Breaking Up and the Value of Commitment

  • Richard Healey (The London School of Economics and Political Science)


While love and personal relationships are the subjects of rich and sophisticated literatures, philosophical writing about the end of special relationships is much harder to come by. However, the end of special relationships is a significant part of our lives and gives rise to a number of philosophical questions. In this article, I explore the normative significance of the end of special relationships, with a particular focus on the case of breaking up in the context of committed romantic relationships. Specifically, I address three questions. First, what does A do when A breaks up with B? Second, what normative effect will A’s breaking up with B have on the relationship-based duties, reasons, and permissions that are partly constitutive of A and B’s relationship? Third, how is the ability to break up consistent with the commitment that many longer-term romantic relationships involve? In response to the first and second questions, I argue that breaking up is a neglected example of a normative power and develop a tripartite account of the normative effects of exercising this power. In response to the third question, I develop an account of the nature and value of commitment within romantic relationships and show how the power to exit a relationship by breaking up is consistent with this form of commitment.

How to Cite:

Healey, R., (2023) “Breaking Up and the Value of Commitment”, Ergo an Open Access Journal of Philosophy 10: 6. doi:



Published on
17 Nov 2023
Peer Reviewed

Developing and maintaining close personal relationships is a central part of most people’s lives. Indeed, such relationships are sites for some of the most important human experiences and activities, including love, care, support, and interpersonal understanding. However, it is not the case that as our lives go on, we form and maintain an ever-greater number of relationships. Rather, relationships come and go, for reasons both within our control and outside of it. Friends and family pass away, and the phenomenon of “drifting apart” is familiar to most. In other cases, we take active and deliberate steps to end a relationship. For instance, we may seek to end a relationship because we have realized that we are “not compatible” with someone or choose to exit a romantic relationship because doing so is a precondition for entering into another.

While love and personal relationships are the subjects of rich and sophisticated literatures, philosophical writing about the end of special relationships is much harder to come by.1 Perhaps there is a natural aversion to what is, very often, a matter of deep regret. But whatever the explanation, the end of special relationships is a significant part of our lives and gives rise to a number of philosophical questions.

In this article, I explore the normative significance of the end of special relationships, with a particular focus on committed romantic relationships. Most of us believe that our participation in such relationships will have a significant impact on the reasons, duties, and permissions to which we are subject.2 For example, spouses are subject to a variety of relationship-based duties, such as duties to care for and support one another, to help the other pursue their goals, and to feel pleased if things go well for one’s spouse and upset if they do not. These duties are pervasive and often demanding. What happens to these relationship-based duties, then, when a relationship comes to an end?

I begin to address this question by focusing on a particularly stark context in which it arises: that in which one person breaks up with another. As I will use the phrase, A breaks up with B when A declares that they are ending the relationship (e.g., “It’s over, B”). Of course, this is not the only way in which a romantic relationship can end. Yet it is an especially interesting case for two reasons. First, as I elaborate in Section 1, we do not normally think of special relationships as things that can be created or dissolved by declaration. Second, many acts of breaking up end (or at least aim at ending) relationships that are avowedly committed relationships. Of course, what exactly commitment amounts to in this context is something that itself requires attention, and I develop an account of the nature of commitment within romantic relationships below. But whatever we say, we normally think that commitments are not things that we can get out of by merely declaring that we are no longer bound.

With these issues in view, the present article seeks to answer three questions. First, what does A do when A breaks up with B? I will argue, in Section 2, that breaking up is a neglected example of a normative power: an ability to alter the normative relationship that one stands in by declaring the intention of so doing. If this is correct, it prompts a second question, namely, what normative effect will A’s breaking up with B have on the relationship-based duties, reasons, and permissions that are partly constitutive of A and B’s relationship?3 An initial possibility, considered in Section 3, is that the breakup simply cancels all A and B’s relationship-based duties. Yet this, I argue, is generally implausible. We continue to owe at least some significant special duties to someone with whom we have broken up or who has broken up with us, at least for a period. And at least one common reason for this, I argue in Section 4, derives from the value of commitment.

Nevertheless, many people recognize the need to be able to exit even long-standing and committed romantic relationships. This gives rise to the third question: how is the power to break up consistent with the commitment that many longer-term romantic relationships involve? To better understand the normative effects of breaking up, and the compatibility of commitment with the power to break up, Sections 4 and 5 take a step back to consider the nature and value of commitment and voluntariness within romantic relationships. With this discussion of commitment and voluntariness in place, Section 6 then develops an account of the normative effects of breaking up. In short, I argue that exercising the power to break up enters a relationship into a transitional phase. While some relationship-based duties will persist for at least a period, others will be dissolved by the breakup, and there will also be a new normative pressure on the participants to take effective steps to transition out of the relationship.

Before proceeding it is important to note that no discussion of this kind could hope to do justice to the wide variety of valuable romantic relationships that people engage in. My focus here is on what I take to be a familiar kind of romantic relationship that is widely valued. With this in view, I hope to trace out the normative presuppositions and implications of such a relationship, rather than offer a template for what all (or perhaps any) romantic relationships should look like. No doubt, there are other valuable forms of romantic relationship that involve, for instance, different ideals of commitment. Moreover, I certainly do not mean to imply that the kind of romantic relationship under discussion is unique to heterosexual couples such as the one involved in my main example. I focus on this stylized case merely to highlight what I think are culturally prominent ideas about the end of relationships and the value of commitment so as to subject those ideas to philosophical scrutiny and enhance our understanding of an important feature of many people’s lives.

1. The Beginning and End of (Some) Relationships

To focus on the phenomenon that is my central concern, and to make progress in understanding that phenomenon, it will help to have an example to work with.

Breakup: Jamie and Fred have been in a committed romantic relationship for fifteen years. However, after careful consideration, Jamie decides that she no longer wants to continue in the relationship. One evening, she sits Fred down and says, “There is something I have to tell you. I am no longer happy in our relationship, and I want to break up. I’m really sorry.”

I take Breakup to exemplify a familiar way in which one person ends (or seeks to end) a romantic relationship with another. To be sure, I am not claiming that all romantic relationships end in this way. Some romantic relationships will, for instance, unravel over a period without anyone ever deliberately communicating that they are ending the relationship. Moreover, many relationships will have begun a process of dissolution (e.g., a weakening in emotional and normative significance) before one or both parties explicitly seek to end the relationship.

My focus, however, is on discrete and deliberate acts aimed at ending a relationship, and I will reserve the term “break up” and related phrases to refer to such cases. Some may be unhappy with this terminological choice. Yet one reason that these cases are especially interesting is that, as noted above, we do not normally think of relationships as things that can be created or dissolved by declaration. As Joseph Raz observes, the obligations of special relationships like friendship are clearly voluntary in that they depend on the development of a relationship that one could have avoided and where one understands the normative implications of developing a relationship of that kind. However, they are importantly distinct from other forms of voluntary obligation, like promises:

Since the obligations are part and parcel of the relationship they cannot be assumed or renounced by one act. . . . The friendship itself, involving an intricate web of reciprocal dispositions and attitudes, practical, emotional, and cognitive, cannot be created by an act of commitment. It has to grow, develop, and cement over time. (Raz 2009: 257)

Raz’s claim in this passage seems right. We do not normally talk about creating a friendship—with all the associated affective and normative elements that will imply—by a single act of commitment.4 To be sure, specific acts of commitment (e.g., to go for dinner, embark on a joint project, etc.) often play an important role in the development and maintenance of relationships. Furthermore, in many romantic relationships, acts of commitment that take the relationship as their object, like marriage, are relatively common. Yet such acts (whether in public or private) are surely not necessary to establish a relationship as a committed relationship. Indeed, as Talbot Brewer observes, setting aside purely legal examples, it would seem very odd to interpret the utterance of marriage vows as the establishment of a relationship and its associated obligations, as if that relationship and those obligations did not already exist (Brewer 2003: 563–64). To be sure, such vows may make an important normative difference to the relationship. Yet they most plausibly do so by publicly revealing or affirming a commitment to the relationship that the couple already mutually recognize.

