2. Meaning in Language and Action

This section lays the groundwork for what is to come by first highlighting two points about the linguistic communication of meaning—that is, our attempts to convey our thoughts through words. First, whether we’re understood often matters to us. Second, our ability to communicate meaning via any particular locution is dependent on various social conditions; context determines what we can communicate. Highlighting how these points play out in a more familiar context will make it easier to see how the same factors are at play in expressive action. Footnote ²

We take it as uncontroversial that being able to make ourselves understood matters. Suppose, for instance, that you tell a friend, “I’m fine,” intending to downplay a minor inconvenience, but your friend misinterprets this as a sign of deeper distress. This reaction might be frustrating. There are higher-stakes cases as well. Some feminist scholars argue that in patriarchal societies, norms, expectations, and power dynamics make it difficult for women to express dissent through the locution “no.” In Rae Langton’s canonical articulation of this idea, “Some speech acts are unspeakable for women in some contexts: although the appropriate words can be uttered, those utterances fail to count as the actions they were intended to be” (Reference Langton1993: 299). This feminist critique highlights a key point for our argument: because much can be at stake in being understood, causing someone to be misunderstood has moral risks. There is a general presumption in favor of preserving the conditions for others’ communicative integrity.

The second point we aim to highlight in this section is that what one is able to communicate depends on various social factors; communicative possibilities are shaped by the context in which they occur. Two aspects of the relationship between social context and communicative efficacy are particularly important for our argument. First, communication depends on reciprocity. This point, combined with the previous—that it matters to us that we can communicate effectively—generates communicative obligations on each of us, which opens up space for specifically communicative wrongs. Investigating such wrongs has been an important project in social epistemology (Fricker Reference Fricker2007; Dotson Reference Dotson2011).

Second, what one can express through language depends on what is taken as presupposed by the relevant discursive community. David Miller illustrates this point with the historical transformation of the term “condescension.” In an era when deference to social superiors was widely accepted, a high-status individual who “condescended” to speak as an equal with an inferior was seen as virtuous—it was “amiable to forgo deference that you can rightfully claim” (Reference Miller1997: 233). Today, however, “condescension” carries a strongly negative connotation. Since contemporary social norms reject the legitimacy of inherited social superiority, to describe someone as condescending is to condemn them for wrongly assuming a status they do not rightfully possess. The very possibility of using “condescending” as a term of condemnation, then, depends on the broader social context—specifically, on the extent to which an egalitarian ethos permeates society.

In this article, we explore some implications these points have for the way we express ourselves through our actions, rather than our words. As a first step, notice that some kinds of actions—things like waving hello or giving a thumbs up—work more or less the same way speech acts do. Both of our points about linguistic meaning apply here. It can hurt to be misunderstood, e.g., when your wave reads as awkward rather than enthusiastic. And what you can do with these actions depends on the local context. In a community where waving hello isn’t a widely adopted practice, you can’t expect your wave to be interpreted as a greeting.

The set of actions that work in exactly this linguistic-like way is fairly narrow, but the set of expressive actions simpliciter is significantly wider. Many actions serve a communicative function, even when they don’t do so in the same way that linguistic expressions do. Suppose that you have been invited over to an acquaintance’s home for dinner. You’d like to express your gratitude for the meal and for the opportunity to begin to move from acquaintances to friends. One way to express your gratitude would be to say something, perhaps “I appreciate the opportunity to spend time together.” Another way would be to do something, perhaps bringing a gift. The latter option works by being an observable consequence of the attitudes you intend to express. In the parlance of creative writing, the gift shows rather than tells.

Both of our observations about linguistic meaning apply to these kinds of actions as well. Just as with linguistic expression, there are emotional stakes. It’d hurt, for instance, if your acquaintance interpreted your gift as an ostentatious display of wealth rather than an act of gratitude. Similarly, to invoke the ideal of communicative reciprocity, if your host is relentlessly cynical, you may struggle to express your intended meaning no matter what action you take. And, just as with linguistic expression, the context matters. In a community where bringing a gift is the default expectation whenever you visit someone’s home, bringing one isn’t evidence of gratitude (or at least, not reliable evidence) (Raz Reference Raz1988: 349). To make your attitude observable in that kind of case, you’d have to do something else.

