The Moral Magic Problem in Theories of Rights

2.1 The Puzzle

Some rights are moral rights: they correlate with obligations that morality alone would recommend, even in the absence of social conventions. The right of every person to be treated with dignity and respect is none other than the translation in terms of rights of the obligation to treat everyone with dignity and respect and does not depend on the establishment of a social convention. So is the right to abortion if you think women possess it even where abortion is unlawful.

Some rights are conventional, as their respective obligations only emerge within artificial social conventions.^[6] Of these conventional rights, some are the conventional translations of the rights moral agents would be entitled to on a purely moral basis. Most conventional rights, however, correlate with obligations that are distinct from, although sometimes overlapping with, moral obligations.

Why do rights violations wrong the right-holder when the right is either moral or, if conventional, sufficiently close in the obligations it generates to a moral right? Here, the justificatory version of the IT offers a satisfactory answer.^[7] According to it, moral rights presuppose the existence of a benefit of such moral relevance that it can by itself generate a duty to contribute to the realization of a state of affairs in which the right-holders may enjoy it. Hence, violating moral rights wrongs right-holders because it frustrates a morally relevant interest of them.^[8]

How can we differentiate between an interest that has and one that does not have the degree of moral relevance that suffices to make it a moral ground of rights-based obligations? A useful test is to consider if there is something morally faulty in societies that do not include protection of that interest among their conventionally recognized rights. If we employ the test to assess the two rights discussed at the outset of the paper, we see that neither reaches the threshold of moral relevance. Note that the test does not depend on the positive effect that the interest satisfaction has on a person’s wellbeing. I am not claiming that, necessarily, the interests of moral relevance are the ones whose satisfaction bring the highest net positive effect on an average person’s wellbeing. The test is moral rather than empirical and hedonistic; its ground, however vague, is the intuitive wrongness of seeing some interests constantly frustrated or never satisfied within particular political environments.

What are the interests in play in the two examples at the outset? Presumably, an interest in spending one’s life in comfort in Scandia and an interest in keeping others under control in Patriarchia. The latter is utterly immoral, and we can give it a negative value on the scale of moral relevance. The first, by contrast, has at least neutral moral value but does not seem to meet the threshold of moral relevance that would make it, by itself, the ground of obligations. Societies that do not recognize the right of Scandia do not seem morally faulty in the same way, say, of societies that do not recognize freedom from torture. If that is the case, the right in Scandia is a conventional right that, though compatible with morality, does not give legal recognition to an independently defensible moral right. But then it seems that the interest itself cannot explain why we feel that right-holders are wronged when their rights are violated. If not from the interest, where does the wronging character come from?

Someone could be tempted at this point to dismiss the puzzle through a deflationary strategy. It could be said that the Scandia example is undertheorized since there are various ways in which one can be prevented from the enjoyment of the right. One way is through some sudden and unexpected legislative move that frustrates the right-holders’ legitimate expectations. Another is by forcibly preventing a single right-holder (or a single group of right-holders) from enjoying the right while letting the rest of society continue to enjoy its benefits.^[9] Then one could argue that, in each different case, different principles explain the wronging character of the violation. In the first case, for instance, one could refer to the frustration of expectations as such, with the right being relevant only instrumentally as the normative device that generates the morally relevant expectation. In the other, we could appeal instead to the value of equal treatment before the law in a community of moral peers.

Let’s agree first that the violation in question has to do with preventing a single right-holder (or a single group of right-holders) from enjoying a right that is part of the publicly acknowledged conventional code. (I will have more to say about expectations later.) The suspicion that the wronging character of this kind of violation could be explained by the value of equal treatment cannot be rejected easily. Yet, equal treatment as such does not seem to explain why citizens deprived of the enjoyment of the right in Scandia, but not in Patriarchia, are wronged. Indeed, both cases can be presented as ones where the principle of equal treatment before the law has been breached. The fact that in one case but not in the other the law is itself discriminatory is irrelevant; even discriminatory laws can be applied equally or unequally. In both cases, the situation of the right-holder can be described as that of a citizen who was guaranteed by the law a certain benefit in the form of a right and was then denied the enjoyment of the same benefit, while other citizens who also qualified for it were treated in accordance with the right-conferring norm. In both cases, then, some citizens were treated unequally. Hence, what differentiates a case like Scandia specifically?