So, thick personal relationships such as romantic partnerships must be developed in myriad ways over time. This is a deliberate process carried out by individuals who want to develop a relationship with one another. But the process itself cannot be bypassed through single acts of commitment like making a promise. Raz continues by suggesting that the same is true of the termination of friendships: “Normally no single act can put an end to the myriad elements making one person another’s friend” (Raz 2009: 257). This also seems plausible, at least regards friendship. Yet, as I have said, acts that appear to be aimed at ending romantic relationships are not uncommon. At least on the surface, then, there is an asymmetry between the possible ways of developing a romantic relationship and ending such a relationship. Yet if we are not able to create a relationship by declaration, how is that we can end or dissolve such a relationship by declaration? Furthermore, how is this possibility consistent with the fact that many romantic relationships profess to be committed relationships?

In developing my account of breaking up and the value of commitment in what follows I aim to make sense of this asymmetry by treading something of a middle way. In an important respect, Jamie’s single and deliberate act of breaking up with Fred will fundamentally restructure their relationship. Nevertheless, Jamie cannot straightforwardly end the relationship, in the sense of dissolving all its associated normative requirements, through this single action. Only the passage of time and the gradual disentangling of Jamie and Fred’s lives will achieve this result.

2. Breaking Up as an Exercise of Normative Power

What is Jamie doing when she says to Fred, “There is something I have to tell you. I am no longer happy in our relationship, and I want to break up”?5 An initial suggestion is that she is communicating that she no longer loves Fred. Yet it seems clear that communicating an end to love is neither necessary nor sufficient for breaking up with someone. We are all familiar with examples of people who break up with someone despite being head over heels in love with them (e.g., because they think they are bad for them, or want to pursue a career elsewhere). We are also familiar with examples of couples who acknowledge that their love for one another has dwindled but who commit to working on their relationship.

Perhaps by breaking up with someone we are communicating a change in our intentions or plans. In our example, we might think that in breaking up with Fred, Jamie is communicating to Fred that, from now on, she no longer intends to spend time with him, engage in various activities together, live with him, and so on. Of course, Jamie could communicate some such changes in her intentions without breaking up with Fred. Perhaps she intends to spend less time working on the garden with Fred to meet an important deadline at work. Yet if Jamie is communicating the intention of disengaging from all or at least most future activities and interactions, perhaps this communication constitutes the act of breaking up.

However, reading Jamie’s declaration as equivalent to her saying, “I no longer intend to spend time with you, live with you, pursue joint projects with you . . .” rings hollow. The main reason for this concerns the normative significance of a breakup. It seems clear that a breakup will bring about a significant normative change in a relationship. Before breaking up Jamie and Fred have relationship-based reasons, duties, and permissions, for example, to care for one another, engage in shared activities together, and share property. Yet the breakup will have a significant impact on this web of relationship-based duties. For example, once Jamie has broken up with Fred, they will no longer have the same reasons to spend time together, pursue joint projects together, and so on.

These normative effects are something that an adequate account of breaking up should be able to explain. But the change of intentions model cannot adequately account for these changes. This is not necessarily because our intentions do not have normative significance (although, to be sure, the normative significance of intentions is a matter of controversy). Plausibly, if I intend to go to the library this afternoon to work, then I have a reason to take the steps necessary to my getting to the library and working, such as checking the bus timetable, putting my shoes on, and walking to the bus stop on time. Even so, the normativity of intentions cannot explain the normative changes that result from a breakup. For one, the normativity of individual intentions is not a matter of interpersonal obligations of the kind that structure Jamie and Fred’s relationship. It is rather the normativity of rationality, which requires, for example, coherence and consistency in means-end reasoning. Furthermore, our intentions and their associated normative upshots are generally revocable, especially in the light of new information. If I discover there is a large parade in town this afternoon and that all public transportation is suspended until tomorrow, then I may have good reason to revise my intention to go to the library this afternoon. However, Jamie is not able to unilaterally nullify the normative effects of the breakup by (re-)revising her intentions. And that is so even if she seriously regrets her decision in light of new information, such as a realisation about how much she misses Fred. Finally, the normative force of intentions follows from an individual’s having the relevant mental state, for example, my intending to go to the library. But the normative significance of the breakup does not depend on Jamie’s simply having certain intentions, such as intentions to move out and disengage from future projects with Fred or even the intention of breaking up. Rather, for Jamie to break up with Fred, she will need to communicate to Fred that she is breaking up with him.

A more plausible proposal is that Jamie’s act of breaking up with Fred communicates the withdrawal or revision of an intention (or intentions) that Jamie contributes to Jamie and Fred’s shared intentions.6 While there are numerous philosophical accounts of shared intention, a common idea, at least in broad outlines, is that A and B share an intention to ϕ when both A and B intend that they ϕ together and these intentions are common knowledge between A and B.7 For instance, you and I have a shared intention to put up the tent in the garden if I intend that we put up the tent, you intend that we put up the tent, and our intentions are common knowledge.

It seems clear that special relationships involve a variety of shared intentions of varying degrees of specificity, for example, to go on holiday together, raise the children together, do the washing up together, and so on. Furthermore, it is widely recognized that a shared intention held by A and B to ϕ often generates a mutual obligation between A and B that they do their respective part in ϕ-ing.8 For instance, all things being equal, if you and I have formed a shared intention to put up the tent, we owe it to one another to play our respective parts in putting up the tent. Given this, if in breaking up with Fred Jamie is declaring the withdrawal of a previously shared intention, this might explain why Jamie’s declaration impacts Jamie and Fred’s relationship-based duties. Furthermore, the shared intentions model might be thought to explain why Jamie would need to communicate the change in her intentions to Fred to bring about the normative effects of a breakup, since such communications are normally how we navigate changes in our shared intentional activities.

An initial question about this proposal concerns the nature of the shared intentional activity that Jamie is withdrawing her intention from. As I just suggested, special relationships involve a wide variety of shared intentions. Whatever the previously shared intention that Jamie is held to be communicating that she no longer has when she breaks up with Fred, it will need to be sufficiently general to capture the idea that in communicating the withdrawal of that intention she is ending her and Fred’s relationship. One possibility is that the parties to an ongoing committed relationship will have a higher-level shared intention to maintain their relationship. Such a shared intention might regulate subsidiary plans, such that when this shared intention lapses it also impacts the subsidiary intentions that structure and organize their ongoing relationship.

While an improvement, I think the shared intentions model still fails to accurately capture what is going on when Jamie breaks up with Fred. To begin, while appealing to shared intentions can explain why Jamie’s declaration is about Jamie and Fred’s interpersonal obligations, rather than intra-personal norms of rationality, the shared intention model does not explain why this declaration should make a normative difference to those obligations. Let’s assume that in virtue of forming a shared intention to maintain the relationship at T1 Jamie and Fred acquire an obligation to one another to maintain their relationship. If this is true, Jamie’s telling Fred that she no longer intends to do her part in maintaining the relationship at T2 looks like nothing more than a declaration that she does not intend to fulfill her obligation. Compare a case in which I promise to pick you up from the airport at T1, but at T2 declare that I do not intend to pick you up as promised. While my communication tells you something important, it does not by itself make a difference to the normative relation that obtains in virtue of my promise (unless, perhaps, I am informing you about an extenuating circumstance that justifies my non-performance). Unless you release me from the promise, I will still owe it to you to pick you up. Similarly, if Jamie’s breaking up with Fred is read as simply communicating a lapse in her intention to maintain their relationship, it is unclear why this communication would do anything more than communicate that Jamie does not intend to act as her obligation enjoins.

Furthermore, I think the claim that Jamie’s aim in declaring that she is breaking up with Fred is solely to inform him about her mental state (i.e., the revision of her intention) misconstrues that aim. On this view, Jamie’s saying “I want to break up” or “I’m breaking up with you” is akin to Jamie’s saying, “I am sad” or “I plan to go to the shops on the way home.” These communications aim to inform Fred about Jamie’s mood or intentions. Granted, such changes in our mental states often supply others with new or altered reasons for action. For instance, such communications may alert Fred to various reasons for action, for example, to give Jamie a hug, or not to buy any extra groceries. Yet it seems that when Jamie breaks up with Fred, she is not merely aiming to inform him about a lapse in her (previously shared) intention, thereby indirectly alerting him to a change in his reasons and obligations. Rather, in declaring that she wants to break up Jamie aims to affect their normative relationship, indeed, to fundamentally redraw the web of relationship-based duties to which they are subject.

