Contractualism, Aggregation and Promising

Contractualism and Promising I: Motivations for Contractualism

  1. Schedule

(1) Motivations for contractualism

(2) Challenges for contractualism

(3) Contractualism and promising

(4) Promising as a normative power

Please also refer back to the lectures last term on aggregation. Notes here:

  1. World Cup.

It’s the evening of the football World Cup final; millions of people are watching the match on television. A worker falls onto some electric cables in the television studio. He’s in agony, though it’s not life-threatening. The only way to rescue him is to turn off the power for the final few minutes of the match. If we do that, millions of people will miss the climax of the World Cup final.  (From T. M. Scanlon, What We Owe to Each Other (Cambridge, MA: Belknap, 1998), 235)

Suppose it’s in your power to stop transmission. What ought you to do?

  1. Utilitarianism and its discontents

How would utilitarians analyse this case? The right thing to do, on their view, is whatever will produce the best balance of pleasure over pain (happiness over suffering, utility over disutility…) But this formulation attributes no significance to the fact that the pleasure/happiness/utility of some could depend on the pain/suffering/disutility of others. So a utilitarian is probably committed to the view that the transmission should continue and that we should just let the worker suffer.

This famously has some very odd, even offensive, implications: e.g. that torture could be right if the torturer got more pleasure out of the torturing than the victim. In Scanlon’s example, the case – which is not nearly as artificial or implausible as some scenarios philosophers have devised – is designed so that the numbers on one side (i.e. the millions of people who’ll be dismayed and angry when the transmission of the match stops in the crucial ten minutes) are so large that however mild their distress, it will outweigh whatever distress the worker feels.

What do you think about Scanlon’s example?

  1. The Individualist Restriction

Intuitively, says Scanlon, we think transmission ought to be halted. (Do we?) If so, why? Scanlon suggests we are committed to an individualist restriction (phrase from Parfit 2003) on ethical principles: ‘the justifiability of a moral principle depends on various individuals’ reasons for objecting to that principle and alternatives to it’ (Scanlon 1998: 229).

This is one way of respecting what John Rawls, a significant influence on Scanlon, called ‘the separateness of persons’.

Indeed, the ethical concerns behind Scanlon’s theory can be traced at least as far back as Kant’s ‘formula of humanity’: roughly, we ought to treat human beings as ends in themselves, not mere means.

  1. Applications of Contractualism


Can Jones reasonably reject a principle (e.g. utilitarianism) requiring him to endure a great pain to allow the transmission to continue? Surely yes! (Put yourself in his place: are you just whining when you reject that principle? Or do you have a reasonable, legitimate claim?) Can a viewer reasonably complain about the end of transmission once he knows what it would have required?

‘A contractualist theory… allows the intuitively compelling complaints of those who are severely burdened to be heard, while, on the other side, the sum of the smaller benefits to others has no justificatory weight, since there is no individual who enjoys these benefits and would have to forgo them if the policy were disallowed’ (p. 230).

As such, contractualism seems better placed to accommodate commonsense views about justice and fairness than utilitarianism, according to which those values are strictly derivative of, and therefore lexically posterior to, utility.

  1. Grounds for Reasonable Rejection


There are, Scanlon says, many possible reasons for rejecting a principle, welfare reasons (to adopt this reason would make me badly off) as well as non-welfare reasons (to adopt this principle would be to treat me unfairly). The standard of reasonableness, as Scanlon conceives it, presupposes that we share the aim of finding principles.

‘The “choice situation” that is fundamental to contractualism as I have described it is obtained by beginning with “mutually disinterested” individuals with full knowledge of their situations and adding to this … a desire on each of their parts to find principles which none could reasonably reject insofar as they too have this desire.’ (Scanlon 1982:  127)

Analogy: think of you and your housemates drawing up a cleaning rota. Someone who wants to be exempted from all cleaning duties, for no other reason than that they don’t like cleaning, is quite naturally described as ‘unreasonable’: but this presupposes that they share some aims with you and your other housemates, e.g. that of having a clean house and for the labour needed to make that happen be shared fairly.