It is not just intuitions about cases like Scandia that give rise to the puzzle. As I wrote at the outset, a significant body of literature takes it almost as granted that all kinds of rights violations generate wronging, or at least that they do so when the right is compatible with morality (as in the Scandia example).^[10] Joel Feinberg, for instance, writes that “one person wrongs another when his indefensible (unjustifiable and inexcusable) conduct violates the other’s right” (1987, 34). Similarly, while responding to Judith Thomson’s (1990) distinction between infringing on and violating a right (with the first being the justifiable breach of a right-based obligation), Frances Kamm writes “suppose A has a right that I not lie to him, but I ought to in order to save B’s life. It seems to me that I have wronged A, though I infringe and do not violate his right” (2004, 478). In defending his Kantian account of private wrongs, Arthur Ripstein holds that “a wrong is an action that is inconsistent with another person’s right to body, property, or reputation” (2016, 8). Finally, in what is probably the clearest statement on the matter, Rowan Cruft writes that “any morally justified directed duty, whether socially created or natural, defines the sphere of moral respect owed to the party to whom the duty is directed. When someone violates morally justified duties owed to me … I am not simply the occasion for wrongdoing committed against the values that justify the structure of the organization in which I am manager. Rather … the wrong committed is a wrong done to me, one which shows me disrespect” (2019, 76).

Cruft helps us dissipate a misunderstanding that may have emerged at this point. RW – the thesis that violations of rights are, ipso facto, forms of moral wronging – is not merely a definitional thesis. The point is not just that we would better reserve the word “wronging,” as opposed to mere moral wrong, to cases where one moral agent has violated the right of another; as Cruft’s reference to disrespect makes clear, the thesis is that wronging is a particular form of moral injury – a directed form of moral injury – which can come about, among other ways, by violating others’ rights. In yet other words, RW is a substantive moral thesis about the relationship between rights violations and moral wronging and not an exercise in semantic tidying up.

If the wronging that figures in the RW thesis is a moral wronging, and if, by implication, RW is a substantive moral thesis, one cannot just claim, as many are inclined to do, that there are distinct senses of wronging depending on whether the violated right is moral or purely conventional. No author subscribing to RW seems to treat wronging as a morally neutral descriptor. They all intend wronging to entail a particular kind of moral injury characterized specifically by second-personal direction.

But how can we capture more specifically this form of directed moral injury? Let me offer an account in the next subsection.

2.2 Toward a Definition of Wronging

What does it mean to wrong someone? If we answered, following Feinberg and some of the other authors mentioned previously (Coleman 2002, 332; Feinberg 1987, 34; Kurki 2018, 444), that a wrong perpetrated against somebody just is a violation of their rights, RW becomes no more than a tautology. But intuitively we recognize that the usage of the expression “X wronged Y” is wider than cases where Y’s right was violated by X.^[11]

We need a definition of when one agent wrongs another that prescinds from the existence of rights. We can start by considering a paradigm of wronging, a general case where we find it particularly apt to apply the concept of wronging. I cannot think of a more paradigmatic use of the concept of wronging than this: a certain agent, A, has done something to agent B that requires, from A, some kind of remedy.^[12]

A simplifying way to make sense of the necessity of the remedy would be to attribute to the remedying a kind of metaphysical power capable of reverting the temporal order so that the damages visited on the wronged party are fully annihilated. This idea of remedy, which we can retrieve in Aristotle’s comments on corrective justice,^[13] is implausible. Firstly, it assumes that remedying a situation simply means restoring a disrupted original order, ignoring the double problem that such restoration might be impossible (Hershovitz 2010, 93–95) and that the original order might be unjust (Coleman 2002, 304–305). Secondly, it pays no attention to the fact that one agent did something to another that they need to remedy. A simplistic model of remedying as restoration focuses the attention away from the morally relevant relation between two agents (the wronger and the wrongee, so to speak) to the morally insignificant fact that a change in the world has occurred that was caused by someone and affected someone else.