We can be more precise about how these “show rather than tell” actions work by introducing distinctions from metaethics and philosophy of action. When discussing reasons for action, it’s common to distinguish at least three categories: “justifying reasons” that rationally count in favor of acting, “motivating reasons” that move an agent to act, and other kinds of explanatory reasons which do not make reference to one’s deliberative processes. Markovits (Reference Markovits2014: 1) illustrates the difference like this. Suppose that Alessandra snaps at Beth. From a third-person point of view, we might explain this by noting that Alessandra hasn’t had enough sleep lately. That might on some level be the correct explanation, but it’s not how things seem to Alessandra first-personally; that’s not what moved her to snap at Beth. In other words, it was not Alessandra’s motivating reason. Suppose further that what moved Alessandra was Beth’s being late for their meeting. Even if so, lateness might not actually count in favor of Alessandra’s snapping. This case illustrates that motivating reasons, justifying reasons, and other kinds of psychological explanation can come apart, but of course, they don’t have to. When things go well for us as agents, we’re motivated to act by the same considerations that justify acting.

T.M. Scanlon (Reference Scanlon2010) uses this distinction among senses of reason to pick out what he calls the “meaning” of an agent’s action. Scanlon notes that there are cases in which someone acts permissibly yet still seems criticizable. For example, bringing a gift for your host is morally permissible, but if you bring the gift to flaunt your wealth, that nonetheless indicates something morally criticizable about you. By distinguishing questions about permissibility and blameworthiness on the one hand from questions about an action’s meaning on the other, he hopes to defend the coherence of saying that there are actions that are subject to moral criticism on the basis of their meaning but are nonetheless not morally blameworthy.Footnote ³

On Scanlon’s view, the meaning of an agent’s action is determined by the motivating reasons that led that agent to do it. For example, if you invite someone to a party and are motivated by a desire to make friends, your action says something different than had you extended the same invitation in hopes that your guest would bring some expensive champagne. Of course, we often act from more than one motivating reason, and Scanlon’s account of meaning accommodates this complexity. Acting on the basis of two different motivating reasons can mean something different than acting on either reason alone. To borrow his example, suppose that you have promised to meet a friend for lunch and that you enjoy seeing this friend. You take each reason as sufficient to show up at the restaurant, but acting on both is important here. If you were to show up only because you want to, your action would manifest a lack of respect for your friend’s time. If, instead, you were to show up only because you promised, your action would manifest a lack of interest in your friendship.Footnote ⁴

This kind of meaning often matters to us. Your dinner host cares whether you bring a gift to express gratitude or to flaunt your wealth. Notably, we also care about the Scanlonian meaning of our own actions. If you feel gratitude, you might want an opportunity to express it in action. That we care about the meaning of our own actions is an unsurprising upshot of the points we made above about linguistic meaning. Just as being understood matters to us in the case of linguistic meaning, it matters to us when we’re expressing ourselves through Scanlonian meaning as well.

We think Scanlon’s account is a useful model for thinking about how “show rather than tell” actions work, and what distinguishes them from actions that express something in a way that’s more closely analogous to linguistic meaning. So, going forward, we’ll use the idea of Scanlonian meaning as a way of picking out an action’s capacity to express something about the agent that performs it by making manifest her motivating reasons. In relying on Scanlon in this way, though, we don’t mean to suggest that his view necessarily captures all of the ways that an action might have a communicative meaning. On that point, we’re ecumenical. What we’ll go on to say applies no matter what range of ways you think an action can have a meaning, so long as the way Scanlon identifies is one of them.

3. Deliberative Autonomy

This article is about a particular threat to deliberative autonomy that concerns expressive actions. In the previous section, we introduced the type of expressive actions we’re interested in; in this section, we’ll say more about deliberative autonomy.

