Commitments: An Overview

There are (at least) three general types of commitments:

(1) individual commitments to ourselves (e.g., my commitment to go running each morning)

(2) individual, interpersonal commitments (e.g., my commitment to call my best friend on a weekly basis)⁶

(3) fully shared commitments, where the commitments in question are those that we take on together (e.g., the commitment between two romantic partners)

The relevant question to consider in what follows is which type(s) of commitments play a role in the context ofjoint action and shared agency more broadly.

To do so, however, we first need to determine what a “commitment” is. Although the con­cept of a commitment figures prominently in a wide variety of philosophical debates, there has been comparatively little direct philosophical inquiry about them. There are exceptions to this generalization (e.g., Arruda ms; Bratman 2004; 2014; Brewer 2003; Chang 2009; 2013; Chartier 2017; Dorsey 2016; Elster 1985; 2000; Gilbert 2013; 2014; 2018; Helm 2017; Hieronymi 2006; 2009; Hinchman 2010; Holton 2009; Killmister 2017; Liberman and Schroeder 2016; Marusic 2015; Michael and Pacherie 2015; Ross 2012; Roth 2004; Schroeder 2013; Sen 1977; 2005; Shpall 2014; Townsend 2017).Yet even in cases where commitments are the focus, the contexts within which they are used also vary widely, from theories of intention, resolution, autonomy, and rationality, among others.

Given that this variety makes it difficult to identify explicit competing views about commitments, I suggest that it is most fruitful to begin from a guiding case. Doing so will help to determine the prospects of using commitments (and potential implicit views about them) to understand shared agency and collective responsibility.⁷

SMARTPHONES: Imagine two close friends, Angela and Jane. They meet monthly for dinner, given that their respective jobs make it difficult to make last-minute plans to see one another.

Although this is an example of an individual, interpersonal commitment rather than a shared or joint commitment, it provides a useful model for answering the following questions: ^{1 2 * *}

13.1.1 The Agency Question

SMARTPHONES points to three characteristics of individual, interpersonal commitments. These characteristics will hold, mutatis mutandis, for commitments present in joint action, and they underscore why philosophers think commitments are important for understanding shared agency more broadly.⁸

First, under normal circumstances, commitments are attitudes that represent what we self- reflectively aim to do in light of the fact that we take this aim to have some import or value. In SMARTPHONES, Angela makes the commitment to Jane to limit her smartphone use during their monthly dinners in order to capture both that she has self-reflectively adopted this aim and that she takes this aim to have volitional significance for her future practical deliberation. This may seem like an obvious point, but there is not universal consensus on this matter. Shpall (2014: 148, 153), for example, denies that what he takes to be the philosophically compelling kinds of commitments have a “volitional” component; rather, the types of commitments that matter are those that establish normative relations between a mental state or action and its object (another mental state or action).⁹ Sen (1977; 2005) denies that commitments are attitudes, and instead takes actions to be committed when they are undertaken for the pursuit of others’ goals rather than for the sake of our own welfare.

Yet for commitments to play a substantive role in shared agency, they should have what Chang (2013) and Shpall (notwithstanding his claim that this feature is not present in the philosophically interesting cases of commitments) characterize as a “volitional” component. That is, they should be the kinds of attitudes or dispositions that figure prominently in our relationship with our own actions or the actions we undertake with others. At the same time, as I discuss in the next section, their role in shared agency is not fully explained by this characteristic.

SMARTPHONES suggests, second, that commitments are future-directed: (1) minimally, they represent what we aim to do rather than what we have done; and (2), in so doing, they add to the strength and the stability of our intentions, thereby putting rational pressure on us to follow through on them.¹⁰ Here again, there is little consensus in the literature about whether commitments are future-directed in this second, more robust sense (Bratman 2004; 2014; 2018; Hinchman 2014; Holton 2009; Marusic 2015; Morton 2013; Morton and Paul, 2019).¹¹ Nonetheless, this feature of commitments is significant for shared agency. One puzzle for theories of shared agency is to explain how and why we should count on one another to follow through on our plans to act together. One way to solve this puzzle is to point to the kind of stability that either intentions or, as I suggest above, commitments provide. As will become apparent in §13.2, both Gilbert’s and Bratman’s respective uses of commitments will depend on the idea that they provide stability of this more robust kind.

SMARTPHONES highlights, third, that commitments have normative force. They estab­lish backward-looking reasons, either other things being equal or all things considered, to follow through on the commitment.

Worries about bootstrapping naturally arise here (Bratman 1987; Chang 2013; Ferrero 2006: 103; 2010; Holton 2009; Marusic 2015).¹² In the individual case, the problem of bootstrapping arises because the fact that one made a commitment to φ is potentially the wrong kind of reason to justify φ-ing.¹3 The fact that someone makes a commitment to, say, be a jerk—or, to invoke an example from Rawls (1971: 432), to count blades of grass—does not provide an independent reason to do so. In fact, one might think that there are overriding reasons not to be a jerk even when one has made a commitment to do so. In the grass-counting case, there are no overriding reasons not to fulfill one’s commitment to grass-counting, but one might wonder whether the fact that one has made a commitment to count blades of grass is sufficient reason to do so. Granting that commitments provide sufficient reasons in either instance is to suggest that we can bootstrap into existence reasons (that do not hold otherwise) simply by what amounts to an act of will. Even if one is willing to grant that we can do this in some circumstances, as Chang (2013) argues in the case of loving relationships, there is a genuine question about whether commitments can generally do so without falling prey to a bootstrapping problem.