  1. Reasonable rejection and justification

A rough gloss on Scanlon’s principle: if for every alternative principle to P, there is some other individual who has stronger reasons to reject this principle than you have reasons to reject principle P, then it would not be reasonable for you to reject principle P.

Can you think of examples to support / undermine this?

  1. The Scanlonian project

Scanlon’s systematic statement of his theory in What We Owe to Each Other (1998) is addressed to three questions: (1) what is the subject matter of moral right and wrong? (2) what kind of reasoning is involved in determining what is right and wrong? (3) why should we take considerations of right and wrong to be significant in our ethical reasoning?

Scanlon starts with questions (2) and (3) and uses his answers to those questions to get at (1). Scanlon answers questions (2) and (3) in terms of agents justifying their conduct to others: ‘When I ask myself what reason the fact that an action would be wrong provides me with not to do it, my answer is that such an action would be one that I could not justify to others on grounds I could expect them to accept’ (Scanlon 1998: 4).

Is this merely a psychological peculiarity about Scanlon? Or do we all recognise something right in this? In other words, does this ring true? Is moral in fact, as Scanlon’s title suggests, ‘what we owe to each other’?

  1. Next week: Contractualism and the question of aggregation

Scanlon’s theory, in line with many people’s intuition about World Cup, entails that we should halt transmission to help the worker, no matter how many millions will be angry or distressed at this:

‘Because grounds for rejecting a principle must come from the standpoint of some individual, contractualism rules out justification for principles that appeal to the sum of the benefits they bring to different people’ (Scanlon 1998: 241–2).

Does this mean that Scanlon’s theory can’t make sense of the quite intuitive thought that if I have to choose between saving one person’s life and saving five – in a situation where it’s impossible for me to save all six – I ought to save five?

Recommended reading for next week

– Michael Otsuka, ‘Scanlon and the Claims of the Many versus the One’, Analysis, 60 (2001), 288–90.

– Martha Nussbaum, ‘Beyond “Compassion and Humanity”: Justice for Nonhuman Animals’, in Frontiers of Justice: Disability, Nationality, Species Membership (Cambridge: Harvard University Press, 2007).

Contractualism II: Challenges for Contractualism

I. Contractualism and aggregation

  1. Can contractualists aggregate claims?

TM Scanlon’s theory, in line with many people’s intuitions about World Cup (recall the structure: should we halt transmission of the last few minutes of the football world cup final to stop someone being electrocuted?), entails that we should halt transmission to help the worker, no matter how many millions will be (slightly) angry or annoyed at this:

Because grounds for rejecting a principle must come from the standpoint of some individual, contractualism rules out justification for principles that appeal to the sum of the benefits they bring to different people. [What We Owe to Each Other, 241–2]

In this, it seems to constitute an improvement on utilitarianism and similar consequentialist theories, in that it respects what is called ‘the separateness of persons’. Does this mean that Scanlon’s theory can’t make sense of the quite intuitive thought that if I have to choose between saving one person’s life and saving five – in a situation where it’s impossible for me to save all six – I ought to save five? We’re thinking here of cases such as:

Lifeboat: A rescuer with a lifeboat can either rescue one person, or five different people, from drowning, but not all six people.

Quite plausibly, the right principle to use here is that in situations of this sort, one should simply save the greater number. But surely the only way to do this is by aggregation. And it looks like the way in which contractualism gets the ‘right’ answer in World Cup is by denying the legitimacy of aggregation. Can a contractualist get the right answer, for the right reason, in both cases?

  1. Contractualism and the Saving the Greater Number principle

In fact, Scanlon (pp. 229-41) supports the SGN principle, but does so in a non-consequentialist way. He claims not to assume the claim about the aggregation of interests.

Consider a Foot/Anscombe/Taurek-style lifeboat case: should the rescuer, unable to save both, save the smaller group or the larger?