David Enoch (Reference Enoch2022) has helpfully elucidated a two-part view of autonomy, wherein autonomy is instantiated in the dual properties of non-alienation and sovereignty.Footnote ⁵ One accomplishes non-alienation when one’s life is shaped by one’s deepest commitments and values. One has sovereignty when one’s considered judgment settles what will happen. Both components are ineliminable, irreducible, and essential elements of a truly autonomous life. Think first of one who possesses sovereignty, but not non-alienation. We might imagine a case where one is stuck at the bottom of a pit with no means of escape and just enough food and water to survive. There is no question who is calling the shots here. The problem for autonomy is that one can make no choice which will cause one’s life to reflect one’s values and commitments. One can make no meaningful relationships, produce no great music, and pursue no novel experiences. One reigns supreme over a kingdom of dirt. Now think of one whose life exhibits non-alienation but lacks sovereignty. Their life is filled with just the kind of achievements, relationships, and pursuits required to give reality to their values, but suppose this only comes about because they are subject to a paternalistic state which determines what they can do, with whom, and when. Such a life, as experientially rich as it may be, is not truly autonomous.

Following Enoch’s later development of his account, we might complicate the above story.Footnote ⁶ These two components of autonomy, though independent, are connected in a crucial sense. The truly autonomous life is one whose shape is determined by one’s values because one’s choices made it so. This understanding of autonomy lends itself to regarding autonomy as an achievement. Those who succeed in living an autonomous life are admirable in the way that, to borrow a metaphor, a skilled archer hitting a bullseye is admirable. The archer both hits the bullseye and is skilled, but their feat is admirable precisely because their skill caused the arrow to hit the bullseye. The autonomous person both has a life shaped by their values and is the final arbiter of what they do, but their life is especially valuable due to the connection between those two things—their life is a reflection of their values because they are sovereign over their choices.

With this picture of autonomy simpliciter in mind, we can explicate the notion of “deliberative autonomy” at the center of this article. Deliberative autonomy is a matter of one having the deliberative space required to live an autonomous life as just described. The distinctive importance of deliberative space becomes clearer when we notice that a choice or decision is an essentially deliberative activity. An action will not be of the type which matters to the value of autonomy if it is not the result of one’s own deliberative process. That is what the case of the paternalistic government above shows. It follows, then, that one cannot shape their life through their choices (in the required way) if they do not also maintain a level of control over their deliberative process itself, the very mechanism by which choices are made. Failures of such control are central to the literature on free will and responsibility. Think of a case where Tina is brainwashed by mad scientists to deliberate in an excessively egoistic fashion (Pereboom Reference Pereboom2014). Most believe that if Tina acts wrongly in such a case, the brainwashing is exculpatory just because she did not have sufficient control over her deliberation to count as autonomous. Taking all this together, we get a picture of autonomy which builds on Enoch’s and looks as follows: the autonomous life reflects particular values and commitments because of the choices made by the one living it, and the deliberative mechanisms by which one made said choices were sufficiently independent of interference. Without the security of this final clause, there is a perpetual worry that one’s choices are not one’s own choices at all, but merely a reflection of another’s interference into one’s deliberation.

How does this connect to what we said about Scanlonian expressions in the previous section? The intuitive thought is that a primary way in which we shape our lives is through the meaning of our actions. The meaning of our actions, taken together and spread out over time, in large part determines the range of coherent narratives we can tell ourselves about our lives. They make a difference to what kind of person we can conceive of ourselves as being and what kind of person others can reasonably believe us to be. Given this, we’ll want some control over the expressive opportunities available to us.

There are at least three ways in which having such control might be constitutive of an autonomous life, each way corresponding to one of the elements detailed above (non-alienation, sovereignty, and their causal connection). First, expressive control can implicate non-alienation. This is easy to see if you imagine an extreme case where you are simply unable to express intended meanings through your actions. Suppose you are a visitor in a faraway land, oblivious to local customs and cultural norms. This is the social equivalent of life at the bottom of a hole. Second, sometimes such control is a requirement for sovereignty. Though meanings are social and cannot be solipsistically determined, there is still a sense in which an autonomous life depends on your ability to settle the matter regarding what you express through your actions. This is part of what is so valuable about a socially recognized lexicon of meanings. Once all the relevant persons are on the same page, so to speak, it is sufficiently within your control in normal conditions to determine what you express through your behaviors. Lastly, this kind of control affects the causal connection between non-alienation and sovereignty. Remember, the fully autonomous life is the life reflective of one’s deep commitments and values because one was the final authority regarding the decisions one made. But plausibly, the sense of “because” used here must involve a deeper kind of connection than mere causality. To see this, think again of the case of the archer. Suppose the archer is set to miss their shot, however, right before they release a fan says, “You’re the best!” This jars the archer, and this slight jerk causes them to release at a different angle and hit the bullseye. We might give a causal explanation of this case by saying that the archer hit the bullseye because they are skilled, for if they weren’t skilled the fan would not have yelled at them. But clearly this is not the kind of connection that renders the achievement especially valuable. The skill doesn’t play the right causal role in the result. Similarly for autonomy, it is not good enough that we make decisions which, by some causal mechanism or other, achieve non-alienation. Rather, our decisions have to play the right role; the life we build has to, in some sense, be the one we were actually attempting to build. This matters because it shows the importance of expressive opportunities and control to deliberative autonomy. If one cannot control the meaning of the actions which make up their life, then however their life ends up looking, their choices will not share the correct causal connection to what they end up expressing.