We can further support this claim by observing that Fred could learn about Jamie’s lapse in shared intention indirectly, but without it nullifying the need for or normative significance of Jamie’s breaking up with Fred. Imagine that Jamie confides in a mutual friend, Mary, that she no longer intends to maintain her relationship with Fred. Feeling that Fred ought to know, Mary relays this information to Fred, before admitting to Jamie that she has done so. In this case, Jamie’s lack of intention to maintain Jamie and Fred’s relationship is common knowledge. While this knowledge will make a difference to the relationship, and such a relationship may be destined to unravel unless something changes, relationships can and do persist in this kind of uneasy state for some time. Indeed, some relationships recover from such difficult periods. However, if at a certain point during this period Jamie sits Fred down and tells him “I want to break up,” it does not seem accurate to claim that this declaration merely restates something Fred already knew (and that Jamie knew that Fred knew). Rather, it seems that Jamie aims to fundamentally alter the nature of their relationship. Before the breakup, unless Jamie and Fred’s relationship had broken down in a more thoroughgoing manner,9 there remains normative pressure on Jamie and Fred to address the difficulties they face, at least in the context of a committed relationship. Yet once Jamie has communicated that she is breaking up with Fred this normative pressure dissolves, and they are instead required to begin taking steps to transition out of their relationship. So, while shared intentions play a significant role in special relationships, and Jamie’s breaking up with Fred will have a significant impact on various shared intentions, I suggest that the revision of these shared intentions is for the most part downstream of the normative changes Jamie engenders by breaking up with Fred.

To recap, Jamie’s act of breaking up with Fred will have a significant effect on the web of relationship-based obligations they owe to one another. To bring about these normative effects, Jamie must communicate to Fred that she is breaking up with him. Furthermore, as I have just argued, it is plausible that Jamie aims to bring about these normative changes by breaking up with Fred. She does not want to hurt Fred, inevitable as that may be. And she is not merely communicating that she no longer has a (perhaps previously shared) intention to maintain the relationship, or that she does not intend to fulfill her relationship-based obligations. Rather, Jamie no longer wants to be in a relationship with Fred, and thus, no longer wants to be bound together by the reasons and duties that structure an ongoing romantic relationship.

If these remarks are correct, then breaking up bears the hallmarks of a normative power.10 Roughly speaking, normative powers—such as the power to consent and the power to promise—enable individuals to bring about changes in their normative relationships deliberately and directly. For example, when I sign a medical consent form, I intend to change the normative relationship between the surgeon and myself by giving her permission to operate. Similarly, when I say that “I promise to pick you up at 8,” I communicate my intention of placing myself under a duty to pick you up at 8. My claim is that Jamie’s communication that she wants to break up with Fred can likewise be interpreted as the communication of an intention to restructure their normative relationship.

To support this claim, it will help to be somewhat more precise about normative powers. Unsurprisingly, there is disagreement about how exactly we should conceptualize normative powers. Yet by and large, I think these disputes are orthogonal to my central line of argument. I aim to avoid these disputes by outlining a set of relatively restrictive conditions that I propose are jointly sufficient for an act’s qualifying as the exercise of normative power. I will then show that paradigm cases of normative power (namely consent and promise) meet these conditions, before arguing that Jamie’s act of breaking up with Fred also meets these conditions.

I propose that A successfully exercises a normative power over B if:

  • i)

    A intends to change A and B’s normative relationship in respect ϕ by communicating her intention to do so,

  • ii)

    A communicates her intention to change A and B’s normative relationship in respect ϕ to B, and

  • iii)

    A intends B to recognize her intention to change A and B’s relationship in respect ϕ, and B does recognize A’s intention, and

  • iv)

    A’s ability to change A and B’s normative relationship in respect ϕ is justified by the value of A’s being able to bring about ϕ through acts of this kind.11

For instance, when I sign a medical consent form, I intend to give the surgeon permission to operate, I communicate this intention by signing the consent form, and I intend for the surgeon to recognize my intention to alter our normative relationship in this way. Moreover, my ability to bring about this normative change is plausibly justified by the value of my having such a power given the importance of my being able to make autonomous decisions about my own life. Similarly, when I promise to pick you up at 8, I intend to place myself under a duty to pick you up at 8, I communicate this to you, and in so doing intend that you recognize my normative intention. And, while there is extensive disagreement about the values that underpin the power to promise, it is plausible that my ability to bring about this normative change is justified by, for example, the values of being able to create a special bond with the promisee and provide assurance that one will act as promised.12 All being well, both you and the surgeon will recognize my communicated intention, and I will thereby succeed in giving the surgeon permission and in placing myself under a duty to pick you up.

With these examples in mind, I submit that Jamie’s communication in Breakup is a further example of the exercise of normative power. We can interpret Jamie’s declaration as follows:

  • i)

    Jamie intends to change Jamie and Fred’s normative relationship in respect ϕ by communicating her intention to do so (“I want to break up”),

  • ii)

    Jamie communicates her intention to change Jamie and Fred’s normative relationship in respect ϕ to Fred,13 and

  • iii)

    Jamie intends Fred to recognize her intention to change Jamie and Fred’s relationship in respect ϕ, and Fred does recognize Jamie’s intention,

  • iv)

    Jamie’s ability to change Jamie and Fred’s normative relationship in respect ϕ is justified by the value of Jamie’s being able to bring about ϕ through acts of this kind.

My argument for this claim is an inference to the best explanation. As I noted above, Jamie’s communication will have a significant impact on Jamie and Fred’s relationship-based obligations. Of course, the nature of the normative change that breaking up occasions remains open at this stage, and I will turn to begin addressing that question momentarily. But unlike the other proposals considered, interpreting Jamie’s action as the exercise of a normative power can straightforwardly explain why it would have a significant normative impact on Jamie and Fred’s relationship. Indeed, normative powers are often held to play an important role in generating, modifying, and dissolving both shared intentions and the obligations associated with shared intentions.14 So appealing to the idea of normative power can help to explain how Jamie’s declaration can remould the pattern of shared intentions involved in their relationship.

Furthermore, we saw that Jamie will need to communicate that she is breaking up with Fred to realize these normative changes. Such communication is a very common (and arguably necessary) feature of the successful exercise of normative power. And since Jamie’s action is not plausibly interpreted as the communication of an emotional state or a change of a sole or joint intention, it seems reasonable to infer that Jamie aims to affect their relationship by revising the normative requirements to which they will both be subject moving forward.

So far, I have not said anything about condition iv) concerning the values that justify Jamie’s normative power to break up. Whether or not we can develop an adequate justification of the power will depend on the nature of the normative changes at issue. As such, a fuller discussion of this condition will have to wait until we have an account of those changes on the table. We can, however, provide initial support for the claim that this condition is met by appealing to the value of voluntariness within romantic relationships. As I discuss below, while many people value the fact that their relationships are committed, they also positively value the fact that their relationships are voluntarily maintained. This implies there must be means of exiting the relationship. The power to break up provides one such means. This raises the question of what is valuable about the ability to exit (or begin the process of exiting) the relationship in this way, that is, through the exercise of normative power. I briefly return to address this question in Section 7.

3. Breaking Up: A First Pass

If I am correct in claiming that breaking up is a normative power, what is the normative effect that exercising that power realizes? How exactly does the fact that Jamie has broken up with Fred alter the normative status of their relationship? In the cases of consenting and promising, the normative effects at issue are relatively fine-grained and determinate. When I give permissive consent, I intend to release another from a duty (e.g., the duty not to operate). When I make a promise, I intend to place myself under a duty (e.g., a duty to pick you up). By contrast, the normative effects of breaking up are somewhat obscure. Moreover, the normative effects of a breakup appear to be much more general. Jamie does not aim to create or revoke any specific duty, but rather to end the relationship, and thus, affect the whole web of relationship-based duties to which Jamie and Fred are currently subject.

It might be suggested that when one person breaks up with another, they simply cancel all the relationship-based reasons and duties that apply to them. In so doing, they take themselves and their partner back to a kind of normative baseline, akin to the normative relationship we stand in with any other person. Thus, while Jamie and Fred will continue to owe one another familiar moral duties (e.g., not to lie, injure, steal, and the like), they will no longer owe one another any special kind of consideration.

On reflection, I think this suggestion is implausible. Consider the following example.

Sudden Illness: Soon after Jamie breaks up with Fred, Fred is taken ill with a serious infection and is hospitalized. Fred’s family live out of town, and he is unable to let them know or to inform work. He would also be much more comfortable if someone were to bring him some home comforts.