In a case with the structure of Lifeboat, either member of the larger group might complain that this principle did not take account of the value of saving his life, since it permits the agent to decide what to do in the very same way that it would have permitted had he not been present at all, and there was only one person in each group. Some relevant quotations from Scanlon:

The presence of the additional person [in the larger group] … makes no difference to what the [rescuing] agent is required to do or to how she is required to go about deciding what to do. And this is unacceptable, the person might argue, since his life should be given the same significance as anyone else’s in this situation… [What We Owe, 232]

[I]f one of the members of the two-person group were absent then the positive reason for saving the one person would be balanced by an identical reason for saving the remaining member of the pair, thus creating a tie, which is broken by the claims of the other member of the pair, if there is one. [What We Owe, 235]

  1. The Worry: Is Scanlon making a suppressed appeal to aggregation?

Compare these two approaches to Lifeboat­-type cases: Scanlonian vs. ‘pair-wise’. The pair-wise comparer first compares saving A to saving B, then compares saving A to saving C. The two options (assuming that every life has the same weight) are on par. But this is not the result that Scanlon wants. If he thinks we should save B and C rather than A, it can’t simply be by C breaking a tie. It’s only by aggregating B and C and finding the balance tipped in their favour. Isn’t this just aggregation by another name?

If this objection succeeds, what follows? At least this much: that a Scanlonian contractualist can’t really avoid aggregation. This means they need to think carefully about how they can continue to get the right answer in cases such as World Cup. What’s the difference between these two cases such that in Lifeboat, the aggregated claims of the five outweigh those of the one, and in World Cup, not even the aggregated claims of millions can outweigh those of the one?

II. Contractualism and non-human animals

  1. Contractualism and rationality

Contractualism tries to derive moral duties (and therefore, moral wrongness) from something it takes to be more basic, viz. the capacity for reasonable rejection. In the contractualist system, one doesn’t reasonably reject principles because they (or what they permit) is in fact morally wrong; this gets things the wrong way around. Rather, why something is morally wrong – what explains the moral wrongness of something – is simply the fact that it can be reasonably rejected.

Evidently then, it’s tricky for the contractualist to make sense of how some being could have moral standing even though it – being non-rational and therefore lacking the capacity to be sensitive to reasons (in the relevant sense) – lacks the capacity to reasonably reject anything at all. This includes, pre-eminently, (most? all?) non-human animals. But it may also include some human beings – small children and disabled people.

  1. Contractualism and utilitarianism

Again, as with aggregation, contractualism faces difficulties in accommodating moral claims that utilitarians have no difficulty with. For utilitarians, the basic category is sentience – i.e. the capacity to feel pleasure and pain. This means that it’s easy for them to say what (for many people) is the merest common sense, that animals, children and some cognitively disabled people, all in different ways, have moral standing. There is something about them that means that – even if their rational faculties are absent or incipient or undeveloped – it is wrong to do certain things to them. Utilitarianism may or may not give the right account of their moral standing, but unlike contractualism, it seems that it can give some account of it.

  1. Contractualism, morality and trustees

Scanlon has two replies:

(1) The subject matter of contractualism. Scanlon can agree that non-rational beings have moral standing but say that their standing is simply outside the scope of his theory. His is a theory of ‘what we [i.e. rational beings] owe to each other’, not the entirety of morality. This leaves it open that there might be things we owe to other beings. Only, those obligations would not themselves be part of a contractualist theory.

(2) Human trustees. Scanlon also points out that there is quite a simple way to make sense of why it is wrong to do things to animals. All we need is to imagine animals as having human trustees, and that the justifications are to made to them, who them accept or reject them based on reasons grounded in the interests of those they represent [What We Owe, 183]. (It has since been powerfully argued – in Kymlicka & Donaldson, Zoopolis (Oxford: Oxford University Press, 2011) – that there’s nothing absurd about animals having, e.g., lawyers.)