If something like this is right, then our interest in having autonomy implies an interest in having a sufficiently good range of opportunities for expressive action. But what kind of range, and what’s sufficiently good? We cannot offer a full answer here. Nonetheless, we can offer a sense of the type of cases where our interest in opportunities for expressive action might be particularly acute.

At a high level of description, our idea is that performing actions with a particular Scanlonian meaning can contribute to making those actions meaningful. An action is meaningful, as we understand it here, when it enjoys a distinctive significance in virtue of its connection to long-term projects, identity-constituting properties, important relationships, and so on. For example, your delivery of a home cooked meal to a sick friend might take on special significance for them as an expression of your care. It can also happen that your action takes on this significance for you, the person doing it. You might relish the opportunity to make your grandmother’s particular spin on Jewish deli-style chicken soup—the one she always made when someone in the family was sick—seeing making it on this occasion as a way of honoring that part of your heritage. This parallels the way an object can be meaningful, as when a gift from a friend takes on special significance for you as a representation of that friendship.

From both directions—an action’s meaningfulness for others and for the performing agent—the meaningfulness of the action depends in partFootnote ⁷ on its Scanlonian meaning. What makes the meal delivery meaningful is that it’s an expression of care, and what makes it an expression of care is that care motivated you to do it. Had your friend ordered delivery, the arrival of the delivery person wouldn’t be an expression of care, and so wouldn’t be the kind of thing that could come to represent being cared for.

To sum up: we have argued that being understood matters to us, that some actions express a distinctive kind of meaning by illustrating our motivating reasons, and that sometimes that form of expression can make an action particularly meaningful. With those points in mind, it’s time to return to the other uncontroversial point about linguistic meaning that we started with: what we can express depends on the context. This point, we’ll argue, has an underappreciated upshot when applied to Scanlonian meaning, one that reveals a distinctive threat to deliberative autonomy.

4. A New Threat to Deliberative Autonomy

Our capacity to express ourselves depends in part on what other people will take our expressions to mean. So, if you are in a position to change the context of someone’s expression, you can prevent her from being understood. In the case of linguistic expression, there is relatively little that any of us could do individually and in a short amount of time to change the context which determines how a locution is understood. However, in the case of Scanlonian meaning, the relevant context can be much more under individuals’ control. We explore that difference in this section.

Ultimately, we think the phenomenon we’re concerned with here is familiar, but to focus attention it’ll help to start with a cartoonishly outlandish case. Consider:

Obviously, there’s a lot in this scenario that is bad for Sasha. Most obviously, her parents are deliberately undermining her. What’s important for our purposes, though, is the way that undermining works. By assigning Sasha to do what she would have done anyway, Sasha’s parents ensure that when Sasha helps out, she is never doing so solely out of generosity. Of course, Sasha can still be generous, but she is robbed of an opportunity to manifest her generosity in her actions, or at least to manifest it in the same way she could if being helpful were her sole motivating reason. So, if one were to tally up the things that are bad for Sasha in this scenario, one would not stop at her parents’ underhanded scheming. In addition, one would include her loss of an opportunity to do something with a particular meaning that she cares about being able to express, something which is meaningful to her.

This particular bad-making feature of the situation is reproduced in more realistic cases. Now consider:

Much about this situation is better for Sasha than in Bad Parents. But even though her parents are well-meaning and could not have known that she’d resolved to do her brother this favor, their assignment of dish duty nonetheless puts Sasha in a position where doing the dishes no longer has the same meaning. Even if she knows her parents did so accidentally, Sasha might reasonably regret not having the opportunity to do the dishes solely out of generosity (or not being able to successfully express generosity in her dish-doing). Of course, to point out that there is this bad-making feature of her parents’ actions in Dish Duty is not to say that Sasha’s parents wrong her all-things-considered. Nonetheless, it’s interesting and important that ordinary actions can have this cost.