Yet it is unclear whether these same worries arise in the case of shared agency. The backward­looking reasons that commitments plausibly involve in the context of shared activities are not reasons to φ; rather, they are reasons to uphold the commitment that one has made to (and with) others, where the end goal of doing so is that they φ together.¹⁴ If this is correct, then, interestingly enough, the problem of bootstrapping is only (or largely) a problem for the case of individual (whether inter- or intra-personal) commitments.¹⁵

Another aspect of commitments’ normative force worth emphasizing—particularly for their import in debates about shared agency—is that their normativity is not necessarily moral in nature.

13.1.2 The Responsibility Question

SMARTPHONES suggests a number of important ways in which commitments are relevant for understanding (moral)¹⁶ responsibility. As will become apparent, however, there are important disanalogies between their role in the individual and collective contexts.

First, SMARTPHONES shows that Angela has forward-looking responsibility¹⁷ to follow through on her commitment to Jane. As Gilbert (2014: 58) notes, we are responsible in a backward-looking sense when we have caused some state of affairs for which responsibility judgments are fitting. By contrast, we are responsible in the forward-looking sense when we have obligations to fulfill.¹⁸ SMARTPHONES shows that commitments, much like promises, set up obligations to the commitment’s addressee (and the addressee’s expectation that the commitment will be fulfilled) and thus are the grounds for forward-looking responsibility. The source of the obligation is twofold: (1) Angela has voluntarily made the commitment such that she has made herself answerable to Jane; and (2) Jane has a claim on Angela to fulfill the commitment given that Jane is the addressee of the commitment.

There are not, however, analogues to (1) and (2) in the collective case. Regarding (1), for at least some views of the role of commitments in shared agency such as Gilbert’s, the obligations associated with commitments may hold simply in virtue of one’s membership in a group or one’s participation in a joint activity rather than solely in terms of whether one explicitly makes a commitment. There is also no clear analogue to (2) in the collective case. Although SMARTPHONES involves an interpersonal commitment, it is not mutual. Angela has made a commitment to Jane (and thus has an obligation to her to follow through on it), whereas Jane does not have a comparable commitment (and concomitant obligation) to Angela. By contrast, if commitments are relevant for understanding collective responsibility, the paradigmatic type of commitments will be mutual (or even fully shared) rather than interpersonal.

Notwithstanding this disanalogy, let’s return to SMARTPHONES to consider what else it tells us about the relationship between commitments and responsibility. SMARTPHONES highlights the conditions under which Angela can be released from her commitment. Since the commitment is interpersonal, one might be tempted to think that only Jane can release Angela. There are, however, two reasons to doubt that Jane and only Jane can release Angela from her commitment: (1) this would require categorizing commitments as a special form of promises; and (2) there is an intrapersonal dimension to commitments that captures Angela’s initial will­ingness to take on the commitment and what it requires. Regarding (1), commitments may share some features with promises. Yet we use them in different contexts and thus we should be able to explain them in ways that reflect these differences. The obligations commitments create must, in at least some cases, be jointly rescinded, whereas promises are typically rescinded only by the promisee. What about (2)? Even ifJane releases Angela from the latter’s commitment to avoid using her phone during dinner, Angela may not want to give up her commitment and may still judge that she has reasons to uphold it.

As was the case above, this feature of individual, interpersonal commitments again stands in contrast to mutual (or even fully shared) commitments in the context of shared agency. As I dis­cuss in §13.2, these commitments cannot be rescinded by one participant; rather, they must be rescinded either by all participants individually or, even, by all participants as a group. Whether there is an intrapersonal dimension to these commitments will vary among the different views of shared agency in which commitments play a role.

Thus far, I have used SMARTPHONES to understand the relationship between commitments and forward-looking responsibility (in both the individual and collective senses). Yet the more familiar kind of responsibility is backward-looking. SMARTPHONES shows that there is an important disanalogy between commitments and individual backward-looking responsibility, on the one hand, and commitments and collective backward-looking respon­sibility on the other. In the individual case, we will likely need a robust theory of (moral) responsibility to explain how commitments are connected in any interesting and inform­ative way to states of affairs for which backward-looking responsibility attributions are appro­priate.¹⁹ In the collective case, by contrast, they often secure important aspects of shared agency, as I briefly outlined in the previous section and as I show in more detail in §13.2. In this regard, commitments will turn out to be directly relevant for the question of collective backward­looking responsibility even if they do not fully settle the question of whether and when it is appropriate to assign it.

Finally, there is one issue that is unique to the question of collective responsibility and for which commitments will be useful, but for which there is no analogue in the individual case. This is the question, to which I return in §13.3, of whether collective responsibility is distribu­tive or non-distributive.

13.2