refugees and stateless persons (168), and he suggests in passing, animals and future generations (169). I will develop and defend this suggestion at a later stage of the argument, but for the time being, it should be clear that when I use the term "vulnerability" I will mean actually being under threat of harm, rather than just the abstract possibility of being harmed.

Given this meaning of vulnerability we should be able to say things like: members of an endangered species a vulnerable, while those of a non-endangered species are not. Prisoners as a class are vulnerable, while those not imprisoned are not. What makes prisoners specially vulnerable is the fact that their condition of incarceration removes the possibility of their defending themselves from threats by fleeing. This is one of the feature that make the practice of torture so morally horrendous -- the person who is being tortured is typically a captive who has no means of escape nor any means of defending himself from assaults upon his person. It is the deliberate infliction of pain upon a person who is specially vulnerable that make torture so appalling. As we shall see, many of the things that we call "human rights" are designed to forestall threats of just this kind, that is, threats upon persons who are, for some reason or another, specially vulnerable. Indeed, oppressed persons generally are an important class of specially vulnerable moral patients. If oppressed persons, as a class of moral patients, are specially vulnerable, and if the VP is a fundamental principle of ethics, then it follows that those moral agents who have the ability to protect the oppressed and prevent them from being harmed, have the moral responsibility to do so.

Under the VP moral agents who have some ability D to forestall or prevent oppressed persons from being harmed have a special responsibility towards them. This is, I believe, the ethical basis of the responsibility to protect that has recently been developed to articulate the requirement to aid those peoples threatened with genocide or ethnic cleansing. Bystander nations have the special responsibility to protect this class of specially vulnerable moral patients. As Goodin, insists, "What the vulnerability model emphasizes is not just their special need, ... but also your special ability to help. That is the crucial factor in imposing the duty upon you in particular" (p .34). This special ability to help is the D factor in the Vulnerability Relation. Moral agents who are incapable of helping the vulnerable can be excused from their moral responsibility to protect the vulnerable, other things being equal. But those who are able to help, and who have no other legitimate excuse, acquire an actual moral obligation to act so as to protect the vulnerable.

We can say that members of the class of moral agents who possess D occupy the role of a potential rescuers. To illustrate this notion, suppose that a fellow airplane passenger suddenly stops breathing while in flight. The persons who are that individual's potential rescuers include those trained in CPR or who have special medical knowledge that might enable them to render assistance. Another passenger, for instance, a philosopher who does not know how to perform CPR, does not have the relevant D factor in this case. The responsibility to attempt to resuscitate the stricken passenger falls more heavily on those who have the relevant D, than it would on someone who lacks that critical skill. The class of potential rescuers might contain the flight attendants, a physician or nurse who is on the plane, or another bystander who has been trained in CPR. While the incompetent philosopher also has a prima facie moral responsibility to rescue the stricken passenger, he may be excused from actually trying to help on grounds of his incompetence, particularly, if there are other, competent moral agents present who rightfully can fulfill the role of potential rescuers because of some special training, skill, or ability they possess.

Some people might find it odd to speak about roles in this context. We normally think of roles as being defined by social conventions, e.g., the roles of doctor, lawyer, teacher, parent, and so on. However, as Goodin suggests, and I will later argue, not all roles are conventionally defined; there are some moral obligations that are brought into being between moral patients and moral agents because of the Vulnerability Relation itself -- that is, because of their special vulnerability and your special ability to help. It is possible to think of these sorts of moral obligations as natural duties, where the term "natural" here should be understood as opposed to "conventional." The responsibility to protect the vulnerable, in cases where one has the ability to forestall an actual threat to their survival, well-being, or freedom, is an example of such a natural duty.

If there is such a natural duty to protect the vulnerable, then the vulnerable have the basis of a moral claim to social protection, that is, they have a right to it. In this way, natural rights (or entitlements) can be derived from natural duties. The VP and the Vulnerability Relation, then, might plausibly provide an account of rights that does not depend on seeing them as either transcendental "God-given" moral properties, nor as mere social conventions. Natural rights, on this view, are derived from the natural responsibility of the able to protect the vulnerable.

