Duty and Deontology

Abstract

A too rarely emphasized feature of modern deontological ethics is the structure of its directives. Faced with alternatives, the question for the moral agent is “which, if either, must I perform (or avoid)?” Getting it right, one is, morally speaking, done…until the next set of freighted options presents. We should wonder whether this makes sense: whether there is not a more complex structure to deontological requirements that resists the “one and done” idea. Rehabilitating the Kantian idea of duty as a value-based deliberative principle, I argue for a more plausible deontology whose requirements are often temporally extended and interpersonally complex.

For many who are drawn to deontology’s side in ethics, “duty” is rarely part of its attraction. There is the overtone of oppression—of demand and constraint—a measure of distance from the good in human life. We accept some things we find unattractive that are also necessary, like going to the dentist. But the moral idea of duty no longer seems to be one of them. This is not to point to a general aversion to moral requirements per se, but to the thought that there is anything the idea of duty does in our moral discourse that is not done better by a mix of more congenial notions. Setting ourselves to act well, we do not look to doing our duty: what we ought to do is what a good person would do, or what we have most reason to do. And where there are moral reasons in play, we should act for them if and as they dominate the balance of reasons according to some practical accounting principle. We may talk about duties—the duties of friendship or the promissory duty—but when it is not just a legacy term, or an emphatic (“But it’s your duty!”), the talk of “duty” in a region of moral importance does not seem informative. I disagree. There is a serious gap in the prevailing view of the interior structure of deontology (and of non-consequentialist theories generally), that a deepened account of duty is well-suited to fill. It is key to a more plausible and much richer idea of deontology than we are used to having.

1. Some Reasons to Be Wary of Reasons

One way to get a bead on the gap is to consider a few things about reasons deontology, where what is to be done is a resultant of weighing and balancing the various kinds of reasons present in our circumstances of action. Being hungry gives me a reason to eat; fatigue to sleep; a friend’s depression a reason to spend time with them. Some reasons reflect moral considerations; some the merely practical. Sometimes, the resultant or sum is a pragmatic recommendation (early to bed, early to rise…); sometimes, it is an obligation (you ought to [you should, you must] keep your promise). The obligation reflects the emergent fact that the weight of reasons to keep your promise defeats competing reasons (finding that it is burdensome to keep the promise, or having a better option, are not sufficient reasons for default). But how does the weighing and balancing go so that the reasons outcome favoring keeping the promise is a sure (or sure enough) thing? So, a question about how to reason with reasons.

Reasons of well-being play a familiar role in our reasons arithmetic, giving weight to considerations of diet and exercise, though sometimes to days at the beach or streaming episodes of Succession. Sometimes the decision-metric is subjective—what I most want or want to want; sometimes not—I defer to my doctor or my guru, taking their wisdom as my guide. I also take myself to have reasons not to do certain things which could be beneficial to me: I should not betray a friend, engage in wage theft, refuse easy-to-meet requests for help. It would seem that there is a different source that explains the strength of these reasons, and explains it in the right way. We have a reason to eat because of nutritive need. The need explains the source and the strength of the reasons. I also have reason to eat because I accepted an invitation to dinner. A social norm does the explaining here. Which might suggest a parallel story for moral norms as the source of moral reasons. But when we think of how the social norm does the reasons explaining—I am embedded in a social world made possible by these norms, a world I value or at least accept—the parallelism that is to get us distinctively moral reasons is hard to make out. Our moral reasons are not well understood as conventions we just find ourselves valuing or that weigh heavily because of social sanctions. They may have connections with social conventions, even give rise to some, but it is reasonable to want them to derive their content and directive significance independent of contingencies of convention and attitudes. ¹

Enter contractualism. As Scanlon presents it, what marks core concerns of morality, what makes its reasons results special, is the value persons place in having their other-affecting actions be beyond reasonable reproach. That anchors what we owe to each other. First-order reasons are subjective, in the grammar of preferences. ² I have my reasons, you have yours. ³ When the actions my reasons support interact with actions or conditions you have reason to care about, that is the moment the moral question is in order: have I acted in a way you can reasonably reproach me for? You may not like the way I acted; you may have suffered a set-back. But I would not warrant reproach if my action (or action-type) is one that would be allowed by any set of principles no one could reasonably reject. ⁴ We could say: moral reasons flow from principles sourced in the shared aim to live in a social world of reasonable regard. The weight and content of moral reasons depend on two factors: value input and the standard of reasonable rejection. So, if, as is reasonable, we value bodily integrity, then there will be strong moral reasons not to harm others, except in circumstances where, for example, it would not be reasonable to reject all principles that permitted self-defense, or where causing a small harm would avert much greater harm. The contractualist function operates pairwise. Forbid or permit: personal preferences (about food or intimate partners), building houses on a flood plain, mandatory public schooling, selling bodily organs, accepting a certain death-rate as the price of achieving herd immunity.