In this situation, I submit that Jamie has a relationship-based duty to help Fred. To be sure, anyone who can help Fred would have a reason to do so. Yet it seems that Jamie has, at the least, a significantly more stringent reason than someone who bears no special relation to Fred. This is not to say that things are no different in light of the breakup. All the example aims to show is that the claim that Jamie cancels all of Jamie and Fred’s relationship-based duties and reasons by breaking up with Fred is implausible.15

Someone might suggest that we can provide an alternative explanation for Jamie’s duty to help Fred in Sudden Illness. For instance, by breaking up with Fred, Jamie will have caused him significant pain and anguish. Plausibly, if A causes B emotional or physical pain then A has a special reason to help and assist B, even when that assistance will relieve pain that has a different source (in this case, pain caused by the illness rather than by the breakup). On this view, the duties to which A is subject after breaking up with B have the same basis as other remedial duties (e.g., duties of compensation). Thus, it might be claimed that while Jamie does cancel all of Jamie and Fred’s relationship-based duties and reasons by breaking up with him, she also incurs further duties to Fred because breaking up with him causes serious emotional distress.

Yet even if Jamie will incur remedial duties for this reason, it would be a mistake to infer that Jamie’s breaking up with Fred means that they are both otherwise free of relationship-based duties. One way to see this is to consider a situation in which the breakup is amicable, and Fred is, if anything, relieved that Jamie has initiated the breakup. He too has been thinking that he wants to take his life in a different direction. Nevertheless, it seems clear that Jamie has a special duty to help Fred, and that Fred would likewise have special reasons to help Jamie, at least for a certain period.

4. The Nature and Value of Commitment

So far, I have argued that Jamie’s breaking up with Fred in the context of an established romantic relationship will cancel neither all of Jamie’s nor all of Fred’s relationship-based duties. There are various possible reasons for this. One reason concerns the grounds of at least some relationship-based duties. For instance, if Jamie and Fred owe one another duties of reciprocity, grounded in their historical provision of benefits to one another, these duties will not (at least usually) be dissolvable through Jamie’s exercise of a normative power.

No doubt there are a variety of further reasons why individuals cannot “opt-out unilaterally” of the duties implicated in different kinds of relationship.16 However, in this section, I argue that one important reason for the normative limits on Jamie’s power to break up derives from the value of commitment. Specifically, I claim that the value of commitment inheres in a relationship partly in virtue of that relationship’s having a certain normative structure. This structure has implications both for the duties that obtain within an ongoing relationship and the way these duties can be dissolved. Of special importance here, I argue that if Jamie or Fred had the normative latitude to cancel all the relationship-based duties to which they are subject at will, they would not be participating in a (certain form of) valuable committed relationship, whatever other values their relationship might realize. To develop this claim, it will help to outline an account of the central features of ongoing committed relationships, before situating these in relation to constraints that apply to the power to break up.

All special relationships involve norms that distinguish those relationships from the normative relationship that we stand in with all people in virtue of their basic moral status. Furthermore, many such relationships are also “attitude-dependent” in that they depend upon the (historical and contemporaneous) presence of a certain pattern of attitudes such as concern and good-will.17 On the view I will propose, committed relationships centrally involve a commitment to a shared future, which has both normative and attitudinal dimensions. Let me elaborate on each of these dimensions in turn.

I assume that we are required to give a romantic partner’s interests and ends a special weight in our practical reasoning, and thus, to be partial toward them. Call this, for want of a label, the relationship-based duty of care and concern. This general duty of care and concern will imply a wide variety of specific duties on a day-to-day basis, evolving in accordance with the evolving circumstances of our lives. We might have a duty, for instance, to help our partner with their latest job application, to cheer them up if they have had a bad day, or to listen to their concerns about a family matter.

One important feature of committed relationships is that the duty of care and concern is open-ended. To better understand this idea, compare a paradigm form of commitment, namely, a promise. Imagine that I promise my neighbour to water their garden when they are on holiday next month. While I am not normally subject to a duty to water my neighbour’s garden, my promise ensures that I now have a promissory duty, which gives me a special reason (perhaps especially weighty or partly exclusionary in structure) to act in protection of their horticultural interests in the future. In this respect, my promise binds me. My neighbour and I both understand that (all else being equal) my future self will be required to give special consideration to their garden and its water levels.

The idea is that within committed relationships, the same structure obtains regarding the much more general duty of care and concern. The cases can be distinguished in terms of the scope and the duration of the duty. In terms of scope, my promissory duty is restricted to making sure my neighbour’s garden has sufficient water, whereas my duty of care and concern requires me to give special weight to my partner’s well-being quite generally. In terms of duration, whereas my promissory duty requires action of me for a specified period (i.e., while my neighbour is on holiday), my duty of care and concern is, at least in most cases, open-ended.

Committed relationships also involve what I will call “relationship-sustaining reasons,” that is, reasons to deliberate and act in ways that will promote the continuation of and health of the relationship. Of course, insofar as we value a relationship, and recognize the specific duties to which the relationship gives rise (e.g., to do our share of the housework, pick up the birthday cake, etc.) we will generally deliberate and act in ways that sustain the relationship. Yet at the same time, we are all too aware that we often face challenges that threaten to derail our relationships. Insofar as a relationship is committed, we have reasons to take actions and cultivate attitudes that guard against this possibility.

Relationship-sustaining reasons might enjoin a wide variety of courses of action and favour various emotional and epistemic attitudes. For example, to sustain their relationship, A and B might have reason to spend time together or to spend time apart. They might have reason to cultivate an interest in a hobby or pursuit that they would otherwise ignore. If A and B are in a sexually exclusive relationship, then they will have a reason not to cultivate feelings of sexual attraction toward a new acquaintance C (Kolodny 2003: 153). Indeed, relationship-sustaining reasons do not only consist in first-order reasons to act or refrain from acting in certain ways, but also in second-order reasons to see certain first-order reasons as irrelevant or less weighty in general or in specific contexts.18 While C’s attractiveness might otherwise give A or B a reason to have sex with C if propositioned by C, it will not do so if A and B are in a sexually exclusive relationship. Alternatively, if A and B have recently had a disagreement, B might have a reason to disregard the reason they have to hold A accountable for another minor wrong for which they are otherwise culpable, so as not to put further pressure on the relationship.

Committed relationships also typically involve two kinds of attitudes on the part of participants. First, being committed to a shared future involves intending that the relationship continue (Nozick 1991: 430). This involves, for example, actively planning and structuring one’s life on the standing assumption that the relationship will persist.19 However, the precise manner in which this intention is manifest will vary depending on the relationship at issue. If A and B are in a long-term romantic relationship it is likely to involve various forms of joint planning and decision-making, ranging from idle speculation to concrete organization about a wide range of short- to long-term goals, priorities, and plans. Furthermore, it might include a mutual understanding between A and B that they have forms of individual and joint authority over some of these decisions (Ebels-Duggan 2008: 162). It might also mean that when A and B individually project themselves into the future, considering the options open to them and the choices they might make, they should do so on assumption that their relationship persists, such that the relationship frames and influences these options and choices.20

A second attitude that participants in a committed relationship generally possess is that of valuing the fact that they are a participant in a committed relationship with this person.21 That is, they value their relationship as a relationship that involves (i) an open-ended duty of care and concern, (ii) relationship-sustaining reasons, and (iii) an active and conscious commitment to a shared future. Valuing the fact that one is a participant in a committed relationship, then, involves a higher-order attitude toward the first-order normative and attitudinal dimensions of committed relationships.22

Marriages offer an example of relationships that at least purport to exhibit the features of a committed relationship. If A and B get married, then part of what they publicly avow is a commitment to a shared future (traditionally, “till death us do part”). Likewise, the fact that marriage implies relationship-sustaining reasons is often manifest in wedding vows. As one source suggests,23 wedding vows may involve A promising B to “[help] our love grow, always [be] there to listen, comfort and support you, whatever our lives may bring.”24

What, if anything, do the features of commitment so far outlined imply about the normative constraints that apply to the power to break up? Strictly speaking, I do not think they imply anything. Jamie and Fred could regard themselves as bound by an open-ended duty of care and concern and relationship-sustaining reasons unless and until one of them decides to end the relationship, at which point those duties simply disappear. Yet whatever the merits of such a relationship, I do not think that it would fully embody the value of commitment. The reason for this, I want to claim, is precisely the presence of the normative option—the normative power—to completely opt-out of the relationship at any time. By contrast, the value of commitment in romantic relationships is partly realized by the very fact that Jamie and Fred lack a normative power to cancel all of the relationship-based duties to which they are subject. The idea is that within committed relationships it would be inconsistent with a source of the relationship’s value if the parties saw the relationship and its associated duties as something they could simply opt-out of or disengage from at will.25 That is to say, if it were always an open possibility for us to completely extricate ourselves from a relationship and its constitutive duties, then our relationship would fail to incorporate the value of commitment, even if neither of us ever takes up (or even considers taking up) this option.26

To make this idea clear it may help to compare another valuable feature of some personal relationships, namely, their egalitarian nature. To say that a marriage or friendship is egalitarian is to say something about the constitutive norms of the relationship and the attitudes of the parties to these norms. Imagine, for instance, that A and B regard themselves as party to an egalitarian marriage. This implies that A and B “accepts that the other person’s equally important interests . . . should play an equally significant role in influencing decisions made within the context of the relationship” (Scheffler 2015: 25). Furthermore, for A and B’s marriage to be egalitarian, it is presumably the case that they have roughly equal power over the character of their relationship and the important decisions that are made within it.27 Thus, if A and B do not regard one another’s equally important interests as playing an equally significant role in influencing the decisions made in the context of the relationship, or the norms that A and B recognize permit A to have the final say about all the major decisions affecting the relationship, then it would be false to describe A and B’s relationship as egalitarian. Consequently, whatever value equality has within marriage would not be realized within A and B’s relationship.