The notion of a natural duty here is meant to indicate that the Vulnerability Relation and the Vulnerability Principle express a kind of theoretically primitive notion of moral obligation, primitive in the sense that it is not derived from any other moral fundamental moral notions or principles. It expresses a moral axiom that can be used as the basis for developing an ethics of global responsibility.

Filial Responsibilities

One might have a filial responsibility derived from one's role as adult child to care for one's aged and infirm parents, and to do things for them like help them do their shopping. Such filial responsibilities create moral obligations which exist whether or not the agent has explicitly made a promise to the parent to, say, take her shopping on a particular day. These sorts of special responsibilities, filial obligations, are associated with the role occupied by the agent, being someone’s adult child, and one does not choose to enter this role.

An adult child with an aged and vulnerable mother may, of course, also explicitly promise to take their morther shopping on such and such a day. Here in the act of making a promise one has both explicitly acknowledged one's special moral responsibility, which pre‑existed the act of promising, and creates a right on the part of the parent under which she can demand that you fulfill what you have promised to do. The act of promising, in this case, also functions as an ascription of responsibility to a particular agent at a particular time, and this might be needed, for instance, in cases where there is more than one child who shares a filial responsibility towards the parent. The voluntaristic aspect in such cases concerns only the ascription of responsibility (i.e. volunteering to assume a shared responsibility on a particular occasion), and the creation of a correlative right. However, the content of the responsibility itself, the filial obligation to render assistance and care to aged parent who is in need of it, derives not from the promise, but from the vulnerability of the parent and the child's capacity to satisfy her needs, and perhaps also in this case from the notion of gratitude and reciprocity, (one can legitimately question whether an adult who was abused by his parent as a child owes that parent any special filial obligations).

The moral claim of the parent on the child's assistance pre‑exists the act of promising, so that act does not create the obligation, but only acknowledges and specifies it. If I make a promise to my mother to take her shopping, then if I fail to do so, I have both wronged her and have acted wrongly. However, I may still act wrongly if I fail to do this for her, even though I have not explicitly promised her that I would. I may act wrongly towards her if I understand that she is vulnerable to being stranded in her apartment because she is fearful of going out alone, and that she has no one else to turn to who will escort her to the department store. Under such circumstances, the adult child of an aged and infirm parent would have a filial responsibility towards her mother, even when she does not explicitly consent to it. The act of explicitly making a promise discursively legitimizes the underlying moral responsibility, but it does not create it.

Moral Status and Personhood

In order to understand the VCP on must grasp the concept of moral status. Moral status is what determines whether something can function as the A and B terms in the Vulnerability Relation, which is, as you will recall,

The Vulnerability Relation: A is vulnerable to B with respect to C because of D.

The B term specifies the bearer or subject of the sorts of moral responsibilities that I content derive from the Vulnerability Relations, while the A term specifies the addressees, beneficiaries, or objects of these kinds of moral responsibilities. The VCP says that moral agents can have moral responsibilities to protect those who are specially vulnerable or in some way depending on them for their care. But the questions which we must now address are:

Who (or what) can function as the bearers of moral responsibilities in this sense? and,

Who (or what) can function as the objects of moral responsibilities in this sense?

I will begin my discussion of the concept of moral status with the work on this topic by several contemporary philosophers.

Mary Anne Warren has characterized the concept of moral status as follows:

To have moral status is to be morally considerable, or to have moral standing. It is to be an entity towards which moral agents have, or can have, moral obligations. If an entity has moral status, then we may not treat it in just any way we please; we are morally obliged to give weight in our deliberations to its needs, interests, or well-being. Furthermore, we are morally obliged to do this not merely because protecting it may benefit ourselves or other persons, but because its needs have moral importance in their own right. [Warren (1997) Moral Status: obligations to persons and other living things. Oxford: Oxford University Press, p. 3]

Personhood is an example of moral status in this sense. To have the status of a person is understood as the basis of having human rights, such as the rights to life, liberty, freedom of expression, religious freedom, and the others. The moral status of persons is typically conferred upon types of individuals who meet certain general criteria, for instance, sentience, rational agency, intentionality, and the like. Normally adult human beings are regarded as persons. I have already noted that moral agency is a precondition for attributions of liability responsibility. So it is reasonable to suggest here that moral agency is likewise a precondition for something functioning as the bearer of moral responsibilities in the substantive normative sense. So, the short answer to the first question is that only individuals who possess moral agency can function as the bearers of moral responsibilities in the vulnerability relation. Since persons possess moral agency, persons can be the bearers of moral responsibilities.