So the challenge for an account of wronging is to make sense both of the “doing something to someone” part without collapsing that into a mere causal relation and to explain the moral necessity of remedying which remains a mystery if we give remedying a purely restorative meaning. My suggestion is that the mental state we need to focus on to explain in what sense wronging means doing something to another that needs to be remedied is one of concern (or lack thereof) for the agent one is affecting. Specifically, I want to suggest, we wrong someone when we act in a way that demonstrates inadequate concern toward them; more precisely, we do not take sufficiently into account in our deliberation the fact that they are, in all relevant respects, our moral peers.

The account of wronging suggested here is at home within contemporary contractualism. Contractualists claim that an agent commits an act of wronging when her conduct toward another is not justifiable to a generic, reasonable person who found herself in the position of the latter.^[14] My conduct toward another agent is justifiable to them (or, more precisely, to a generic, reasonable agent in their position) if there is no alternative course of action that, without creating an unreasonable burden to me, would have led to lesser harm for them or would have subjected them to a less significant risk. If, on the other hand, an agent undertakes a course of action that she should be aware leads to significant harms or risks for others, disregarding an alternative course of action that is less risky and that does not impose excessive costs on her, her action is unjustifiable to whoever suffers the consequences. And it is unjustifiable, I want to suggest, precisely because those who deliberate that way demonstrate inadequate concern for others.

We can further analogize my proposal to the idea, which goes back at least to P.F. Strawson’s statement in Freedom and Resentment, that wronging makes it fitting for some moral agents to express the characteristically second-personal reactive attitude of resentment. Strawson speaks in particular of “some degree of goodwill or regard” that we demand from others who stand in certain relationships with us, including “as chance parties to an enormous range of transactions and encounters.” (1974, 6). Strawson’s “goodwill or regard” corresponds to my idea of adequate concern; it represents at the same time what we can legitimately demand from others and what we can hold them accountable for.

The adequate concern account explains wronging at a significantly high level of abstraction. The scope of adequate concern will vary dramatically depending on the type of relationship existing between two agents; the adequate concern a parent has to show their child is not just different but more onerous than the adequate concern two perfect strangers ought to demonstrate to each other on the street. Despite its abstraction, the main advantage I see in the account is that it can easily explain why remedying is morally obligatory for the wronger and can thus explicate the inherent link between wronging and canonical remedial practices such as apologizing or asking for forgiveness. The remedial obligation is triggered because, as the wrong consists of a lack of adequate concern, the best wrongful parties can do to make up for it is to dedicate themselves fully to remedying it, thus offering, so to speak, an extra dose of concern to a party they have previously neglected.

Consider, in this respect, apologies. Apologies succeed at responding adequately to wronging, when they do, because they negate the message, inherent in each act of wronging, that one “can be treated in this way, and that such treatment is acceptable” (Hieronymi 2001, 546). By apologizing, the wrongdoer not only acknowledges her having previously shown inadequate concern to the victim, but she also implicitly commits to respecting the other as a moral peer in all future interactions (see Martin 2010).

Now that I have offered a general account of wronging, the puzzle can find a more precise presentation. Can we vindicate the intuition according to which the citizens of Scandia whose right had been violated have been wronged? The answer is yes if we can redescribe the violation as an act whereby the violator demonstrated inadequate concern for the right-holder and is, because of that, now liable to remedial acts. It is natural to redescribe the situation this way for many rights violations. The violator is aware, or should be, that the presence of a right indicates that an important interest of the right-holder is at stake. Neglecting this in one’s practical deliberation qualifies one’s action, in the view of others, as inadequate moral concern.

But does that apply, necessarily and without exception, to all cases of rights violations? We want to answer “No” at least with regard to the Patriarchia example, which is however excluded by the “at least consistent with morality” clause of RW. But I am not sure violating the right in Scandia would demonstrate a lack of adequate concern for right-holders either. Or, at least, we need an argument for that, as all we have in the description of the case is a social rule assigning a level of protection to an interest of specific right-holders; an interest that, I have showed, does not meet the threshold of moral relevance.