  1. Direct and indirect duties

Is it plausible to think that the best explanation of why it’s wrong to cause needless pain to animals involves justification to the animal’s trustee? Isn’t there a simpler explanation available, viz. because it makes the animal suffer?

A possible reply: Wronging another rational being is the paradigm case of an obligation; our obligations to animals just are different, precisely because it doesn’t involve justifying our actions directly to them. Rather, we have to understand the relevant sense of wrongness in a different way. It is a virtue of contractualism that it makes it obvious that our obligations to other human beings are grounded in a different set of facts than our obligations to animals.

How persuasive is this?

Next week: TM Scanlon, ‘Promises’, in What We Owe to Each Other (Cambridge MA: Harvard University Press, 1998), 295–327.

 

Contractualism III: Contractualism and Promising

0. The mystery of promising

I shall further observe, that, since every new promise imposes a new obligation of morality on the person who promises, and since this new obligation arises from his will; it is one of the most mysterious and incomprehensible operations that can possibly be imagined, and may even be compared to transubstantiation or holy orders, where a certain form of words, along with a certain intention, changes entirely the nature of an external object, and even of a human creature. (Hume, Treatise, 3.2.5–14/15–524)

  1. Non-contractualist views about promising

Act-consequentialism: Breaking a promise is wrong when (roughly) it produces a worse state of affairs than keeping it. On the face of it, this is dangerously contingent and fails to give the sort of assurance a promise is supposed to give (i.e. it may follow that you shouldn’t believe a promise given to you by an act-consequentialist!) It doesn’t ground the kind of robust duty we generally think a promise involves.

Rule-consequentialism: Breaking a promise is wrong because it is forbidden by a rule which is itself justified by the good consequences of its being generally followed.

John Rawls in ‘Two Concepts of Rules’ makes a distinction according to which the wrongness of breaking a promise is understood in terms of the constitutive rules that define the practice of promising, but that practice itself is justified in terms of the practice’s effects on social utility.

This view has some obvious advantages: it demystifies how simply saying some words could generate an obligation; it also improves on the act-consequentialist view in making sense of why we might have obligations to keep promises even when no one is benefited by our doing so.

It also has some well-known disadvantages: it fails to capture the directed nature of promissory obligation, i.e. the fact that the obligation I fail to meet when I break a promise is an obligation to the promisee, not to the institution of promising.

  1. Promising in the state of nature

Imagine a state of nature, defined for our present purposes as some community of human beings who do not already have an established practice of promising. Thomas Scanlon invites us to imagine this situation:

In a state of nature, I accidentally throw my spear across a river. A stranger also accidentally throws her boomerang to my side. We communicate to each other that we will swap the weapons back. The stranger throws back my spear, and I walk off without returning the boomerang.

Scanlon says I have wronged the stranger in the same way that I would wrong him if a promising practice existed and I had promised him, using whatever words one uses to say this once such a practice exists. (Do you agree? Why or why not?)

If promissory obligation exists even in the state of nature, as Scanlon alleges, then it’s a mistake for our account of it to rely so strongly on the social utility or whatever of a promising practice. The notion of a promise can, rather, be related to a more basic set of human interests (in getting robust assurances, the ability to form intentions that rely on other people carrying through on their stated intentions, etc).

  1. Side-note: A genealogy of promise-keeping?

Scanlon doesn’t put it this way, but this could be thought to be a ‘genealogy’ of promising: i.e. a naturalistic understanding of how promises emerge from more basic human needs and interests, but which doesn’t try to reduce them to those needs.

As Bernard Williams puts it in Truth and Truthfulness, a book which explicitly adopts a genealogical method, the hope is to get ‘explanation without reduction’: i.e. we come to understand the basis for valuing such things as truth-telling or promise-keeping in terms of other things, but without simply rendering these notions superfluous. Further, it allows to relate the initially mysterious idea of ‘promissory obligation’ to other, more intelligible and humanly recognisable sentiments, such as expectations, assurances, trust, resentment, blame, anger, betrayal, disappointment.