We think that what happens in both Bad Parents and Dish Duty is that Sasha suffers a loss to her deliberative autonomy. Her parents change her context such that they remove an opportunity for her to express herself in a particular way. Specifically, we think the loss works like this. Sasha’s parents’ chore assignment gives her a new justifying reason to do the dishes: something like “this is my chore” or “my parents told me to do this” is now part of what counts in favor of doing the dishes. Now that this new justifying reason is on the scene, Sasha’s possible set of motivating reasons changes as well. She might act only on the reasons she originally took to justify doing the dishes, only on her reasons to do her chores, or on a combination of the two. But given this change in the available motivating reasons, her action cannot have the same Scanlonian meaning it did before. If Sasha acts only on the reasons she had previously taken to justify doing the dishes, she is now conspicuously not acting on her reasons to do her chores. So, what would have been an action that manifests generosity to her brother now manifests indifference to her parents’ authority. This has the very same structure as the case discussed earlier, where attending lunch with your friend only because you want to manifests indifference to your promissory obligations. So, it’s no longer possible for Sasha’s action to have the meaning it would have had prior to getting assigned chore duty. To return to our example of bringing a gift when you’re invited to dinner, it’s as if on the way to your acquaintance’s home, the local customs change so that gift-giving is now the default.

These cases illustrate that our ability to communicate Scanlonian meaning is vulnerable to a type of interference that other kinds of communication are not. Sasha can still communicate in words and she can still communicate with actions that have a conventionally recognized, linguistic-like meaning (she might give her brother a hug). These kinds of communication don’t run through her motivating reasons. In contrast, Scanlonian meaning does run through our motivating reasons, and our possible set of motivating reasons is affected by what justifying reasons there are, so our ability to communicate with Scanlonian meaning is vulnerable to others’ ability to change the circumstances of our decision in ways that change the justifying reasons we’re confronted with.

There is a potential worry about our explanation of what’s going on in Dish Duty that needs to be addressed before moving forward. Why can’t Sasha just act from her previous motivations? It may seem possible that, despite her new justifying reason, she could still act from the very same motivations. This may be, but it doesn’t follow that Sasha’s action can have the same meaning as before. Recall Scanlon’s lunch with a friend: the meaning of one’s actions is not determined only by what motivates them, but also by what doesn’t motivate them. Before her parents assigned her dishes, failing to act from a sense of duty would have no expressive significance to Sasha, as she was not duty-bound to do dishes. However, now, if she is to avoid manifesting rejection of her parents’ authority, she must act from duty.

One might grant this but note that Sasha may still be able to act from duty and generosity. Wouldn’t that have a close enough meaning to acting from generosity alone? We offer three responses. First, even if Sasha can maintain the self-facing expressive function of her action (she still knows she was being generous), she now has a more difficult time preserving the intended other-facing function. After all, absent additional explanation, Sasha’s brother and parents will assume she is doing the dishes because she was told to. She might overcome this by telling her family, “Just so you know, I was going to do the dishes anyway!” but this risks coming across as objectionable moral score-keeping. At the very least, the change in Sasha’s context has created an interpretive burden.

Second, we sometimes care about the “purity” of intentions. For an extreme case, suppose your partner tells you “I married you for two reasons. First, I love you. And second, right before I met you, I flipped a coin and told myself I’d get married to the next person I see if it comes up heads, which it did!” In this case, your partner’s decision to marry you was overdetermined, but you might wish that your partner married you only for love, or, at least, that had they not loved you they wouldn’t have married you. But you lose this counterfactual dependence in both the marriage and Sasha cases of mixed motivation.