4.1 First Attempts at Using the IT to Solve the Puzzle: The Basic Version

At this point, one could propose the following interest-inspired solution to the puzzle. I have suggested that wronging consists in neglecting another in one’s deliberation. The interest theory of rights, on the other hand, holds that rights always protect interests of the right-holder, which may or may not play a justificatory role in the explanation of why the right exists. But then, by implication, whenever I violate a right, I also frustrate an interest of the right-holder. And isn’t that possibly the explanation we are looking for? After all, deliberately frustrating others’ interests might seem a rather paradigmatic way of neglecting others in one’s deliberation.

As simple as it might seem, this explanation of the wronging character of rights violations is doomed. And it fails, as I will try to prove in this section, even if we expand significantly on what it means to frustrate another’s interest or if we combine the interest thesis with other features of rights, such as their directionality.

Let’s start, however, with the simplest explanation, which derives the wronging character of rights violations from the recognition that all rights protect interests of the right-holder. This simple thesis fails because the interests protected by conventional rights may not meet any threshold of moral relevance or may not belong to the right-holder in any meaningful sense.

I noted before how the justificatory version of the IT could easily explain the wrongingness of rights violation for moral rights. If I am aware that somebody’s right protects a morally relevant interest of theirs, and still decide to ignore this in my deliberation, I am showing neglect toward the right-holder; I am failing to let my conduct be determined by something relevant about others. But, I also noted before, this justification is unavailable for purely conventional rights.

First, note that, in the way I have presented it, the justificatory version of the IT is a thesis about the inevitability of interests in the justification of why a specific right (ought to) exist. The thesis merely holds that if we want to offer a justification for why a particular right exists or ought to exist, we must mention the interest protected by the right. But the thesis does not claim that the interest must be either the sole element in the justification or the morally relevant element; in fact, supporters of the justificatory version are happy to accept that, at least for legal rights, the moral relevance of the interest may be null.

Joseph Raz, for instance, writes that “an individual has a right if an interest of his is sufficient to hold another to be subject to a duty” but that “his right is a legal right if it is recognized by law, that is if the law holds his interest to be sufficient ground to hold another to be subject to a duty” (Raz 1984, italics added). If I interpret correctly the italicized part, it implies that the law can artificially elevate whatever type of interest to an interest of “sufficient importance” to ground a duty. As such, the justificatory relationship between interests and rights for legal rights is mediated by the law, which may assign a level of relevance to the interest that it would not have outside the law.^[17] The interest is still part of the justification of why the right exists – in its absence, we would merely have a legal duty but no legal right – but it is neither the sole ground of the right nor an element that gives us any additional moral reason why the right ought to be recognized; it is merely an irreplaceable element in justification, without which the legal right becomes meaningless (see Fornaroli 2022).

Let’s go back to our initial example. The justificatory version of the IT would hold that the interest in spending one’s life in comfort is an ineliminable element in the justification of the right. Yet, we continue to assume that the interest does not meet the threshold of moral relevance that would make it, in itself, the ground of obligations, even in the absence of legal recognition. Do we still have reasons to claim that individuals who are denied the enjoyment of the right in Scandia are wronged? If we still do, these reasons cannot depend on interests in the same immediate manner they do for moral rights.

In general, I do not need to make up for what I did when I frustrate others’ morally irrelevant interests, even when such interests may play a significant part in advancing a person’s wellbeing. Consider participants in an auction or a fair competition that assigns a single prize. Whoever succeeds in winning the prize or outbidding the others will frustrate their interests, often remarkably so. But because the interest in succeeding in a specific enterprise does not meet the threshold of moral relevance, winners do not need to make up for the frustration of losers’ interests. They might and sometimes should feel sorry in specific circumstances but there is no correction to be put in place. The reasoning is simple: by participating in the competition and competing to win, I am not demonstrating inadequate concern for other participants’ moral status. That would be the case only if I took advantage of a structural unfairness of the competition or if I knowingly participated in competitions whose results would burden participants to an excessive extent.