Adult human beings are, for the most part, persons. Philosophers sometimes call human moral agents 'undoubted persons' because there is no doubt that they satisfy the conditions for moral agency. However, not all human beings are also persons in this sense. Individuals in deep comas are not, nor are infants and young children, nor are embryos and fetuses. In none of these kinds of cases does it make sense to ascribe moral responsibilities to these kinds of human beings. There are also some borderline cases in which one might legitimately doubt whether someone is a person, for instance, when dealing with severely psychotic or schizophrenic individuals. People who are deranged in these ways are not normally held to be morally responsible for their actions, even though they look just like undoubted persons.

But there are also certain kinds of things which are persons but are not human beings. The prime example of this category are corporations and other sorts of collective entities. I will return to this topic at a later point, but for the time being, I want to note that corporations and other kinds of collective entities can be ascribed moral responsibilities even though they are not human beings, and in fact, are not even alive. Corporations do not bleed and they do not die. Nevertheless it is important to understand that corporations can satisfy the necessary conditions for moral agency, and hence be regarded as moral as well as legal persons.

If there are extra-terrestrial intelligent races out there somewhere in space, then it is quite possible that we would classify them as persons who are not also human beings. Think, for instance, about the alien in Steven Spielberg's movie E.T. -- not human to be sure, but capable of moral agency and hence moral responsibility nonetheless. It is certainly imaginatively possible to conceive of other kinds of beings which are 'persons' in the sense of possessing moral agency, but which are not human beings. Think about hobbits.

Science fiction provides other examples of possible non-human persons. Issac Asimov's story "The Centennial Man" (later made into a movie called the Bicentennial Man starring Robin Williams) contemplates the possibility that some day we will have robots that claim their civil rights as persons under the law. For the time being, this is just a fictional possibility, but some authors, such as Ray Kurzweil, have predicted that the day in which we have such forms of artificially intelligent robots is not as far off as many people think. If and when these robots do come into being we will need to think of them as nonhuman persons.

Some people also think that some of our biological relatives, higher primates such as chimpanzees, or cetaceans such as bottle-nosed dolphins, might quality as nonhuman persons. Other people doubt this is true, which is why it might be possible to call these members of other species "doubted persons". Since I think of personhood as closely connected with those capacities necessary for moral agency, I am comfortable with this designation. Philosophers, by and large, have tended to over-emphasize the evolutionary discontinuities between human beings and other intelligent species. I, on the other hand, am more persuaded by the arguments of the biologists and evolutionary psychologists that there are significant continuities between humans and other members of the animal kingdom in the underlying psychological capacities that make moral agency possible.

Philosophers have spent a good deal of time and lots of ink trying to figure out exactly where to draw the boundaries of moral personhood. While interesting, I am not going to say much more about this topic here other than to note that I believe this is a matter for decision rather than discovery. That is, we will ultimately need to decide to draw the boundary somewhere or another, and where ever we draw it is going to be politically contested, as it is, for example, in the debate over the morality of abortion. We can fix on various kind of natural or psychological features of individuals as the basis for drawing the boundary, but this does not alter the fact that we construct and create it.

Definitional boundaries are rather like political borders in this sense; we can use the presence of a river or a coastline as the marker for a political boundary, say between the states of New Jersey and Pennsylvania, but it is still our decision to use this natural feature as a boundary. The ability of the river to function in this way depends upon our collective intentionality to treat it as a political boundary. The same thing is true of the definitions of terms like "person" or "moral agent". We construct the meanings of our symbols.