The point I want to draw out by appealing to this case concerns the relation between a relationship’s constitutive norms and the values instantiated by that relationship. Many valuable features of relationships are realized by the normative standards that the parties to the relationship endorse. As I have just suggested, egalitarian relationships require participants to recognize egalitarian constraints on how decisions are made within the relationship. Similarly, I wish to claim, committed relationships involve constitutive norms that are (at least generally) endorsed by the participants in the relationship. Specifically, if A and B stand in a committed relationship, then A and B (i) have an open-ended duty of care and concern regards one another, (ii) have reasons to sustain their relationship, and (iii) lack a normative power to opt-out of all the relationship’s duties at will.28 That a relationship is partly constituted by these norms is what makes it a committed relationship (at least on my usage of “committed relationship”). And the fact that a relationship is committed in this sense endows it with a distinctive value.29

5. Commitment, Voluntariness, and Exit

The argument of the previous section should not be misunderstood. I am not claiming that Jamie is unable to exit her relationship with Fred because their relationship is a committed relationship. Rather, I am making two more moderate claims. First, a common form of romantic relationship involves commitment, and this is something we often take to be a valuable feature of those relationships. Second, the fact that a relationship is committed has normative implications, both while the relationship persists, and concerning the possible dissolution of the relationship. Of particular significance now, the value of commitment can help to explain why someone in Jamie’s position cannot simply cancel all the relationship-based reasons and duties that apply to her and Fred by exercising the normative power to break up.

Nevertheless, we generally assume that someone in Jamie’s position can instigate the dissolution of their relationship. Of course, some traditions and cultures view relationships such as marriage as committed in a stronger way which implies that it is very difficult, if at all possible, for one or both parties to exit. I in no way aim to assess these traditions here. Rather I offer an account of the values that explain the need, recognized by many, to be able to exit even committed relationships.

One obvious reason that we think people should be able to exit relationships is because we value personal autonomy. Ultimately, individuals should be free to choose how to live their life, and this includes the freedom to enter and exit relationships as they see fit. Yet while true, this idea should not be overstated.30 The most significant point for present purposes is that all of the relationships that are the focus of my discussion are, I am assuming, voluntarily formed.31 Thus, so long as it were sufficiently clear to both parties that their relationship was to be committed in the stronger sense that exit from the relationship would not be possible (perhaps when some point has been passed, as marked by an official ceremony such as marriage), then there is no obvious objection from autonomy, insofar as the binding duties of the relationship have been voluntarily entered into.

For this reason, I think it is a mistake to see the ability to exit committed relationships as a straightforward consequence of the value of personal autonomy. Rather, I think our ability to exit relationships reflects the fact that we value many of our relationships in part because they are voluntarily maintained. To be sure, we plausibly value this fact about our relationships out of a more general concern for autonomy. My point is just that it is not the case that we simply think that no autonomous agent can be irrevocably bound in these pervasive ways, and thus grudgingly accept our partner’s capacity to dissolve our relationship.32 Instead, we positively value the fact that our relationships are voluntary, where that implies the possibility of the relationship’s dissolution. For instance, I value the fact that my partner stays with me and continues to pursue a shared future with me in part because she could instigate a breakup. Of course, if she were to exercise this power it would devastate me. But I would not value the relationship as much, nor in the same way, if this were not an option for her, or if I thought she maintained the relationship because she believed herself to be inescapably bound. Thus, the ability to exit a relationship is a condition on the existence of a certain valuable kind of committed relationship.33

To help illustrate this idea, consider the availability of divorce within marriage. Where divorce has not been recognized as possible, some may have thought that introducing the possibility of divorce devalued marriage because it removed the possibility of making the kind of binding commitment that marriage was held to involve. Whether or not this is true, on my view it is at least also true that the possibility of divorce introduced (at least at a formal institutional level) a new positive value into marriages. Individual marriages are, past this point, voluntarily maintained, a fact that the parties to a relationship have reason to value about their relationship.

It might be objected that this account of the value of voluntariness within romantic relationships is inconsistent with a plausible view of the nature of love.34 Specifically, it may seem inconsistent with Harry Frankfurt’s account of the necessity of love. On Frankfurt’s view, love is a “disinterested concern for the well-being or flourishing of a beloved object” (Frankfurt 1999: 167) coupled with a pattern of second-order desires and volitions that centrally involve a commitment to the first-order desires born out of love and a desire that those desires successfully move one to action (Frankfurt 1999: 162). Importantly, according to Frankfurt, love “is not under our direct and immediate voluntary control” (Frankfurt 2004: 44). Indeed, Frankfurt claims that “The lover cannot help being selflessly devoted to his beloved. In this respect, he is not free. . . . The will of the lover is rigorously constrained. Love is not a matter of choice” (Frankfurt 1999: 135).

The idea that who we love is not a matter of choice is familiar. Indeed, many people who break up with someone, or have been broken up with, struggle for long periods with the fact they continue to love their ex-partner. Given this, the claim that we positively value the voluntariness of a romantic relationship may appear inconsistent with a defining feature of romantic relationships, namely, romantic love. In assessing this concern, we should first note that the voluntariness of love and the voluntariness of a romantic relationship are distinct.35 Love is an attitude or set of psychological attitudes towards the object of love. A romantic relationship may be partly constituted by such attitudes, but it also consists in shared activities and relationship-based obligations. Thus, even assuming that our love for a romantic partner is not voluntary—that we simply find our will constrained by our love for them—there is no reason to think we cannot exercise control over whether our normative relationship persists.36 Indeed, someone may exit a relationship precisely to try and stop loving someone, something which is not a direct object of possible choice.37

Nevertheless, within the context of an ongoing romantic relationship, valuing the fact that either my partner or I could exit our relationship may appear inconsistent with, or at least in tension with, the higher-order volitional structure of our loving one another. This structure requires that I am committed to my desire to act in my beloved’s interest in the sense of “being active in seeing to it that the desire is not abandoned or neglected” (Frankfurt 1999: 162). This idea resonates with the account of commitment developed in the previous section. And one might question whether I can be committed to maintaining my desire to act in my beloved’s interest while simultaneously valuing my and my partner’s option to leave the relationship.

Let me make two points in response. First, the fact that we love someone surely does not mean that we are indifferent to the normative structure of the relationship that we have with them. As I suggested in the last section, many people value the fact their romantic relationships are egalitarian, where this requires, among other things, egalitarian constraints on the way in which decisions affecting the relationship are made. Thus, if an egalitarian A finds themselves loving B and establishing a relationship with B, but B does not share A’s egalitarian views or live up to this ideal, this will very likely affect A’s view of B and their relationship. A may seek to convince B of the importance of equality within the relationship, and failing that, decide to leave the relationship. This is perfectly consistent with A’s continuing to love B, at least on Frankfurt’s account of love. True, on Frankfurt’s overall picture it may be that we should describe such a scenario as one in which A’s love for B competes with or is conditioned by other things that A loves or cares about, including egalitarian ideals within romantic relationships (Frankfurt 2004: 41). My point is just that love for a person can be necessary in the way Frankfurt describes—training our attention on persons (or things) that we cannot help caring about, and that mark our “volitional limits” as persons (Frankfurt 1999: 138)—and yet compete with or be conditioned by other concerns about the kind of relationship that one has with one’s beloved.