Moreover, I have tacitly assumed so far that all cases of rights violations necessarily involve the frustration of an interest of the right-holder, however morally irrelevant. But this assumption does not hold. There is no implication from IT, in either version, to the idea that all rights violations frustrate right-holders’ interests.

To see this, consider that, in both versions of the IT, right-holders are identified as generic individuals or types.^[18] Both moral rights and the rights inscribed in conventional norms do not aspire to protect the interests of specific individuals, but only of types of individuals picked out through some generic designators. Even the rights that are contingently and temporarily enjoyed by a single individual (think of the right to issue executive orders in the United States) aim to protect the interests of types (the president of the United States).

Because the interest the right serves is presumed to be characteristic of types of individuals, it might happen that some real right-holders do not have an interest in whatever the right entitles them to. The right to property, for example, provides various options and benefits associated with what one owns. A generic interest protected by the right is not too hard to identify. For example, we can take rights to property to protect an interest in controlling or having authority over external things (Owens 2019),^[19] or in being able to benefit from what one takes care of. But there are real-case scenarios in which one has no interest in controlling what one owns. Take somebody who is burdened with the ownership of an unsellable, unrentable, and uninhabitable house which will cost her a lot to refurbish and on which she needs to pay taxes. We can hardly say that the concrete property right over the house protects one of the owner’s interests. And we can say that without compromising on IT, as IT merely holds that for a right to exist there must be an interest of the generic right-holder that the right is supposed to protect, not that the right is going to protect equally the interests of each real right-holder.

Hence, we should ask ourselves once more: if it does not come from interests as such, where does the wronging character of rights violations come from? Merely affirming the wronging thesis is tantamount to saying that conventional rights possess a “moral magic” capable of turning an otherwise morally irrelevant interest into a morally charged particle that, when left unsatisfied, produces an instance of wronging.

Postulating a moral magic in conventional rights is problematic not just for its dubious meta-ethical grounds but because it does not allow us to make any distinction among cases of rights violations that do and do not wrong the right-holder. If there is indeed moral magic in the conventional proclamation of a right, then all rights possess it. Hence, we cannot say any longer, in line with the intuition at the beginning of the paper, that only the citizens of Scandia are wronged when their morally justifiable right is violated. If a morally trivial interest is elevated to a morally relevant one when served by a conventionally recognized right, why shouldn’t the same happen with immoral interests?

To sum up, there are two circumstances under which we are inclined to question an immediate implication from IT to RW. The first case concerns conventional rights that, though morally justifiable, protect morally trivial interests. Let’s call rights protecting morally trivial interests “morally discretional” as their recognition is neither recommended nor prohibited by morality. The second case, which can sometimes coexist with the first, concerns concrete rights that do not protect any interest of real – as opposed to ideal or generic – right-holders. Both cases invite the question: how is it that violations of (morally justifiable) conventional rights wrong right-holders, when no morally relevant interest of theirs, or no interest whatsoever, is frustrated? In the next subsection, I will consider how more sophisticated versions of the IT can cope with the puzzle.

4.2 Further Attempts at Solving the Puzzle: Legitimate Expectations, Rule of Law, and Normative Interests

Sometimes, admittedly, the law does elevate some morally irrelevant interests to the threshold of moral relevance without that involving any kind of moral magic. Consider, for instance, my interest in driving on the right side of the road.^[20] An interest of this kind either does not exist outside specific conventions or, if it does, is very far from meeting the threshold of moral relevance. Yet, under a jurisdiction where everyone is supposed to drive on the right, my interest in driving on the right side of the road acquires a distinct moral relevance; if another driver occupies the wrong side of the road thereby putting me in danger, she seems to be wronging me personally on top of violating a coordination convention.

Cases such as these can be described as ones where the law altogether creates a novel morally relevant interest which would not exist outside of legal institutions. An analogous case would be the interest of every adult US citizen in voting for their representatives in the US Senate. An interest described in such fine-grained manner (as opposed to, say, a generic interest in political representation) would obviously not exist outside an already established institutional structure. Yet, within the structure, it does acquire the status of a morally relevant interest and its frustration qualifies as a form of wronging.