But note that the work of the contractualist principle is not to provide guidance about how to be a friend, or how and when to benefit others, or what kinds of treatment count as fair. Given pairs of reasons about actions that affect friends, or helping, or distributing some benefit, it offers a framework for explaining why one of the options in the pair-wise comparison might be wrong. What contractualism cannot do (and is not meant to do) through its pair-wise renderings is deliver deliberatively robust moral content. Finding out that we may not make a deceitful promise does not tell us what is involved in making one responsibly. This is the gap. I would say: What is missing is a competent grasp of the duty of promising.

2. Duty, Value, and Having a Say

Approaching the gap on a different track, we might note that morality as we live it comes with a distinctive phenomenology. There are some things we take ourselves to know if we know anything. Pain is bad; suffering to be averted. Yet, some pain may be necessary to avert greater pain, or is on the route to some serious gain in physical or mental power. We might then think: wouldn’t it obviously be better if we could realize the various things we aim for without any pain? It is both imaginable and better that I hammer the nail without injuring my thumb. But—is it also imaginable, would it be better, that we love finite and vulnerable beings without pain at their loss? Our idea of love is shaped by that vulnerability, and not something we can avoid if we are going to love people, dogs, even works of art. ⁵ Some of the value of our physical triumphs comes from pain understood and overcome. If vulnerability to pain and suffering is integral to some of our most valued activities (and not minimizable in the way of more careful carpentry), then we need something more than principles about avoiding and averting.

So, imagine a trip to the dentist. Someday there may be painless dentistry, and that would be good. But for now, if the dentist is a good one, in addition to the poking, scraping, and drilling (I will stop with that), there is talking. “This won’t hurt.” or “This will hurt just a little.” or “Let me know if this hurts too much, I can numb the area.”—and so on. The pain felt will be the same if the dentist is careful but silent, yet the moral situation seems different. It is not about consent, which can be granted globally at the outset. It is about the talking. What the talking accomplishes that global consent cannot is an on-going relation of reciprocal permission or authorization. From the dentist—I know that you know that I know…that this (what I am doing to you) will hurt. Knowing that, I want you to have a say about how it is for you as things go along. It is a recognition that things are happening that might make prior consent moot, either because one can be wrong about one’s pain tolerance, or because it is the nature of incursions into the body that pain and its effects are unpredictable. With or without the talking, the pain is (unfortunately) part of an appropriate or even necessary course of treatment. With the talking, the patient is a participant, her feelings and her agency remain in view, salient throughout. ⁶ The direction of attention is not the pain, but the person in pain.

Suppose you agree with this. Wouldn’t it just show that the dentist has a good reason to engage with their patient this way? And isn’t that enough? The option for participation and agency-recognition supports a reason, even a strong one, to take that path. But there are other reasons that can be in play—about efficiency, stress-avoidance, uniform treatment protocols, controlling compassion—reasons which call for more weighing and balancing. Would it make a significant difference if the dentist had a duty to engage the patient as a participant?

Both the balance of reasons and a contractualist verdict can tell you whether to act or not. And having acted as you should, then what? You are now in new circumstances, uncharted territory, looking for the next balance or verdict, which may or may not be connected to what you just did. What is missing is some connection to underlying value that can continue to direct us as our actions place us in new circumstances. It may be, at the end of a long dental procedure, that the patient has had enough of a say and just wants it to be over. This is not well described by saying that at t ₁ there was reason to talk and at t ₂ there was reason not to, though that is true. Our dentist acted throughout on a duty of recognition that gave the patient a say which they could waive if that is what they wanted. One might put it that the parties act and continue to act working from the same (somewhat improvisatory) script.

We tend to think of duties as a source of obligating rules. A duty can play that role, but duties can also be understood as the action-oriented face of fundamental moral value. The difference that duty then makes is in its requirement of on-going deliberative attention to the value that sets the duty’s form and content. Let us suppose the value is something about our agency. ⁷ The role of duty, then, is to manage contexts of agential vulnerability that cannot readily be avoided (and maybe not avoided at all if we are talking about the normal human life-form we now share). Pain and its vicissitudes will show up in many circumstances of action (some of them pleasurable) that a value-responsive duty manages. Pain needs to be understood, integrated, not just avoided or medicated, in both self- and other-affecting action. The duty resists passive enduring. It can set an extended discipline, sometimes a shared practice. In the case at hand, it directs us to differentiate the role of interpersonal communication as a way of working through a morally fraught situation from, say, using soothing music to allay anxiety.

Looking at duty this way signals its role in a deontology. Rawls defined deontology simply, as a kind of moral theory that “either does not specify the good independent of the right, or does not interpret the right as maximizing the good.” ⁸ Like justice as fairness, anchoring duties in the value of rational agency takes the second way. It is a mark of deontology that duties provide evaluative and directive standards for actions. It is a feature of the deontology I am excavating here that the object of duties is not just actions as causes, but “complete actions”: temporally extended and including motive. These are the elements that bear a duty’s value.