Second, Frankfurt is explicit that he does not regard romantic love as an authentic or paradigm case of love (Frankfurt 1999: 166). Central among Frankfurt’s reasons for this include the fact that romantic love is generally less disinterested and more conditional than what he regards as paradigm cases of love, such as parental love for a child. Frankfurt remarks that romantic love:

is nearly always mixed up with, if not actually grounded in, a hope to be loved in return or to acquire certain other goods that are distinct from the well-being of the beloved—for instance, companionship, emotional and material security, sexual gratification, prestige, or the like. (Frankfurt 2004: 83)

This should lead us to say one of two things. On the one hand, if we accept Frankfurt’s general account of love, then the widely recognized importance of voluntariness within romantic relationships may be one of the reasons why romantic love is not an authentic example of love. In this case, the tension lies not between the value of voluntariness within a romantic relationship and romantic love, but rather between romantic love and a purer or more authentic examples of love. On the other hand, this limitation in the scope of Frankfurt’s account may be thought to put some pressure on it. Indeed, we might follow Natasha McKeever in arguing that rather than regarding these features of romantic love and relationships as demonstrating that romantic love is not an authentic kind of love, we should view them as part of the distinctive value of romantic love (McKeever 2019: 214, 224). As McKeever shows, one can incorporate significant insights from Frankfurt’s overall account of love into an account of romantic love, and regard them as on an equal footing, so long as one is willing to recognize that not all forms of love are identical.38

Thus, I do not think my account of the value of voluntariness within romantic relationships is inconsistent with a plausible view of romantic love. Nevertheless, someone might suggest that there remains a tension in my view of commitment. After all, the idea that we not only recognize but value the opportunity to exit a relationship may seem inconsistent with the idea that we value the fact that the relationship is committed. However, I think this tension is more apparent than real. For one, a commitment need not be irrevocable or absolute to have value. For instance, a promise to attend my best friend’s wedding can have value even though we both recognize the fact there are conditions (e.g., a family emergency) under which I will not regard myself as duty-bound to attend. Similarly, while I would think it inconsistent with the commitment that I hold my relationship to involve if my partner admitted that she did not recognize reasons to make a special effort to maintain our relationship, or suggested separating at the first sign of trouble, this commitment does not require that my partner and I are irrevocably bound to one another.

What is more, I am proposing that a certain valuable kind of committed relationship would not be possible without the possibility of exit. For instance, the meaning of the relationship is entirely different precisely because the possibility of exit means that the parties to a relationship are constantly reaffirming their commitment to the relationship and its role within their lives (Westlund 2008). Thus, while the possibility of exit may preclude an alternative kind of valuable committed relationship, that does not show that another kind of committed relationship, in which exit remains an option, is not valuable qua committed relationship.

6. Breaking Up

With the foregoing discussion of commitment and voluntariness in place, we are in a better position to consider the normative effects of breaking up. As we have seen, an account of those effects must be responsive to considerations that pull in different directions. On one hand, it must reflect the fact that a breakup fundamentally redraws the normative boundaries that partly constitute a relationship. This is (at least in part) a consequence of the value of voluntariness within romantic relationships. On the other hand, an adequate account of the normative effects of breaking up must also reflect the fact that participants in committed romantic relationships are not able to opt-out of all their relationship-based duties at will.

Before offering my account, however, it is important to note that the precise effects of any breakup will be heavily context sensitive. One reason for this is that, while familiar types of relationships such as romantic partnerships involve common normative features, every token of a relationship type will have unique normative contours. This is the upshot of, for instance, the length and history of a relationship and the way in which the parties have deliberately shaped their relationship. Furthermore, the normative effects of a breakup will be conditioned by facts about the cause of the relationship’s dissolution. For example, if a breakup has been prompted by one of the parties acting wrongfully then the wronged party may have fewer ongoing relationship-based duties than would otherwise be the case.39 Yet with these observations in mind, I think we can identify three normative effects that a breakup will have within paradigm cases of committed relationships.

a)  Cancellation of Reasons and Duties

While breaking up will not result in the cancelation of all relationship-based duties instantaneously, it seems that in most cases a breakup will immediately cancel at least some relationship-based reasons and duties. For instance, if it is soon to be Jamie’s birthday, and Fred would have had a duty to buy her a nice gift, that duty is plausibly cancelled by Jamie’s exercising the power to break up with him. Furthermore, it may weaken many of the reasons that persist. Even after the breakup Jamie and Fred are likely to have stronger relationship-based reasons to help and support one another, and safeguard one another’s interests, as compared with the reasons they would have to serve the interests of a stranger in a relevantly similar situation. Nevertheless, the breakup may decrease the stringency of some of these reasons. For example, while Fred still has a special reason to visit Jamie in hospital and make calls on her behalf if she breaks up with him and is hospitalized shortly after, these reasons may be less stringent than they would have been prior to the breakup.

b)  Cancellation of Relationship-Sustaining Reasons

An instance of a) that is especially noteworthy, a successful breakup will cancel relationship-sustaining reasons. For instance, Jamie’s breaking up with Fred means that they will no longer be required to deliberate about their futures on the assumption that their relationship will persist, or engage in joint decision-making about what their futures will involve. Furthermore, Jamie and Fred will no longer have reasons to refrain from cultivating feelings of romantic love or sexual attraction towards others to help sustain the relationship, because it has been decided that the relationship will not be sustained. This aspect of a breakup is significant because it can be seen as curtailing the self-perpetuation of the normative standards that a relationship involves. Thus, while we cannot straightforwardly extricate ourselves from the normative demands of a relationship, we can undercut the normative pressure toward the reproduction of these normative demands.

I say that a “successful” breakup will cancel relationship-sustaining reasons advisedly. That is because, in at least some cases, it might seem too quick to say that a declaration by A that she is breaking up with B will immediately cancel all relationship-sustaining reasons. As I suggested above, at least part of what it is to participate in a committed relationship involves recognizing a duty to work to resolve difficulties that arise within the relationship. Thus, if there has been no indication of a problem between A and B, and thus no attempt to address whatever problems or issues have prompted A’s declaration, then A may not be able to immediately cancel A and B’s reasons to try and address these issues in order to sustain the relationship. At the very least B may be entitled in such a situation to ask questions and seek explanations, with a view to resolving the underlying problems. And A herself may continue to have reasons to address these problems in light of the valuable relationship she has with B. Whether these reasons can persist in the face of attempts to exit the relationship will be heavily context-sensitive. And in any case, at a certain point, a sincere affirmation by A that she intends to end the relationship will cancel A and B’s reasons to sustain the relationship, and trigger the other normative effects of breaking up.

c)  Duty to Transition

Finally, I want to propose that exercising the power to break up will generate a new duty for both parties, namely, a duty to transition out of the relationship. Before offering support for this claim, it will help to specify some of the actions that the duty to transition might enjoin. Most obviously, perhaps, a duty to transition out of the relationship will require various omissions, such as spending less time together and relying on each other less for help and emotional support. Yet it may also require actively assisting the other in taking these steps. For instance, it might require Jamie to go and stay with family or friends for some period, to allow Fred the time and space he needs to process the breakup. It may also require having conversations about the relationship and the reasons behind its break-down, perhaps with a view to the kind of emotional recalibration that is required when a central relationship in one’s life comes to an end. Furthermore, the duty to transition might require working to alter ingrained habits and patterns of reasoning and emotional response. For instance, Jamie and Fred will need to alter the habit of regularly contacting one another and may need to take active steps to this end, such as talking to a friend or writing a diary. And they will need to work toward a state in which they are not emotionally invested in the life of the other person in the same way or to the same degree.

We can motivate the claim that a breakup generates a duty to transition out of the relationship by highlighting two features of committed relationships and their dissolution. First, given the central and pervasive role that committed relationships play in our lives, the end of these relationships is usually a difficult and fraught process. For example, loving relationships are often held to play a significant role in constituting our identity, to involve a pooling of well-being (such that what is good or bad for you is good or bad for me, and vice versa), a pooling of autonomy (such that partners can make certain creative demands of one another,40 as well as being required to make many decisions collectively), to shape our evaluative and practical outlook, and, for these reasons among others, to make us especially vulnerable to our beloved. No wonder, then, that a breakup can leave individuals distraught and disoriented, uncertain about who they are and what kind of life they should live. For these reasons, a failure on one or both parties to take the steps that will allow them to transition out of the relationship could be seriously detrimental to their well-being and their ability to go on to form future relationships. For instance, if Fred were to accept that Jamie has broken up with him, but neither of them could bring themselves to disentangle their lives, then they are likely to continue in an unhappy and harmful “grey area,” or cycling through an on/off relationship, in which they are unsure where they stand with each other and what their future lives might look like.