In general, artificial conventions have the ability to transform the normative landscape and to bring about morally relevant interests. They can do that either by creating factual conditions that make some interests morally relevant within the convention – arguably, the situation with the interest in driving on the left – or by giving a determinate content to morally relevant interests that are already there, but with a relatively indeterminate content, in the state of nature. But not all interests protected by the law as a matter of right are automatically turned into morally relevant interest. Consider the difference between the interest of every adult US citizen in voting for the US Senate and the interest of Scandia’s 18-year-old citizens in receiving assurance regarding spending a life of comfort. Both are protected by the law as a matter of right but only one gives determinate content to the more abstract interest in political representation, which is itself a morally relevant interest. Moral magic does not emerge as a problem when we already have an explanation for why a particular interest is morally relevant, including when the moral relevance only emerges under specific, contingent institutional arrangements.

Let’s now move to consider other explanations that could be proffered to buttress the simplistic interest-based explanation. One suggestion would be to focus on the expectations that necessarily emerge when conventional rights are recognized. There is now a burgeoning literature on the kind of legitimate expectations that citizens can claim about the law and the normative significance we should assign to them.^[21] And the wide consensus is that right-holders are indeed among the agents who have legitimate expectations in the law. In fact, rights and promises are usually taken among the most canonical normative devices that can generate legitimate expectations. Could we then rely on legitimate expectations and their frustration to explain why, if at all, violations of rights wrong the right-holder?

There is a simple objection against expectation-based explanations of the wrongness of rights violation. Rights do create expectations, but not for right-holders alone. To see this, let’s follow the supporters of the right-holding version of the interest thesis on the correct manner to identify the right-holder – the so-called Bentham test (Kramer 2017; Kurki 2018; McBride 2020). Following the Bentham test, citizens who turn eighteen qualify as right-holders in Scandia because their not receiving the benefit is part of the minimally necessary set of facts establishing the breach of the correlative duty. By contrast, consider the canonical case of a shop where some 18-year-olds habitually do their grocery. The shop owner fails the Bentham test and is not, because of that, a right-holder: her not receiving an increased profit is a redundant fact among those establishing the breach of the duty since it can be evidence of the breach of the duty only if coupled with the fact that one or more 18-year-olds have not received the benefit. Yet it is rational for the shop owner to form expectations about her increased profit in case the benefit is delivered and such expectations are indeed frustrated when the right is violated.

If we believe that the wrongingness of rights violations depends on the frustration of legitimate expectations, we have to accept that the shop owner too is wronged when the right is violated. But this way RW is undermined; we do not have a reason any longer to explain why the right-holder specifically is wronged when one of their rights is violated. Even worse, we do not have a reason to assume that all right-holders are wronged as there might be right-holders who form no expectation about the right being delivered. Paradoxically, this psychological attitude tends to be frequent when rights are ordinarily violated. Hence, we would have to say that when right-holders lose any expectation that their right will be respected any further violation stops wronging them.

One could object, nonetheless, that when compared with expectations that merely result from the law, right-holders’ expectations deserve a kind of priority in the normative response that is most adequate to them. The law, for instance, gives very limited recognition to expectations that are not protected by rights, whereas it gives priority to those expectations that are generated by the presence of rights (Pölzler et al. 2022).

I find this response indefensible once we adopt a purely moral viewpoint. From a purely moral perspective, we cannot merely presuppose that all expectations that have been merely induced by the presence of a right (the shopkeeper case) count less than expectations cultivated by right-holders themselves. Consider a case from interpersonal morality. I have promised my ailing father that, once he passes away, I will leave his fossil collection to a dear friend of his who is, unlike me, a fossil expert. The friend knows about the promise, has no reason to suspect I would not meet its terms, and therefore forms an expectation about receiving the fossil collection – he even creates a nice display in his living room. Suppose that, when the unfortunate time comes, I decide to keep the collection for myself. Leaving complications about promises to the dead aside, I agree that, by frustrating the promise, I seem to wrong the promisee in a way that is qualitatively distinct from the way I may be accused to be wronging the third party. But that distinction cannot be explained by expectations since, by assumption, the only one who raised an expectation in this case is the third party beneficiary. We can imagine similar cases in which it is third parties alone, rather than right-holders, that raise expectations about the delivery of a particular right-associated benefit. And, again, if we want to say that it is right-holders alone that are wronged when the right is violated, or that wronging them is qualitatively distinct, we cannot rely on expectations alone. To make sense of that difference, we have to rely on a moral principle that is still elusive.