Obligations don’t disappear in this picture, but they cease to have their metaphysically sui generis quality: they are intelligibly related to other things in life and psychology, not some weird special entity that just exists ‘out there’. The basic idea is not our ‘having obligations’ – like we have (say) hair on our heads – but that of it being proper/right/appropriate/rational for us to regard ourselves as obligated. The objectifying style of discourse about obligations – as if they’re mind-independent objects – comes out as derivative of the more basic idea.

This mirrors the so-called buck-passing account of value that Scanlon favours: value isn’t strictly speaking a property of the valuable thing; it’s rather that we have reason (or alternatively, it is fitting for us) to value certain things.

  1. Elements of an account of promise-keeping

As Scanlon understands it, there’s no magic in the words ‘I promise’ that has the power to ‘create obligations’. Rather, the idea of a promise is a good way of unifying a set of more basic considerations about certain ways of wronging people.

Suppose I were to get you to help me by misleading you (by lying or other means) into thinking I’m going to help you to return, it’s not clear why it matters specially whether I explicitly invoked the notion of a promise. It would be wrong, and wrong in the same way, if I’d only said ‘I’ll help you because it’s in my interest’, ‘I’ll help you because I’m sentimental’, ‘I swear on this holy book that I’ll help you’, and then failed to help.

The words ‘I will help you’, uttered in a certain way in a certain context, are doing exactly what the words ‘I promise’ or ‘I give you my solemn word of honour’. This is reflected in the structure of our sentiments and reactive attitudes: the sense of irritation, resentment, disappointment, betrayal (etc) doesn’t seem to depend on whether the words ‘I promise’ were used. The important thing is that I gave you to believe/expect something and then frustrated or disappointed that expectation. ‘Giving you to believe/expect’ is particularly relevant/useful in situations where there’s reason to think the expectation may not be fulfilled.

Imagine you’re stood up by someone who’d said they’d meet you for coffee at four (leading you to cancel other plans and take a long bus journey into town): would your emotions change if you realised that that words they’d used was ‘I will see you at four’ or ‘Expect me at four’, not ‘I promise you I’ll be at the café at four’. The words ‘I will’ in English can express both a prediction and the statement of an intention: the latter, under certain circumstances, is not all that different from what’s expressed in making a promise.

We can expect a lot of things of each other without the need for any explicit statement to that effect. Consider the slight oddness of me saying ‘I promise you my account of contractualism will be, to the best of my knowledge, accurate’, or ‘I promise you I won’t use this lecture to preach Fascist ideas’, or more darkly, ‘I promise not to kill you’. Good assurances to have, no doubt, but they really shouldn’t be necessary!

  1. Scanlon’s contractualist account of promise-keeping

A contractualist thinks that an act is wrong if it is prohibited by a principle that no one can reasonably reject. What principle could there be that prohibits the set of actions described above?

Scanlon works his way through a number of candidate principles and ends up with a quite complicated principle (is it a problem that it’s so complicated?):

If (1) A voluntarily and intentionally leads B to expect that A will do x (unless B consents to A’s not doing x); (2) A knows that B wants to be assured of this; (3) A acts with the aim of providing this assurance, and has good reason to believe that he or she has done so; (4) B knows that A has the beliefs and intentions just described; (5) A intends for B to know this, and knows that B does know it; and (6) B knows that A has this knowledge and intent; then, in the absence of some special justification, A must do x unless B consents to x’s not being done. (What We Owe to Each Other, p. 304)

Next week: Some challenges to contractualist views of promising; promising as a ‘normative power’

Recommended reading: David Owens, Shaping the Normative Landscape (Oxford: Clarendon Press, 2012), chs. 5 & 8.

