JUSTICE, BENEFICENCE, AND THE NEEDY

As Peter Singer (Reference Singer1972) famously put it, it would be morally unacceptable for bystanders to let a child drown in a shallow pond, when saving her would only cost them the inconvenience of getting their clothes wet. Similarly, it would be morally unacceptable for the world's privileged to cling to their surplus resources, when transferring them to the world's poor would save many lives. In short, it is beyond doubt that we have duties to help needy strangers when this is not too costly to us. What is at stake in the present discussion is the nature of these duties. Are they a matter of beneficence or also one of justice? Since justice and beneficence are different kinds of moral concerns, our answer has important implications.

Duties of justice have a specific function. They demand that we respect each other's entitlements to an equal sphere of personal sovereignty or freedom, within which we may pursue our conceptions of the good protected from interference (Lamont Reference Lamont1941; see the account of justice in Kant Reference Kant and Ladd1999/1797; Rawls Reference Rawls1999; Stilz Reference Stilz2009). The function of duties of justice, in turn, explains their distinctive features: rights correlativity and rightful enforceability (Buchanan Reference Buchanan1987; Loriaux Reference Loriaux2006; Valentini Reference Valentini2013).

First, duties of justice are owed to particular others, who are entitled to their performance: they are correlative to rights. For instance, when a burglar breaks into my justly owned property, he violates my rights: he interferes with my sphere of freedom and attempts to take possession of my means to pursue his own conception of the good (Ripstein Reference Ripstein2004a, 9–10; Reference Ripstein2006; Feinberg Reference Feinberg1986, chaps. 18–19). Second, duties of justice are rightfully enforceable: no wrong is committed when the bearers of these duties are forced to act on them.Footnote ⁴ After all, when agents are coerced in the name of justice, they are coerced for the sake of protecting or restoring others’ spheres of freedom (Kant Reference Kant and Ladd1999/1797; see Ripstein Reference Ripstein2009, 55ff.).

Unlike duties of justice, duties of beneficence are neither correlative to rights, nor rightfully enforceable (henceforth “enforceable” for short). They do not demarcate our rightful possessions, but demand that we lead our lives with appropriate concern for the good of others, by sometimes using what we rightfully possess to benefit them (Barry Reference Barry1991). When I am under a duty of beneficence, none of the recipients has a right to be helped by me. Moreover, duties of beneficence may not be enforced without wrongdoing. Their enforcement is always pro tanto wrong—i.e., wrong in at least one respect, though susceptible to being outweighed by other considerations—since it involves invading duty bearers’ spheres of freedom. To illustrate, while it is certainly wrong not to act on our duties of beneficence, so long as we are rightfully entitled to what we possess, it is also wrong to steal resources from us to promote others’ good, even if promoting others’ good is a morally obligatory end we wrongfully neglect. Two wrongs, after all, do not make a right.

Since the target of duties of beneficence is—within limits—dependent on duty-bearers’ understanding of what it is to promote the good of others, in a world with many competing conceptions of the good, these duties are a diverse set. For instance, planting new trees in areas plagued by forest fires and donating money to art galleries may each be done in fulfillment of one's duties of beneficence. The duty to help the needy, then, pertains to cases of particularly urgent beneficence, where “promoting the good of others” in effect amounts to rescuing them from misery, whether through bodily actions—as in the drowning-child case—or resource transfers—as in the case of the poor (Fabre Reference Fabre2002, 128–9).

Crucially, even the more urgent duties of beneficence, targeting needy strangers, retain the features of non-rights-correlativity and non-rightful-enforceability. Now, the idea that people have a right to do wrong—e.g., to vote for political parties that fail to promote just causes, to refrain from donating to worthy charities, and so forth—is familiar and prima facie plausible (Waldron Reference Waldron1981). But the suggestion that a wrong so serious as a failure of easy rescue could be protected by a right seems problematic. If we go back to Singer's scenario, it is hard to resist the intuition that bystanders have justice-based duties to rescue the child. The opposite suggestion, namely that bystanders have a right to refrain from rescuing the child, seems morally dubious.

To establish whether this intuitive judgement can be vindicated, we need to answer the following question: Under what conditions, if any, does the duty to help the needy qualify not only as one of beneficence, but also as one of justice? Different answers to this question, in turn, offer different accounts of “samaritan justice.”