So far, I have merely considered non-normative interests. But moral agents also have what David Owens calls normative interests or interests in exercising normative control, through such practices as consenting or promising, over a portion of the external reality;^[22] in Owens’s words, they are the interests that people satisfy when they “impose obligations … and make certain actions into wrongings” (Owens 2012, 143).

I agree that people have normative interests and that such interests, as Owens has recently argued, can explain – at least in part – the value of those social practices, such as friendship or neighborliness, that involve mutual obligations (Owens 2022). The problem is to see whether we can appeal to such normative interests to explain why violating rights is wrongful to right-holders. I think there are two problems with the move, depending on which kinds of normative interests we attribute to right-holders.

Commonly, although with exceptions (Cruft 2019, ch. 3), rights assign to right-holders the normative power to demand that others act in a certain way to them. As a right-holder, I can demand from each bearer of the correlative duty that they discharge such a duty as an act that is due to me. Can we say, then, that every rights violation frustrates the normative interest each right-holder acquires in demanding that the correlated obligation is respected? If we do, we are trapped in a dilemma. Either the normative interest in demanding always reaches the threshold of moral relevance, which leads to counterintuitive implications, or its moral relevance will vary depending on the non-normative interest that the right aims to protect. In the first case, we are forced to say that citizens in Patriarchia and other holders of immoral rights are wronged when their rights are violated because their normative interest in demanding that the right be respected is frustrated. But we are excluding that option from the beginning – it would clash too deeply with our intuitions. The only alternative is to say that the moral relevance of the normative interest in demanding is a function of the moral relevance of the non-normative interest the right aims to protect. If the non-normative interest the right aims to protect is sufficiently morally relevant, so is the normative interest in demanding that the right be respected. But this way the normative interest loses any explanatory power: the wrongingness of rights violation is already explained by the moral relevance of the non-normative interest.

But maybe we can appeal to normative interests more broadly, not as the interest each individual right-holder has in demanding respect for her specific rights but as the interest each moral agent has in the maintenance of a convention that assigns to them the ability to “make certain actions into wrongings.” It is the possibility itself of exercising these wronging-originating normative powers, rather than its actual exercise, which satisfies my normative interest.

If this is the best way to understand the role of normative interests in a theory of conventional rights, however, we are very far from understanding in what sense the violation of a right is a wrong perpetrated against the specific right-holder. The normative interest is satisfied through the existence of a convention that gives moral agents some normative powers. The violation of a specific norm of the convention cannot in itself jeopardize the convention; a single violation will certainly not deprive the right-holder of the ability to enjoy the normative benefits conferred on her by the convention. Moreover – and this is a crucial point – the violation of a conventional norm that assigns directed duties is not necessarily, by itself, a directed wrong.

Directed wrongs make it appropriate for the victim to display certain characteristic second-personal reactive attitudes such as resentment. Wronging is not just deeply personal but, in the terms introduced by Stephen Darwall, “second-personal” in the sense that it is part of that section of morality that concerns interpersonal relations and mutual accountability (see Darwall 2006).^[23] But how can the violation of a conventional norm that satisfies the normative interests of a plurality of agents wrong one agent specifically?