Contractualism/Promising IV: Promising as a Normative Power

0. Recap.

Scanlon’s principle of fidelity: If (1) A voluntarily and intentionally leads B to expect that A will do x (unless B consents to A’s not doing x); (2) A knows that B wants to be assured of this; (3) A acts with the aim of providing this assurance, and has good reason to believe that he or she has done so; (4) B knows that A has the beliefs and intentions just described; (5) A intends for B to know this, and knows that B does know it; and (6) B knows that A has this knowledge and intent; then, in the absence of some special justification, A must do x unless B consents to x’s not being done. [What We Owe, 304]

  1. Two ways of thinking about Scanlon

Scanlon’s underlying normative theory is contractualist: that provides his general way of assessing all moral principles (specifically: moral principles related to ‘what we owe to each other’). That means that his account of promissory obligation is derivative of a more basic set of obligations, viz. obligations not to ‘unfairly manipulate’ other people, in particular, to manipulate them by deceiving them. This covers both (1) making false or insincere promises, and (2) breaking a sincere promise. In both cases, one has brought it about that others believe that one will do what one has promised to do, and thereby, raised expectations that one will fail to meet.

Scanlon thinks his principle of fidelity cannot be reasonably rejected. Why? Crudely: because the interests promisees have in not being deceived, and therefore the reasons grounded in those interests, are generally weightier than the interests promisors have in deceiving, and the reasons grounded in those. Thus, an ‘infidelity’ principle that permitted promise-breaking could be reasonably rejected; the fidelity principle can’t.

The idea that is central to his account of promissory obligation, specifically, is that of an expectation. Call views of this general kind expectationalist views; one an affirm this view without being a contractualist. One could, e.g., be a consequentalist who thinks that promise-breaking is wrong, when it is wrong, because of the harm it causes the promisee whose trust has been betrayed.

  1. Distinguishing promises from other things

In its barest form, an expectationalist conceives of a promise as a device used (by the promisor) to produce expectations (in the promisee). But one immediate difficulty presents itself: things other than promises produce expectations. Compare ‘I promise to φ’, ‘I threaten to φ’, ‘I advise you that…’, ‘I warn you that I’m going to φ’ These all could (depending on what φ is and some feature of the context) produce expectations in the hearer. But not all of them create obligations and the corresponding rights.  One way of thinking about this is to think about whether you feel wronged – or more specifically, betrayed, disappointed or merely surprised – if the expectation isn’t met. Consider these examples: ‘I give you my word – I’ll pay back the money.’ ‘The cheque is in the mail.’ ‘It’ll be sunny tomorrow, don’t bother taking an umbrella.’ ‘I’m going to be sick!’

A side-note: it’s not obvious that an account of promissory obligation has to mark a sharp contrast between promises and other ways of raising expectations, inducing trust, etc. Maybe it’s a virtue of expectationalist views that it shows that there’s a deep continuity between breaking a promise and other wrongs (e.g. telling lies, giving misleading advice, leading someone on…) [E.g. Elinor Mason, ‘We Make No Promises’, Philosophical Studies (2005), 123(1–2): 33–46.]

  1. Vicious circularity?

Consider the following chain of events: (1) A says to B that he will φ. (2) B now believes A will φ (trusts, expects A to φ, etc). (3) A now has a (promissory) obligation to φ – because of (1) and (2).

Focus on (2). What reason has B to believe/trust A? Intuitively, it is the fact that A has promised to φ, that A has acquired a promissory obligation to φ, and that B believes A not to be the promise-breaking sort. But expectationalists can’t say this, because on their view, promissory obligation itself is explained by the prior fact of trust. If they do say this, then on their view: promissory obligation is grounded in trust; trust is grounded in promissory obligation. This looks like a vicious circle.

How to break out of the circle? An expectationalist has to give an alternative explanation of why the promisee should trust the promisor. This may involve supplementing the expectationalist account with elements drawn from a different theory (e.g. conventionalism – though Scanlon himself won’t want to draw on this given that he rejects it for independent reasons).