DESIDERATA

Before moving on, let me make explicit the desiderata that a good account of samaritan justice should meet. Three are particularly important.

For instance, if our account of samaritan justice implied that slavery is sometimes just, we would probably reject it on grounds of lack of fit with the strongly held judgement that “slavery is always unjust.” To be sure, considered judgements only provide prima facie evidence in support of a given account of justice. On reflection, some such judgements might have to be abandoned in the course of the process of “reflective equilibrium.” Yet, an account that systematically clashed with a wide range of strongly held judgements about its subject matter would be unsatisfactory (Daniels Reference Daniels and Zalta2013; Rawls Reference Rawls1999; see McDermott Reference McDermott, Leopold and Stears2008).

There are many ways in which an account of a given subject matter X may fail to explain the evidence. For instance, the explanation offered might: (i) account for only part of the evidence and thereby be incomplete, (ii) fit the evidence but in a far-fetched or unilluminating way, (iii) simply assert the phenomenon it aims to explain, thereby being merely stipulative, and/or (iv) fit a specific type of case without being generalizable to relevantly similar cases, thus being ad hoc. An account of samaritan justice that displayed one or more of these explanatory inadequacies would be unsatisfactory.

For example, the account should satisfy the “ought implies can” proviso, according to which we can have a moral obligation to do X only if we are capable of doing X. Similarly, it should be consistent with the claim that only moral agents—as opposed to inanimate objects or agents not capable of moral reasoning—can bear duties; and so forth.

D1 to D3 are familiar desiderata that any normative theory should satisfy. When it comes to debates about samaritan justice, the particular piece of prima facie evidence we are trying to explain—or “ground/justify,” as philosophers sometimes say—is the intuitive judgement that, in easy rescue cases, bystanders have justice-based duties to help the needy. A satisfactory explanation, in turn, will have to bear the right relation to the evidence, and not clash with theoretical fixed points. If no account meeting these desiderata can be found, the intuitive justice-based duty to assist will remain unvindicated, leading us to conclude that there are no easy rescue cases in which that duty exists—despite our intuitions to the contrary.

INDIVIDUALIST SAMARITANISM

Some scholars start from the uncontroversial observations that (i) needy strangers’ ability to pursue their conceptions of the good is severely compromised, and (ii) third parties have duties to help them at reasonable costs. They then conclude that considering these duties a matter of justice, hence enforceable and correlative to rights, is most in tune with their aim and stringency.

In this vein, Pablo Gilabert (Reference Gilabert2010) argues that, even from a Kantian perspective, need that can be easily met generates justice-based duties to help.Footnote ⁵ Similarly, Cécile Fabre claims that if “the needy have a right, as a matter of justice, to some of the resources of the well-off,” then “the imperilled [also] have a right, as a matter of justice, to the bodily services of those who are in a position to help” (Fabre Reference Fabre2002, 129, 142, added emphasis; see also Feinberg Reference Feinberg1984, 60).Footnote ⁶ To return to the tragic event with which we started our discussion, on this view, Ms. Moore had an enforceable right to assistance against those bystanders who could have easily helped her. If we assume that one of them was named Lily, it would then follow that Ms. Moore had a right to be assisted by Lily, among others.

Since this view conceptualizes the justice-based duty to help the needy as an obligation that binds given individuals, I name it individualist samaritanism.

Despite its undeniable intuitive appeal, individualist samaritanism fails to vindicate the rights it claims to establish. Why? Because it grounds the needy's rights to assistance in third parties’ “being in a position to help at reasonable costs.” Problematically, for any provider of help there are countless candidate recipients. For instance, Ms. Moore is not the only needy person whom Lily—or anyone similarly positioned—is in a position to assist without too much sacrifice. There are hundreds of needy others Lily could help, say, by making small donations to targeted charities. Satisfaction of the condition “being in a position to help at little cost” does not allow us to identify Ms. Moore (or any other needy person) as the specific individual whom Lily (or any other potential rescuer) ought to help. The necessary link, matching givers and recipients of help, is missing. So stated, individualist samaritanism fails to explain the moral phenomenon it aims to vindicate.