Let’s focus on a specific example. Suppose someone enters my property without my consent. My ability to put others under an obligation is unaffected by the single trespass. In fact, property law still gives me a significant array of options for enforcing my right, such as suing the trespasser and demanding compensation or forgiving the occurrence. I still hold, then, significant normative discretion. That is enough to weaken the view according to which a single rights violation has negative effects on the normative interests of the right-holder. But, more generally, my normative interest in enjoying a convention wherein I can be the recipient of others’ obligations and have some control over them does not vary depending only on whether my own rights are respected; if the single violation of a right does not jeopardize the convention, protracted violations will damage everyone’s ability to enjoy the convention. With protracted violations of conventional rights, it is the normative interests of all those who benefit from the convention, and not just the normative interest of the specific right-holders whose rights have been violated, that are damaged.^[24]

What I have been denying so far is the existence of a morally relevant feature, specific to the individual right-holder, which is frustrated every time a right is violated. One way to resist the conclusion would be to focus on the assurance that rights offer right-holders and, in turn, to the role such assurance may play in promoting individual autonomy.^[25] Leading one’s life autonomously means, among other things, being able to adopt meaningful choices on a wide range of relevant matters; it is this sense of autonomy that prompts some to see autonomy as the ability to see one’s life through the lens of self-authorship (see Raz 1986, ch. 5; Wall 1998, 127–130). But one can achieve self-authorship, the argument goes, only if she lives in an environment where at least some of her interests are protected stably and robustly; the absence of such stable protection would force one to be overly preoccupied with guaranteeing one’s basic safety, which would then prevent one from ever adopting those meaningful choices that can make one’s life self-authored.

This argument, although at first sight individual-centered, suffers from the same problems as the ones I have just discussed. First, let’s dispel some possible confusion regarding which interests ought to be robustly protected to promote autonomy. Of course, the robust, stable protection of morally relevant interests promotes autonomy. If I am aware that my morally relevant interests are protected, I can dedicate more of my time and energy to the pursuit of those objectives that make my life unique and are informed by values that are characteristically my own. But, if that is the extent of the argument from autonomy, then it has no bite against my main claim here, since I have been arguing throughout that violating rights that protect morally relevant interests is, ipso facto, a wrong perpetrated against the right-holder.

When it comes to legal rights protecting morally irrelevant interests, on the other hand, I may concede that they too have a role to play in promoting citizens’ autonomy in the same way that, say, the rule of law may promote citizens’ autonomy: by giving citizens assurance regarding the stability of legal norms – and, more specifically, about the stability of interest-protecting legal norms – the law can support citizens in the pursuit of their autonomy-enhancing objectives. But, if this is the argument, we encounter once more the problem of explaining the directionality of the wrong whenever a right is violated.

When it comes to rights protecting morally irrelevant interests, what matters for autonomy is that the system of rights as a whole is in place, so that citizens are given some assurance regarding which interests of them, however morally irrelevant, are protected robustly. But a single violation does not make the system less secure, unless minimally. And, to the extent that it does make the system as a whole less secure, it does so for all citizens and not just the one whose specific right was violated.

Once more, what we are lacking is an explanation for why the right-holder specifically, and not all those individuals who rely on or take advantage of the right, is wronged. Or, in other words, what is missing is an element that may help us qualify the wrongdoing of rights violating – which at most damages everyone’s normative interest in enjoying a conventional system of rights, or everyone’s ability to rely on the rule of law for their everyday choices – as “directed” toward the right-holder.

Nor can the puzzle be solved if we combine the IT with a commitment to the idea that rights always presuppose duties directed to right-holders (see Cruft 2013; Schaab 2018; Zylberman 2014). Directionality is a description of the normative relationship rights create between right-holder and duty-bearer. Each duty-bearer is required to conceive of her duty as owed to the right-holder, rather than as something she ought to do because third-personal reasons so require.

The directionality of rights is not only compatible with the existence of immoral (conventional) rights but can explain why immoral rights, such as the right in Patriarchia, are particularly egregious and are a distinct mark of indecent societies. It follows from directionality that any right assigning a privilege to a category of individuals demands from others that they conceptualize the privileged as people for whom they are required to act. The peasant is not just required to bow in front of the lord, he is required to do so as an act that is owed to the lord. One intuitive reason behind the disgust we feel at privileges and honors of this kind is precisely that they assign a kind of normative pre-eminence to individuals that have done nothing to deserve it.

Regardless of whether we understand interests in a more sophisticated manner or we pay attention to other features that characterize rights in general, the problem remains: the interest theory seems unable, by itself, to explain why rights violations wrong the right-holder specifically.