Lastly, we shouldn’t take for granted that Sasha can still act out of generosity after her chore assignment. Generosity still counts in favor of doing dishes, but it might not motivate Sasha in the same way. What motivates us is not voluntaristically chosen (Audi Reference Audi2021; Ross Reference Ross1930).Footnote ⁸ Nor are one’s motivations set by a post-hoc explanation of what might have driven one to act. Instead, an action’s meaning is determined by the considerations which, in fact, moved the agent to action. It appears possible that the injunction from Sasha’s parents may, as a matter of pure psychological contingency, motivate her in her dishwashing. Insofar as this new motivation swamps or boxes out her previous motivation, then the meaning of her action is affected. It’s easier to see how this might happen if one ratchets up the stakes. Suppose that the context of Sasha’s action was changed not because of chore duty but because a burglar broke into her home and told her, “If you do not wash the dishes immediately, I will kill you and your family.” Sure, it is possible for Sasha to act from multiple motivations; she might act so as to save her family’s lives and to express generosity towards her brother. However, we contend that such a motivational profile is unlikely. More likely, Sasha will simply find herself compelled to wash the dishes because of the threat. It is obvious that Dish Duty is far less dramatic than the example just given. However, if one concedes that great stakes might significantly shift motivations, then it seems at least plausible to think smaller stakes might shift motivations to some degree. And that is all that is required for affecting an action’s meaning.

None of this logically entails that Sasha is any worse off because of her parents’ demand. She might successfully explain her prior motivations without coming across as petty. Perhaps her case is not one where purity of motivation matters, either to her or her family. And, lastly, maybe she is psychologically constituted such that her new reasons do not overshadow her desire to act from generosity. All of this is possible. But the opposite is possible too. And so, if one accepts that expressive meaning is something we should care about, then this gives us a reason to be careful when changing the context of another’s expressive actions.

So far in this section, we’ve identified an under-discussed threat to deliberative autonomy. One might reasonably wonder, though, whether it’s a big threat. After all, it might not be a big deal to Sasha whether she can express generosity on this particular occasion, or whether she does so by way of Scanlonian meaning or some other means. Whether this loss really matters depends on how we fill in the details of the cases, and we chose to leave them open for the sake of focusing attention on the mechanism by which the loss takes place. Nonetheless, we think that this threat to deliberative autonomy is sometimes quite significant. That’s because, as we argued earlier, sometimes our ability to express a particular Scanlonian meaning is what makes an action meaningful.

To help articulate why losing the opportunity to express certain parts of ourselves in action might be important, we turn to Mill. While propounding the value of the free exchange of ideas, Mill notices the risk that our beliefs can get isolated from our day-to-day reasoning, thereby diminishing something about us as agents. In his words:

In the larger context of Mill’s argument, this risk is meant to illustrate a way things go wrong in the absence of regularly encountering good faith intellectual disagreement. By being put in a position to entertain doubts about one’s core beliefs, or by defending them in conversation with others, one is more likely to feel the force of those beliefs because one has appreciated the arguments for them. This, Mill thinks, makes them less likely to end up isolated from the rest of our practical reasoning. But encountering disagreement is only one way to maintain a belief’s efficacy in our practical reasoning. Plausibly, another way to maintain a belief’s power in our minds is to seek out opportunities to act so as to express that belief. Saying “I am my brother’s keeper” doesn’t necessarily direct one’s attention to what follows from that in the same way that taking care of one’s brothers does. So, if we are concerned to avoid ending up in the situation Mill describes—wherein our commitments are impotent vis-à-vis our actual choices—then we’d be well-served both by regularly encountering good faith disagreement and by regularly encountering situations where we act intentionally on reasons that flow from our ideals. Our moral commitments and personal ideals risk becoming untethered from our agency unless we employ them in thought and action.

Seana Shiffrin (Reference Shiffrin, Wallace, Pettit, Scheffler and Smith2004) makes a similar point in an argument for providing religious accommodations. She notes that one role formal protections for religious practices can play is ensuring that people have opportunities to make religious decisions on religious grounds. For example, suppose that Aditi is considering whether to fast as part of a celebration of Navaratri. Her family values their Hindu identity, but since going away to college Aditi has felt herself feeling more ambivalent about her religious commitments. If Aditi’s university has a religious accommodation policy, then Aditi can make this decision just based on her assessment of whether she still identifies as Hindu and what she thinks follows from that commitment. In contrast, if there is no accommodation policy, then other considerations might swamp Aditi’s reflection on religious questions. For instance, Aditi might decide not to fast because doing so helps ensure that she’ll be feeling her best at the time of an exam. We think our observations about expressive actions help to explain why something would be lost without the accommodation policy. A decision not to fast that had to balance Aditi’s introspection about who she is with pragmatic considerations about preparation for her exam wouldn’t be as meaningful to her. And as with Mill’s worry, the force of this observation doesn’t seem to depend on the fact that Aditi is resolving a religious question. If Allyson is deciding whether to fast as part of a political protest, then one might hope for her that she has the opportunity to decide this based on the way she sorts out her political ideals. Similarly for other kinds of commitments that are important to who we are.