It might be suggested that individualist samaritanism can be easily repaired by grounding the needy's rights to assistance not in third parties’ ability to help simpliciter, but in their being “in a unique position” to help.Footnote ⁷ Appeal to uniqueness allows individualist samaritans to match particular givers to particular recipients of help and is, in this sense, an improvement. However, it is also implausibly ad hoc: it confines the applicability of the view to a very narrow set of cases, involving no more than one party in a position to offer assistance. It suffices for a second or third capable bystander to turn up, that individualist samaritanism can no longer ground the needy's putative right to assistance, since none of the third parties is in a unique position to help them.Footnote ⁸

This modification would thus render individualist samaritanism in principle unable to vindicate the needy's rights in most real-world easy-rescue scenarios, where multiple bystanders are involved. Ms. Moore's story is, of course, a case in point, as is the infamous murder of Kitty Genovese, which reportedly occurred in the presence of several idle bystanders (Martin Reference Martin1989).

In response to this difficulty, proponents of individualist samaritanism might replace reference to third parties’ being “in a unique position” to help, with reference to their being in a “privileged position” to do so. This would allow them to extend the coverage of their view to a much larger set of easy rescue cases. These are the cases in which, as Barbara Herman (Reference Herman1984, 595–6) explains, failures to help reveal a lack of commitment to responding to others’ needs.Footnote ⁹ Although we often have discretion regarding when and how to help others—e.g., by donating to this or that charity, by volunteering for this or that organization, on this or that day, and so forth—in easy-rescue scenarios, this discretion typically disappears. One cannot (i) take the moral imperative of helping the needy seriously, (ii) be in a privileged position to rescue someone without much sacrifice, and (iii) remain inert. This explains why bystanders have a duty to assist those needy whom they are in a privileged position to help. But does it also explain how those needy acquire rights to be helped by those bystanders? I doubt so.

From a formal point of view, the rights asserted by individualist samaritanism are special: they hold between particular third parties and particular needy strangers, not between all human beings. Formally special rights, in turn, are typically grounded in special relationships—whether voluntary or nonvoluntary. For instance, children have special rights against their parents by virtue of their family membership; contractual parties have special rights against each other by virtue of having voluntarily signed a contract; fellow citizens have special rights against each other by virtue of their common societal membership, and so forth (Hart Reference Hart1955, 183; O'Neill Reference O'Neill1996, 152). But individualist samaritans assert that needy strangers have formally special rights against third parties in the absence of special relationships. It thus remains unclear where these rights “come from”—namely what grounds them—other than raw intuition or simple stipulation. While it is beyond question that bystanders have a duty to help the needy, this, per se, is not sufficient to ground the needy's entitlement to their help. As Frances Kamm (Reference Kamm, Coleman, Himma and Shapiro2002, 484) explains—referring precisely to easy rescue cases—the mere “fact that someone has a duty stemming from another's [e.g., a needy person's] interest gives no one in particular a moral entitlement to his fulfilling the duty.”Footnote ¹⁰

To illustrate, consider two foreign travellers: Tom and Greg. They independently embark on the same excursion, but Tom hires a local guide, Jack, for assistance, while Greg goes it alone. On the excursion, both travellers get injured. Jack can easily assist both of them and, in the circumstances, he ought to. But are they both entitled to Jack's assistance? In Tom's case, the answer is straightforward: he has a right to Jack's assistance by virtue of their contractual relationship. In Greg's case, by contrast, no contractual or otherwise special relationship can be invoked. While Jack clearly ought to assist him, it is not clear where Greg's putative entitlement to assistance would come from. On what basis could Greg insist that Jack owes him assistance? On what basis could he claim: “I need not be grateful to Jack for his help; after all, he owed it to me” (but cf. Feinberg Reference Feinberg1984, 58)?

Granted, intuitively, we may well want to assert that Greg has a right to be rescued. But this intuition is not self-validating, and individualist samaritanism—in all the versions discussed so far—fails to offer the required validation. It stipulates the existence of a right, without appropriately grounding it. In light of this, it fails to meet D2.Footnote ¹¹

This type of explanatory inadequacy does not plague what I call social samaritanism, to which I now turn.