  1. Profligate pal?

‘Your friend has been borrowing money from you, and from others, for years, always promising solemnly to pay it back but never doing so. Finally, you refuse to lend him any more money, and others do so as well. This precipitates a crisis of shame. Your friend is humiliated by the realization that others have lost all respect for him, and he struggles to retain the last vestiges of respect for himself. He is also in great need of money. Finally, he comes to you on his knees, full of self-reproach and sincere assurances that he has turned over a new leaf. You do not believe this for a minute, but out of pity you are willing simply to give him the money he needs. You realize, however, that it would be cruel to reject his promises as worthless and offer him charity instead. So you treat his offer seriously, and give him the money after receiving his promise to repay the loan on a certain date, although you have no expectation of ever seeing your money again. Does he have an obligation to pay you back?’ [What We Owe, 312]

Scanlon’s response: Intuitively, yes. Scanlon accepts that his principle gives the ‘wrong’ answer and that this is a problem. But he says this doesn’t matter; after all, this is an ‘impure’ case. Certainly, the contractualist account needs to be supplemented with other principles. But what matters is that it can deal with all the ‘pure’, paradigmatic cases.

What supplementary principles might these be? Maybe: You have a right to expect repayment, even if you don’t actually expect it. Or maybe what you’ve done is waived your obligation by writing off any expectations. Or maybe your pal doesn’t actually have an obligation to pay things back, though – given that he doesn’t know you have no expectations of him – he should continue to believe that he does. So even if you won’t be surprised/betrayed/etc if (i.e. when) he doesn’t cough up, you can still be disappointed with him. And indeed, if (per impossibile!) he did pay up, it would be odd to say – ‘I wasn’t expecting it, so you should keep the money’. Surely you should just accept it!

  1. Another approach: Promise as a normative power

A long tradition in western philosophy (Aquinas, Locke, Reid, Pufendorf) conceives of promising as a kind of ‘normative power’, a power we human beings have over our ‘normative landscape’, i.e. the distributions of rights and obligations. Promissory obligation is, on this view, self-conferred; it comes of the special exercise of a ‘normative’ power.

One way of understanding promisor’s obligations and promisee’s rights is by using the classic ‘Hohfeldian’ analysis of rights and duties. The idea is that our ordinary thoughts about rights and duties have a complex, but entirely recognisable, internal structure that we can represent in terms of four components (or ‘Hohfeldian incidents’).

Privileges: A has a privilege to φ if and only if A has no duty not to φ.

Claims: A has a claim that B φ if and only if B has a duty to A to φ.

Powers: A has a power if and only if A has the ability to alter her own or another’s Hohfeldian incidents. (E.g. Ppromising, ordering, sentencing, waiving, consenting, selling, abandoning, etc)

Immunity: B has an immunity if and only if A lacks the ability to alter B‘s Hohfeldian incidents.

Claims and privileges define (what HLA Hart later termed) ‘primary rules’, rules requiring that one do or refrain from doing something; i.e. it specifies obligations and permissions. Powers and immunities define ‘secondary rules’, i.e. rules for altering primary rules. When this analytic framework is used to make more substantive claims about what rights people have, the basis that is generally invoked is people’s interests. Recent work in moral philosophy (e.g. Seana Shiffrin, David Owens) has conceived of promissory obligation as generated by a normative power which is itself grounded in certain human interests. E.g. our interests in forming certain kinds of inter-personal relations (Shiffrin), or our authority interest, i.e. the interest we have in having a certain practical authority over others, in binding their wills in a certain way (Owens, Shaping the Normative Landscape). The normative power of consent, on the other hand, is grounded in our permissive authority – to release people from obligations they would otherwise have. Owens:

the obligations one takes on when one promises to do something are significantly different from those one acquires by expressing the intention to do it or predicting that one will do it or by raising expectations that one will do it in some other way. In promising one hands over the authority to decide whether one does it to one’s audience; one does not merely incur the obligation to take their expectations into account when deciding what to do. [Landscape, 207]