Notice that the situation Aditi and Allyson are in parallels Sasha’s situation in Dish Duty. Their courses’ exam schedules change the justifying reasons in a way that has downstream effects on their motivating reasons, which in turn changes the possible meanings of their actions, and thereby affects those actions’ meaningfulness to them. So, taken together, Mill’s and Shiffrin’s points suggest that our interest in the meaning of our actions can be morally weighty. We have reason to seek out or create our own opportunities for expressive action and to avoid gratuitous limitations of others’ like opportunities. Doing so helps preserve the meaningfulness of our actions and thereby our lives.

5. Implications for Individual Morality

Our argument has had a lot of moving pieces, so let’s take stock. Our ability to express ourselves matters. So, we have (pro tanto, defeasible) reason not to interfere with others’ ability to do so. Sometimes, we express ourselves by performing actions that reveal our motivating reasons. That kind of expression can be particularly meaningful, to us and to others. Like other kinds of expression, our ability to express ourselves in this way depends on context. Often, though, the kind of context relevant to those expressive actions is sensitive to changes in what justifying reasons are available to us. As a result, our reasons not to interfere with others’ ability to express themselves extend to being careful not to interfere with the context of their expressive action. This will often mean arranging things so that certain kinds of decisions can be made by focusing on reasons of a certain type (e.g., when religious accommodation policies protect the meaningfulness of actions based on religious reasons).

One might worry that awareness of this fact will be morally paralyzing. Interactions whereby we change the reasons relevant to someone’s choices are ubiquitous. As we’ve argued, this means we frequently risk costs to others’ deliberative autonomy. By pointing out these costs, however, we don’t mean to suggest that we should let fear permeate all of our social interactions, policing every request, recommendation, or demand, lest we inadvertently rob another of an expressive opportunity. We mean only to suggest there is a level of care owed when engaging with others in ways which are apt to affect their expressive options; we should tread carefully when trekking through the space of reasons. An analogy with giving advice is illuminating here. Giving unsolicited advice can go wrong, e.g., by being unwanted, intrusive, or condescending. But, it needn’t follow that we should never offer advice unless we’re asked. Instead, we should develop a sensitivity to the ways unsolicited advice can go wrong and be careful when we’re tempted to offer it. We think the same is true about actions that risk undermining the meaning or meaningfulness of others’ actions. Indeed, one goal of this article is to draw attention to the need for further theorizing about what being appropriately sensitive to this possibility requires.

In the spirit of starting that conversation, we offer some initial thoughts about when it’s particularly important to preserve the kinds of expressive opportunities with which we’ve been concerned. First, opportunities to express something are more important when they are thinner on the ground. Suppose that when you go out to dinner with your parents, they always pay. Against that background, you have fewer opportunities to express reciprocal care. So, your other opportunities to do so take on greater importance. Second, having an opportunity to express something by revealing your motivating reasons takes on extra importance when other ways of expressing it are less effective. For example, suppose that in your linguistic community, people routinely say “let’s catch up sometime” when they run into old acquaintances, whether they actually want to do so or not. If you want to communicate a sincere desire to reconnect, you’ll be better served by Scanlonian meaning than linguistic meaning. So, it’s particularly important that you have opportunities to do things that could express that desire. Third, the importance of having an expressive opportunity scales with the importance of what is being expressed. It’s important to have opportunities to tell your loved ones that you love them; it’s less important to have opportunities to inform them of what you had for breakfast. By “importance” here, we mean to be ecumenical between objective conceptions of what’s important—fill in your favorite account of the good—and subjective conceptions. Communicating your love for your loved ones might be important because such relationships are a central part of a good life, because your relationships with them are central to your sense of who you are, or—plausibly—both.