SOCIAL SAMARITANISM

On this view, a version of which has been defended by Arthur Ripstein (Reference Ripstein2000), helping the needy is one way for citizens to discharge their duty to comply with, and support, just institutions.Footnote ¹² Just institutions provide citizens with equal freedom to pursue their ends and goals, including by protecting them from the kinds of misfortunes of which Ms. Moore has been a victim—through emergency services and welfare provisions. Citizens, in turn, have duties, owed to society—i.e., all other citizens—to sustain just institutions. These duties can be discharged in a multiplicity of ways and, depending on the circumstances, they might include supplementing emergency services by assisting needy fellow citizens (Ripstein Reference Ripstein2000, 774–9). As Ripstein emphasizes, though, “[t]hose who fail to [assist the needy] breach a duty to society as a whole, rather than to the particular person who is not rescued” (Ripstein Reference Ripstein2000, 775; see also McIntyre Reference McIntyre1994, 184). Succinctly put, this social samaritan perspective can be characterized as follows.

Social samaritanism improves on individualist samaritanism by grounding the justice-based imperative to help the needy in the special relationships that exist between fellow citizens. These relationships are governed by an ideal of reciprocity, whereby, in Ripstein's words, “the misfortunes which stand in the way of each person being able to live a self-directing life [are] held in common, so that all have the wherewithal to participate as full and equal members of society” (Ripstein Reference Ripstein2000, 765; cf. Sangiovanni Reference Sangiovanni2007). Just institutions are the mechanisms through which the sharing of misfortunes and the maintenance of equal spheres of freedom is achieved. Fellow citizens owe it to each other, as a matter of justice, to support just institutions. On this view, considerations pertaining to what each citizen is in a privileged position to do only affect the mode of performance of this justice-based duty, i.e., the types of actions each citizen should undertake to fulfil it. The duty itself is grounded in the special relationships existing between each citizen and the rest of society.

While social samaritanism improves on individualist samaritanism in one respect, this improvement comes at a cost. As Fabre (Reference Fabre2002, 130–2) points out, it commits proponents of social samaritanism to the explanatorily inadequate—because far-fetched—view that, by failing to help the needy, designated helpers wrong society, rather than the needy themselves. This is an unsatisfactory way of thinking about situations such as those involving Lily and Ms. Moore in my earlier example. It would seem morally mistaken for one of Ms. Moore's fellow citizens to reprimand Lily by saying “You have unacceptably wronged society!” A much more plausible complaint would be “You have unacceptably violated Ms. Moore's rights.” Yet social samaritanism regards the former complaint, not the latter, as the correct one.

It might be objected that it is (arguably) consistent with social samaritanism to hold that those who fail to help the needy commit both an institutional wrong of justice—i.e., by failing to discharge the obligation, owed to society, to support just institutions—and an interpersonal wrong of beneficence—i.e., by failing to respond to the plight of the needy.Footnote ¹³ What social samaritans argue is that only the former, institutional wrong is a violation of an enforceable right, specifically of one of society's rights against its citizens.Footnote ¹⁴ The interpersonal wrong, if it exists, falls outside the realm of rights. From an interpersonal, as opposed to institutional, perspective, the wrongdoer has a “right to do wrong” (Waldron Reference Waldron1981).

This response mitigates, but does not remove, our initial concern. It still remains explanatorily implausible to hold that, from the point of view of justice, someone who, say, fails to rescue a wounded child, violates a right of society, and not of the child himself. The locus of the injustice, if an injustice exists, must be the violation of the rights of the child, rather than of society. Adding one layer of explanation, by invoking wrongs that are not “directed,” hence not rights violations, does not eliminate the original difficulty. I thus conclude that, like individualist samaritanism, social samaritanism is explanatorily inadequate—it fails to meet D2—though for different reasons. Let us then turn to universalist samaritanism.

UNIVERSALIST SAMARITANISM

On this view, failure to help the needy wrongs the needy themselves—not society—and the duty to help the needy binds humanity as a whole. Suitably placed individuals, in turn, have obligations to enact the duty “qua agents of humanity,” subject to fair compensation.

Consider the following analogy. University X is under an enforceable duty of justice to pay for the dinner expenses of academic visitor Jones. In other words, Jones has a right against University X to have his expenses paid. Professors Smith and Kelly are representatives of University X at dinner with Jones, and thereby in a privileged position to discharge their university's duty.Footnote ¹⁵ In the circumstances, either Smith or Kelly (or both) ought to pay for dinner on the university's behalf, and later be reimbursed by it. Universalist samaritanism suggests that individual helpers stand in relation to humanity and the needy in roughly the way in which Smith and Kelly stand in relation to University X and Jones.