These considerations suggest that it will be prudent for the individuals involved in a breakup to take active steps to transition out of the relationship. However, I also argued above that a breakup does not simply cancel the relationship-based duties implied by a committed relationship. Indeed, the parties will continue to have, for at least some period, a mutual duty of care and concern. Given this combination of reasons and the difficulties involved in separating, it seems reasonable to think that there will be normative pressure toward individually and collectively taking the transitional steps that will reconstitute the relationship (as, say, a friendship, or as no longer a special relationship of any kind). Taking these steps will, over time, weaken and dissolve the relationship-based duties that were not cancelled instantaneously by the breakup. Thus, while a breakup will effect some immediate changes in the normative structure of the relationship, it will also open a period of transition within which individuals can work to disentangle their lives, adjust their plans and intentions, and recalibrate their emotional responses and practical deliberations.

I think we can evidence the existence of a duty to transition out of the relationship by noting the plausibility of the claim that if individuals in something like Jamie or Fred’s position were to refuse to take the necessary steps then, at a certain point, they would be wronging the other. For instance, if one of them were to refuse to move out, or continued relying on the other for emotional support, or were frequently to contact their ex-partner’s friends or family, then they would be failing to act in ways that they are (given a reasonable amount of time that has passed) required to act in virtue of the breakup.

Some might have the intuition that any such duties must fall solely on the shoulders of the person who instigates the breakup (in our example, Jamie). Yet while it may often be true that Jamie will bear a more stringent duty to transition, requiring them to do more to facilitate the transition, it is a mistake to think that the individual in Fred’s position will bear no such duty, or indeed that Fred will never bear the more stringent duty to transition. Even if Fred has more leeway than Jamie, there will come a point at which Fred is not entitled to depend on Jamie in the same ways. Recognizing the prerogative of our partner to exit the relationship is something that those who accept something like the model of committed relationships I have been sketching readily endorse. Furthermore, if Jamie’s breaking up with Fred was a consequence of Fred’s repeatedly wronging Jamie, then Fred may bear responsibility for the breakdown of the relationship, and thus owe Jamie more by way of ensuring that she can move on and flourish outside the relationship.

As this last point highlights, the normative effects of a breakup may often be distributed asymmetrically between the parties. This is true not only of the duty to transition but also of the reasons and duties that survive the breakup. Often, for instance, Jamie will have a more robust duty to provide emotional support to Fred after the breakup given that it is Jamie’s decision to break up with Fred. But once again, the precise nature of the continuing reasons, as well as their distribution between the parties, will depend on a variety of interacting considerations.

7. Conclusion: The Value of Breaking Up

I can now summarize my answers to the three questions with which we started. First, I have argued that when A breaks up with B, A exercises a normative power over the relationship-based norms that partly constitute A and B’s relationship. Second, the exercise of this normative power within committed relationships will typically have three normative effects: the cancellation of some (but not all) relationship-based duties, the cancellation of relationship-sustaining reasons, and the generation of a new duty to take steps to transition out of the relationship. Third, the possession (and exercise) of this power is compatible with a valuable form of commitment within romantic relationships. That is because the norms partly constitutive of committed relationships involve a normative pressure towards the maintenance and health of the relationship and constraints on the ability to opt-out of the relationship’s duties (such as the open-ended duty of care and concern) by breaking up. These dimensions of commitment play an important role in shaping and structuring the relationship, but they need not be irrevocable or absolute to have value.

With these answers in place, I want to conclude by offering some brief remarks about the values that justify the power to break up. At several points above, I have suggested that while valuing commitment, many people also positively value the fact that a relationship is voluntarily maintained, and this implies that there must be an ability to exit the relationship. Yet the need to be able to exit a relationship does not imply that there must be a normative power to break up. Jamie could, quite intentionally, start behaving differently in ways that will, over time, weaken and eventually dissolve the normative and emotional bonds implicated in the relationship. For instance, she can start spending less time with Fred and more time with other people, withdraw from joint projects with Fred and invest more time in individual projects, stop communicating love and affection to Fred, and so on. Given this, what values explain the need for someone in Jamie’s position to begin drawing Jamie and Fred’s relationship to a close by exercising a normative power to break up?

I think that both parties to a committed relationship have a significant interest in their being a clear a direct mechanism through which they can manage their normative relationship. That is because committed romantic relationships usually involve deep and pervasive forms of entanglement and emotional vulnerability. For instance, romantic partners often live together, share money, are sexually intimate, have children together, plan their careers around one another, share special relationships with others who they both regard as central to their lives (e.g., children, mutual colleagues, and friends), and serve legal functions for one another (e.g., next of kin). For all these reasons and more, our well-being, autonomy, and practical identities will be shaped by, and come to depend upon, these relationships. Consequently, the dissolution of these relationships will have a significant impact on our lives and is very often a fraught process.

For these reasons, there is great value in our having the power to remould our normative relationship by directly communicating our intention of so doing. Normative powers often play an important role in situations within which it is especially important that A and B have a clear and shared understanding of changes in their normative relationship. For example, A’s power to give or withhold sexual or medical consent to B is important because it is important in these contexts that A and B have a shared understanding of whether B is permitted to act in normally impermissible ways. Similarly, A’s power to promise B to pick him up at the airport is important partly because it is valuable that A and B know whether B can rely on A to pick him up. Given the significant forms of practical and emotional entanglement at stake in committed romantic relationships, the end of those relationships seems like another context in which the ability to directly establish a shared understanding that the normative contours of the relationship are being redrawn is of significant value.


  1. For some important exceptions see, Kolodny (2003: 164ff.), Lopez-Cantero (2018), Lopez-Cantero and Archer (2020) and Saunders (2022).
  2. For a classic statement of this idea, see Scheffler (1997).
  3. Throughout the article, I often refer only to “relationship-based duties” for the sake of brevity. This phrase should generally be taken (as determined by context) to refer to the whole package of relationship-based reasons, duties, and permissions.
  4. For further discussion, see Scheffler (1997) and Brewer (2003).
  5. I assume that in communicating that she “wants” to break up with Fred, Jamie intends to communicate (and Fred understands her as communicating) that she is hereby breaking up with Fred. Thanks to Jonathan Parry for prompting me to clarify.
  6. Thanks to an anonymous reviewer for pushing me to address this possibility.
  7. See, e.g., Bratman (1999a) and Alonso (2009).
  8. See Bratman (1999b), Alonso (2009) and Gilbert (2018: 221).
  9. As I noted in Section 1, I am not claiming that the only way in which relationships can come to an end is through acts of breaking up.
  10. On normative powers generally see, Raz (1999: 98–104; 2022), Owens (2012), Enoch (2014) and Tadros (2020).
  11. On the motivation for this final condition, or something close to it, see Raz (1999: 102; 2022) and Essert (2015: 140–41). Of course, some (e.g., Darwall) might reject the value-based account of normative powers altogether. I cannot defend the value-based account here. However, most of what I argue about the case of breaking up is consistent with an alternative account of the ultimate grounds of normative powers.
  12. See, for example, Raz (1977) and Shiffrin (2008).
  13. Someone might wonder how this claim is consistent with my earlier rejection of the view that in breaking up Jamie communicates an intention. The important difference is that the rejected proposal concerned future-directed intentions with a non-normative object as opposed to reflexive recognition-directed intentions with a normative object. For helpful discussion see, Raz (1977), Enoch (2014) and Manson (2016). Thanks to an anonymous referee for prompting me to clarify this point.
  14. See Roth (2017).
  15. This view seems to be consistent with the view of many contributors on a similar subject at Quora:
  16. Wallace (2012: 190). See also, Kolodny (2003: 163).
  17. See, for instance, Kolodny (2003: 149).
  18. For discussion see Scanlon (1998: 52–54).
  19. This need not be an assumption that the relationship will persist in perpetuity, although in many cases this may be the default position, and in some cases (e.g., a marriage based on traditional vows) such a commitment is made explicit.
  20. This is not meant to rule out, for example, idle daydreaming about other possible futures.
  21. For general discussion about the attitudes involved in valuing see Scheffler (2011) and Theunissen (2020: 88–96). For discussion of valuing a relationship see Scheffler (1997).
  22. Of course, an individual can remain within a committed relationship (and thus subject to its normative demands) despite the occasional and perhaps even prolonged absence of these attitudes. Whether and when the absence of these attitudes signifies that a relationship is no longer a committed relationship is a question I cannot resolve here.
  24. While I cannot explore the possibility in detail, I think that other kinds of special relationships, including friendships and some familial relationships, plausibly involve a form of commitment that shares the central features of the account of commitment I set out here.
  25. Cf. Kolodny (2003: 163) and Wallace (2012: 190).
  26. To be clear, I am not claiming that only committed relationships are non-instrumentally valuable. Rather, I am claiming that committed relationships involve a particular value that non-committed relationships lack. This is consistent with its being the case that non-committed relationships can possess other values that committed relationships necessarily lack. Thanks to Ben Colburn for discussion.
  27. For discussion see Scheffler (2015) and Viehoff (2019).
  28. One might say, using Hohfeldian terminology, that A and B have an immunity regarding themselves and one another concerning certain relationship-based duties. See Wenar (2005: 232).
  29. Again, this is quite compatible with their being other valuable forms of commitment within romantic relationships. In proposing this account, I merely hope to capture what I think is a relatively widespread view of commitment (at least from a certain Anglo-American perspective) that helps to articulate its nature and its possible value.
  30. See, generally, Scheffler (1997: 204–5) and Brewer (2003: 557).
  31. Non-voluntarily formed relationships, such as arranged marriages, raise complications that I cannot address here.
  32. Various examples, such as the obligations that flow from voluntarily conceiving children, show that this is false.
  33. An anonymous referee plausibly suggested that we might also value voluntariness within committed relationships because it helps to ensure that the parties remain on an equal footing within the relationship, thus contributing to a valuable egalitarian relationship. For reasons of space, I leave more detailed consideration of this possibility for another occasion.
  34. Thanks to an anonymous reviewer for pushing me to consider this possibility.
  35. Indeed, on Frankfurt’s view, love does not necessitate nor even necessarily call for a relationship with the beloved (Frankfurt 2006: 41).
  36. Frankfurt might say that our love for our beloved would sometimes stop us from exercising this option. Perhaps, but that need not necessarily be the case, if (to assume Frankfurt’s general framework) we love other people, activities, or ideals more, or if we think our beloved would be better off without us.
  37. See McKeever (2019: 213) and Kolodny (2003: 138).
  38. See also Setiya (2014: 266).
  39. Cases like that of abusive relationships provide clear examples of contexts in which an individual will have no ongoing reasons or duties of the kind I describe below. See Scheffler (1997: 199).
  40. That is, to direct their agency in ways not pre-determined by the norms of the relationship.