With this sense of when it’s particularly important to protect expressive opportunities, we’re better able to think through when to be wary of interference with expressive options. As we noted above, accommodations for religious practice serve to protect opportunities to express religious commitments. This makes sense in light of our second and third observations above: religious commitments are the kind of thing it’s easy to say and so more informative to show, and they’re often central to people’s self-conceptions. That these kinds of things can be anticipated in advance suggests one further principle about how we should navigate cases where what we do might constrain others’ expressive options. Interference is worse when it’s more foreseeable. It’s one thing if Sasha’s desires are news to her parents; it’s another if they know she’s been on the lookout for opportunities to express her generosity. When they know such opportunities are important to Sasha, there’s more of an onus on them to take care to preserve them.

6. Political Implications

In this article, we have focused primarily on cases where one person’s actions change the expressive opportunities for another person. That’s no accident: we think that susceptibility to this kind of interference is one distinctive feature of expressive action. The context that determines what we can express through words typically involves social facts (like what a linguistic token means around here), while the context that determines what we can express through Scanlonian meaning is often a person-level fact (like whether Sasha has been assigned chore duty). This disanalogy between linguistic expression and expression in action is what gives rise to the moral questions we discussed in the previous section. Nonetheless, we think that systemic features of a society, like the political or economic system in which we live, can also shape what we can express through our actions. We turn our attention to those kinds of cases here, and the ways our arguments in this article might help to illuminate larger, ongoing projects in philosophy. None of what follows is meant as an endorsement of any particular view. Our goal here is just to show that what we’ve argued has the virtue of fecundity.

First, there is a critique of capitalism, going at least as far back as Marx, that capitalist production infects our motivations and thereby impoverishes our lives.Footnote ¹⁰ In the Economic and Philosophic Manuscripts of 1844, Marx laments that under capitalism our fundamental life-activity of productively transforming the external world presents itself to us “merely as a means of satisfying a need—the need to maintain […] physical existence” (Reference Marx, Engels and Robert1978: 75-6). “Life itself”, he continues, “appears only as a means to life” (76). In contrast, what would it be for us to produce in a way consistent with our humanity? He states that such a mode of production would be one wherein we affirm ourselves by doing work that directly satisfies the “human need” of another (Reference Marx and O’Malley1994: 95). In particularly romantic prose, Marx writes that in working under ideal conditions he would “be felt by you as a complement of your own being, as a necessary part of yourself, and so would know myself to be confirmed in your thought and in your love” (Reference Marx and O’Malley1994: 96).

Of course, much here hangs on Marx’s perfectionistic conception of the human good, but one can extract a core which is less robust in its commitments. At least usually, a central good in a human life is engaging in purposeful, productive activity. This activity is especially meaningful when one engages in it to make a difference in one’s community—to help others, i.e., to serve “a human need.” But under capitalist production and a distribution of goods mediated through an impersonal market economy, it becomes quite difficult for us to produce in this way. Or, to bring it around to this article, it becomes quite difficult to engage in productive activity with this motivation. We tend to see our working life as a necessary evil, required to do other things that actually bring us fulfillment, or just a way of keeping a roof over our heads. And in this way, what might be an especially meaningful human activity becomes “merely a means.” If one finds this plausible, then it looks like one can fruitfully employ our analysis here to give additional content to Marx’s critique of capitalist society. Capitalism, the thought would go, systematically shifts our justifying reasons for working, and, in doing so, alters our motivating reasons for engaging in productive activity. This, in turn, makes our lives less meaningful.

A second, and related, application concerns questions about the moral limits of markets. Authors like Debra Satz (Reference Satz2012) and Michael Sandel (Reference Sandel2013) have argued that the sale of certain goods like organs, sex, and political power has problematic social implications. Our analysis here is well-suited to play a role in explaining why this might be. There are two possible ways one could incorporate what we say into such an argument. First, a society wherein sex, for instance, is sold on a large scale may engender attitudes and norms which change the social meaning of sexual activity from a significant form of connection to something more transactional. Second, such a social context may affect the motivations of those engaging in sexual activity. Each of these might plausibly be seen as a loss of expressive options.

Again, we note that in this article we have not argued for this particular critique of capitalism or these particular views about the moral limits of markets. Our aim in this section isn’t to endorse anything specific; rather, we aim to show that the arguments earlier in this article can inform those debates by helping to identify what’s at stake.