Although appealing at first, universalist samaritanism proves unsustainable, and collapses back into individualist samaritanism (with its associated difficulties). This is because “humanity”—unlike a university—is not the kind of collective that can be a duty bearer in anything more than a metaphorical sense. The only entities that can in principle bear duties are moral agents, namely agents with a capacity for moral reasoning. For instance, cats and dogs are not fit for bearing duties because they are not moral agents. Equally, stones and chairs cannot bear duties because they are not agents in the first place. From this it follows that a collective can bear duties in its own right only if it qualifies as a group moral agent (see, e.g., French Reference French1984; List and Pettit Reference List and Pettit2011; Tuomela Reference Tuomela2013; cf. Gilbert Reference Gilbert1989).

Group moral agents are sets of individuals whose internal organization—i.e., decision-making processes—is complex enough to warrant the ascription of moral agency to them. For instance, states, corporations, and universities typically meet the criteria for group moral agency, hence they are fit for bearing duties. By contrast, unorganized collectives such as a set of people sunbathing on a beach, or the passengers sitting in a train carriage, do not together form an agent. While group moral agents can be duty bearers in their own right, unorganized collectives cannot (French Reference French1984; List and Pettit Reference List and Pettit2011; Collins Reference Collins2013). Only the individuals within unorganized collectives, in their personal capacities, can bear duties, since only they qualify as agents.

It is transparent that humanity in its entirety does not constitute a group moral agent, but is only a very large collection of individuals belonging to the same species. Universalist samaritanism, then, clashes with a well-established piece of theoretical apparatus: namely that only moral agents can be duty bearers. It thus fails to meet D3. Expressions such as “humanity has a justice-based duty to help the needy” are meaningful at most as shorthand for the claim that “suitably positioned individual human beings, members of the human species, have an enforceable duty to assist the needy, correlative to the needy's right to assistance.” But this only takes us back to our original individualist picture, and its explanatory difficulties.

Our discussion so far has alerted us to a number of theoretical pitfalls that any plausible account of “justice-based help to the needy” must avoid. A revised version of social samaritanism, I suggest, avoids them.

REVISED SOCIAL SAMARITANISM

I first set out the main features of revised social samaritanism, and then relate this view to my initial remarks about the function of justice.

ANTISAMARITANISM(S)

Some are averse to any form of samaritan justice, defending what I call unqualified antisamaritanism.

An argument in support of this view is implicitly suggested in the work of H. M. Malm.Footnote ²² The argument—adapted to the context of the present discussion—goes as follows. A society organized around social samaritan principles requires a public criterion of the good on the basis of which duties to help the needy at reasonable costs are established. This criterion would determine (i) what counts as a reasonable cost, and (ii) what types of situations require remedying. Problematically, under the circumstances that make justice possible and necessary—i.e., people's adherence to different and conflicting conceptions of the good—a public conception of the good is unavailable. Enforcing the duty to help the needy would thus deprive at least some individuals of an adequate space within which to pursue their own conceptions of the good, and compel them to act on the basis of conceptions other than their own (Malm Reference Malm1995; Reference Malm2000).

To make this point more vivid, consider the following scenario.

Proponents of unqualified antisamaritanism argue that, by adopting general guidelines to decide between these cases—as the ex post mechanism demands—a social samaritan state would privilege some conceptions of the good over others. Should the religious devotee be excused, but not the environmentalist, or vice versa? No matter what general criterion is adopted, appeal to some conceptions of the good (e.g., religious ones) will successfully defeat the enforceable duty to help, while appeal to other such conceptions (e.g., green ones) will not (Malm Reference Malm1995, 21–4).Footnote ²³ This generates an unfair burden on the pursuit of some conceptions of the good and, perhaps even more troublingly, implicitly establishes a public hierarchy between different such conceptions (cf. Patten Reference Patten2012). Similar concerns arise, mutatis mutandis, when priorities are set for the allocation of state-based redistributive assistance, as in the ex ante mechanism described in the previous section.