For helpful discussions about versions of this article I am grateful to audiences at the University of Glasgow, the Society of Applied Philosophy Annual Conference (Cardiff, 2019), and University College London. For written comments or extended discussion special thanks go to Jessica Begon, Susanne Burri, Ben Colburn, Carl Fox, Ulrike Heuer, Jan Kandiyali, Lucy O’Brien, Tom O’Shea, Jonathan Parry, Angie Pepper, Joe Saunders, Kristin Voigt, Neil Williams, and two anonymous referees for Ergo.


1 Alonso, Facundo (2009). Shared Intention, Reliance, and Interpersonal Obligations. Ethics, 119(3), 444–75.

2 Bratman, Michael (1999a). Shared Intention. In Faces of Intention: Selected Essays on Intention and Agency (109–29). Cambridge University Press.

3 Bratman, Michael (1999b). Shared Intention and Mutual Obligation. In Faces of Intention: Selected Essays on Intention and Agency (130–41). Cambridge University Press.

4 Brewer, Talbot (2003). Two Kinds of Commitment (And Two Kinds of Social Group). Philosophy and Phenomenological Research, 66(3), 554–83.

5 Ebels-Duggan, Kyla (2008). Against Beneficence: A Normative Account of Love. Ethics, 119(1), 142–70.

6 Enoch, David (2014). Authority and Reason-Giving. Philosophy and Phenomenological Research, 89(2), 296–332.

7 Essert, Chris (2015). Legal Powers in Private Law. Legal Theory, 21(3–4), 136–55.

8 Frankfurt, Harry (1999). Necessity, Volition, and Love. Cambridge University Press.

9 Frankfurt, Harry (2004). The Reasons of Love. Princeton University Press.

10 Frankfurt, Harry (2006). Getting it Right. In Debra Satz (Ed.), Taking Ourselves Seriously and Getting it Right (27–52). Stanford University Press.

11 Gilbert, Maragaret (2018). Rights and Demands: A Foundational Inquiry. Oxford University Press.

12 Kolodny, Niko (2003). Love as Valuing a Relationship. The Philosophical Review, 112(2), 135–89.

13 Lopez-Cantero, Pilar (2018). The Break-Up Check: Exploring Romantic Love through Relationship Terminations. Philosophia, 46(3), 689–703.

14 Lopez-Cantero, Pilar and Alfred Archer (2020). Lost Without You: The Value of Falling Out of Love. Ethical Theory and Moral Practice, 23(3–4), 515–29.

15 Manson, Neil (2016). Permissive Consent: A Robust Reason-Changing Account. Philosophical Studies, 173(12), 3317–34.

16 McKeever, Natasha (2019). What Can We Learn about Romantic Love from Harry Frankfurt’s Account of Love. Journal of Ethics and Social Philosophy, 14(3), 204–26.

17 Nozick, Robert (1991). Love’s Bond. In Robert Solomon and Kathleen Higgins (Eds.), The Philosophy of (Erotic) Love (417–32). University Press of Kansas.

18 Owens, David (2012). Shaping the Normative Landscape. Oxford University Press.

19 Raz, Joseph (1977). Promises and Obligations. In Peter Hacker and Joseph Raz (Eds.), Law, Morality, and Society: Essays in Honour of H. L. A. Hart (210–28). Clarendon Press.

20 Raz, Joseph (1999). Practical Reason and Norms (2nd ed.). Oxford University Press.

21 Raz, Joseph (2009). Respect for Law. In The Authority of Law (2nd ed.). Oxford University Press.

22 Raz, Joseph (2022). Normative Powers. In Ulrike Heuer (Ed.), The Roots of Normativity (162–78). Oxford University Press.

23 Roth, Abraham (2017). Interpersonal Obligation in Joint Action. In Marija Jankovic and Kirk Ludwig (Eds.), The Routledge Handbook of Collective Intentionality (45–57). Routledge.

24 Saunders, Joe (2022). Loves Passed. In André Grahle, Natasha McKeever, and Joe Saunders (Eds.), Philosophy of Love in the Past, Present, and Future (131–44). Routledge.

25 Scanlon, T. M. (1998). What We Owe To Each Other. Harvard University Press.

26 Scheffler, Samuel (1997). Relationships and Responsibilities. Philosophy and Public Affairs, 26(3), 189–209.

27 Scheffler, Samuel (2011). Valuing. In R. Jay Wallace, Rahul Kumar, and Samuel Freeman (Eds.), Reasons and Recognition: Essays on the Philosophy of T. M. Scanlon (23–42). Oxford University Press.

28 Scheffler, Samuel (2015). The Practice of Equality. In Carina Fourie, Fabian Schuppert, and Ivo Wallimann-Helmer (Eds.), Social Equality: On What It Means to be Equals (20–44). Oxford University Press.

29 Setiya, Kieran (2014). Love and the Value of a Life. The Philosophical Review, 123(3), 251–80.

30 Shiffrin, Seana (2008). Promising, Intimate Relationships, and Conventionalism. Philosophical Review, 117(4), 481–524.

31 Tadros, Victor (2020). Appropriate Normative Powers. Proceedings of the Aristotelian Society Supplementary Volume, 94(1), 301–26.

32 Theunissen, L. N. (2020). The Value of Humanity. Oxford University Press.

33 Viehoff, Daniel (2019). Power and Equality. In David Sobel, Peter Vallentyne, and Steven Wall (Eds.), Oxford Studies in Political Philosophy (Vol 5., 3–38). Oxford University Press.

34 Wallace, R. J. (2012). Duties of Love. Proceedings of the Aristotelian Society Supplementary Volume, 86(1), 175–98.

35 Wenar, Leif (2005). The Nature of Rights. Philosophy and Public Affairs, 33(3), 223–52.

36 Westlund, Andrea (2008). The Reunion of Marriage. The Monist, 91(3–4), 558–77.