Does unqualified antisamaritanism irremediably undermine the case for revised social samaritanism? Unqualified antisamaritanism raises an important objection, but in its current form, it is overstated. Although justice demands respect for individuals’ pursuit of their conceptions of the good, theories of justice themselves have to rest on some such conception. A purely formal theory of justice, which does not rest on any account of agents’ most important interests, is simply empty. Unsurprisingly, most existing conceptions of justice make some assumptions about what each person has reason to want to have. To be sure, this gives rise to particularly “slender” accounts of the good—whose value is instrumental to people's pursuit of their own conceptions of the good—but accounts of the good nonetheless. John Rawls (Reference Rawls1999, 54ff.), for instance, famously defends a list of “primary social goods,” including liberties, opportunities, income, wealth, and the social bases of self-respect, as central to his theory of justice. Martha Nussbaum (Reference Nussbaum2000), by contrast, advocates a much-debated list of basic capabilities, which each individual ought to have in a just society.

In the absence of considerable convergence between different conceptions of the good, anything a state does in the name of justice is bound to be in tension with at least some such conception. Even something as uncontroversial as a legal prohibition on murder may interfere with an ecocentric extremist's pursuit of his conception of the good, by preventing him from killing people in defence of the environment. Yet very few would challenge such a legal prohibition on the grounds that it places an unfair burden on the extremist's pursuits.

In other words, if we believe that the duty to help the needy should not fall under the purview of justice because this would make the state unjust, then we should also think—counterintuitively—that other demands of any plausible conception of justice are problematic because they make the state unjust. By this line of reasoning, any account of justice with some content is (paradoxically) unjust: a conclusion that is simply too strong, and clashes with a wide range of considered judgements, thereby failing to meet D1.

Although the unqualified rejection of a justice-based duty to help the needy has absurd implications, there is a lesson to learn from it. In some—but not all—instances, revised social samaritanism leads to unjust outcomes, which unduly shrink some people's spheres of sovereignty. This is when the needy may plausibly be held responsible for their plight, through rampant negligence or repeated risk-prone behavior. When this is the case, considering the state under an obligation to assist them on grounds of justice, and considering the “reckless” needy entitled to help, misconstrues the moral position of the parties involved.

As we saw in the previous section, from the perspective of justice, disadvantage for which fellow members cannot plausibly be held responsible should be compensated for by the state. Redistributive taxation and taxation supporting public services are aimed precisely at this; and so are duties of immediate rescue performed on the state's behalf. This is what establishing a fair background, in which each has roughly equal spheres of sovereignty, demands. Once such a background is in place, however, each is responsible for what happens in their lives (Rawls Reference Rawls1999, 86ff.; Ripstein Reference Ripstein2004b).

To see this, consider the following case.

What is the most morally accurate characterization of her situation? A proponent of revised social samaritanism is committed to saying that if Jenna becomes badly off and others can help her at small cost to themselves—e.g., by paying taxes to contribute to Jenna's subsistence—then, if they refuse to help, a fair background is no longer in place. Jenna's gambling habits automatically turn her more prudent and better off fellow citizens into “thieves” who unjustly hold on to resources that are actually Jenna's, i.e., part of her fair share. Similar conclusions would follow if Jenna's reckless risk-taking were to put her in immediate danger (e.g., a road accident), with bystanders in a position to rescue her at small personal cost. According to revised social samaritanism, bystanders would have a duty of justice to help, acting on the state's behalf.

This conclusion counterintuitively implies that some people's (the nonreckless's) spheres of sovereignty should be restricted for the sake of benefiting “the reckless.” It may well be that a morally good person should help unreasonable risk-takers for whom things have not gone well. Others’ neediness, independently of its causes, typically makes a difference to what we should do with our entitlements: it activates concerns of beneficence. However, it seems odd—i.e., at odds with considered judgements (in violation of D1)—to suggest that others’ recklessly self-inflicted neediness makes a difference not only to what we should do with our entitlements, but also to what those entitlements are.

If this is correct, then we have reason to defend a qualified form of antisamaritanism, setting out conditions under which the duty to help the needy should not be viewed as a matter of justice.

In the next section, I combine the conclusions reached so far to develop a plausible account of when the duty to help the needy qualifies not merely as a demand of beneficence, but as an enforceable, rights-correlative duty of justice.

QUALIFIED SOCIAL SAMARITANISM

Our discussion has shown that justice-based duties to help the needy are best construed as social, rather than individual, ones—in accordance with revised social samaritanism. Our discussion has also revealed that only those needy who act responsibly, and may not plausibly be held accountable for their plight, have justice-based entitlements to help. The “reckless” needy, on the other hand, are only appropriate recipients of beneficence.Footnote ²⁴ Taken together, these observations speak in favor of a view I call qualified social samaritanism (assuming the revised version of social samaritanism as my starting point).

From the perspective of qualified social samaritanism, nonrecklessness and societal membership generate justice-based entitlements to help against the state. Fellow membership matters because, as we have seen, those who place constraints on each others’ agency, by belonging to the same comprehensive rule-governed social system, owe it to each other to shape their common rules consistently with everyone's equal right to freedom. Nonrecklessness matters because, once a fair background is in place, each person is to be held responsible for her choices (cf. Ripstein Reference Ripstein2004b; Ripstein Reference Ripstein2000).

But what about those needy whose plight is caused neither by socially mediated bad luck (e.g., Ms. Moore, or the victims of an economic crisis) nor by personal recklessness (e.g., Jenna), but rather by others’ unjust behavior? To see how qualified social samaritanism handles this case, consider the following scenario.

According to qualified social samaritanism, qua private persons, better-off fellow members like Sarah are innocent bystanders, unconnected with Jenny's and Tom's plight. Yet, qua members of the same state, they share a small fraction of responsibility towards the victims of injustice. As we have seen, the state has duties to secure citizens’ equal spheres of sovereignty; and when some suffer injustices, the state ought to restore their equal sovereignty through compensation. If compensation cannot be extracted from the culprits—in this case, Henry, who has a primary, and personal, duty of justice vis-à-vis Jenny and Tom but has fled to Canada—then it will have to be obtained elsewhere. Fellow citizens may therefore appropriately be called upon to contribute to the costs of assistance to the victims on remedial grounds of justice, even though, qua private individuals, their duties towards the victims would be at most a matter of beneficence (on remedial responsibility, see Miller Reference Miller2001).

In summary, qualified social samaritanism allows us to offer nuanced and plausible answers to the question of when the duty to help the needy is not only a matter of beneficence but also one of justice. When the needy's plight is recklessly self-inflicted, they only qualify as recipients of beneficence. When, by contrast, they are not responsible for their predicament, in addition to reasons of beneficence, help is mandatory on either remedial or primary grounds of justice, occasionally discharged by specific individuals, on behalf of the state.

Table 1 schematically represents the verdicts of qualified social samaritanism, in relation to the various scenarios encountered up to this point.

TABLE 1

Justice, beneficence, and the duty to help the needy

Let me expand a little on the cases occupying the bottom row, involving Hurricane Sandy. In both cases, as I have described them, the needy are nonreckless.Footnote ²⁶ From the perspective of qualified social samaritanism, the hurricane's victims thus have justice-based claims to assistance on the part of the state, on top of being appropriate recipients of private beneficence. This means that their claims are correlative to rights and in principle rightfully enforceable. If meeting these claims demands resource transfers towards the victims, then setting aside a substantial portion of the federal budget to meet them is perfectly in line with the demands of justice. It is what organizing our common social rules so as to respect everyone's equal right to freedom demands. The U.S. Congress's decision to set aside $51 billion is beyond moral reproach. Qualified social samaritanism commends current practice in this respect.

What about assistance to be delivered through direct actions? Here qualified social samaritanism tells us that fellow citizens who could have helped nonreckless victims at reasonable costs but refused to do so are, qua members of the state (as opposed to personally), guilty of injustice. Of course, ascertaining whether, in any given circumstance, the conditions for condemning the relevant individuals hold—i.e., can help be offered at reasonable personal cost, and are the needy really nonreckless?—is epistemically difficult. But, as Feinberg (Reference Feinberg1984, 66) and Fabre (Reference Fabre2002, 140) point out, so is ascertaining whether, in any given circumstance, an act of self-defense qualifies as reasonable: something the law does routinely.

The scenario involving Ms. Moore, as I have described it, assumes this epistemic difficulty away. According to qualified social samaritanism, then, legal practice in the state of New York ought to be altered such that “bad samaritans,” like those who failed to assist Ms. Moore, could be rightly prosecuted for their omission.

OBJECTIONS

I have argued that qualified social samaritanism sets out plausible conditions for the duty to help needy strangers to qualify not only as one of beneficence, but also as one of justice. Below, I consider and respond to three objections against this view.