This is enough to warn us against the temptation to use the language of agent-to-patient in describing moral interactions, where the designated agent is the salient actor, the one with authority or expertise and so, primary responsibility. ⁹ The default should be agent-to-agent. Even though “right now” one party is acted upon by another, we do better regarding them as in a potentially cooperative enterprise, regardless of who is the cause of the immediate fraught activity. As with the case of the careful dentist, further back it was the receiving agent who initiated the relationship, the end being promoted is their end, the one thought of as the “active” agent is the receiving agent’s agent, their delegate or proxy.

When the pipes burst in my house, I engage a plumber to stand in for me, to be my agent in making the repair. I authorize the work. Too often, the plumber acts like “the doctor” to my damaged house. ¹⁰ But things could be different. The plumber and I might talk about how the work should be done: what materials to use; the recycling possibilities; the hourly wages and benefits for any temporary workers. I provide information about work that has been done before, whether I will be open to expanding the job were the plumber to find new vulnerabilities or shoddy repairs in the water system. We could agree: the plumber will offer triage information; I will decide. The values that guide this conversation belong to an agent-to-agent moral space where we each have a duty to enable the other to act responsibly.

3. Filling Out the Job Description for Duty

Taking this idea seriously—that we have a responsibility toward ourselves and others that the terms of our interaction be lucid and forward looking (even if we are responding to something that has already happened)—will inform the way we think about duties and their directives. Suppose we ask: “What’s it for?” about a duty. One familiar kind of answer points to the value of having sanction-bearing rules in areas of contested interpersonal activity, rules that can resolve indeterminacies in outcomes, providing assurance of fair treatment or repair. On the conception of duties I am advancing, they are transformative of the activities they address. The model for this is suggested by Kant’s moral defense of juridical duties. There is a moral challenge that arises from things that persons will unavoidably do for production and reproduction that, considered as natural actions (using land, exchanging goods, entering intimate domestic relations), threaten persons’ equality and freedom by making them vulnerable to others’ power. ¹¹ Having a good will cannot fix a threat that is integral to a kind of relationship. Kant argues that a legal system of the right sort can fix this because its rights and duties change the moral form of the natural relationships, giving public status to the parties and defensible standards for the activities constitutive of their interactions. The parties thereby acquire enforceable agent-to-agent responsibility for managing their public roles.

We can extend this model to the non-juridical interpersonal world where we encounter different vulnerabilities. We do not have to be slaves to enter or just wind up in relationships of dependency (or co-dependency) that efface our effective agency. This can happen as my choices become entangled with others’ helping efforts, or where my dissent makes little difference to how we go on, or my information and advice fail to register in a shared inquiry, or where the tasks I am assigned in some cooperative activity impose hardship that could have been avoided if I’d had a real say in the distribution of jobs. Or, as I earlier suggested, in the ways one person is doing something that causes me pain, for my benefit, even with my prior consent.

The common duty-evoking element is non-malicious activity that exploits or just ignores vulnerabilities in embodied, socially situated, agency. The activities in question are normal, sometimes natural, often reasonable, not impermissible. They can occur in situations where we intend to act well, even helpfully (with or without sacrifice). Their structure is familiar in situations where we act paternalistically, choosing not to engage with or working around another’s agency, substituting our own judgment or action for what is properly their own. ¹² The moral resolution is that, in our interactions with others, when we are engaged with their interests (promoting or avoiding a set-back to them, balancing benefits and burdens), or when we aim to do for them, perhaps pre-emptively, what we judge they would if they are reasonable, we have a duty not just to take steps to remain connected to agency values in our manner of acting, but also to be explicitly engaged with them in our acting, in the instance and as things evolve.

The object of duty here is the ubiquitous acknowledgment across our engagement with others of each of them as an equal and authoritative deliberator. It is a duty of respect. When we act agent-to-agent, the moral baseline is shared authority over how things go. ¹³ That is the moral force of the “having a say” idea. If Scanlonian contractualism works off the motive of acting only in ways we could justify to those affected, the duty of respect requires that, where possible, one actually engage in a justificatory conversational practice. So not just an action-directive, but the engagement of a complex, value-tracking motive.

4. Colloquy

This points to an under-explored subject about the role of talking as an element of value-directed duty compliance. Consider helping duties. Sometimes, a need can be obvious, but determining how one should help is difficult, given the complexity of the need or the opacity of preferences about the delivery of aid. Other times, because of the intimate or long-term nature of the need, what is at stake is how someone’s life will go, once the help is initiated. We may be experts or have lots of experience. We could just act and do what we quite reasonably judge to be best. But an agency-respecting duty requires that we talk and, where possible, cede or share authority with the receiving agent. (Advance directives work in this moral space.) However, letting someone have a say, acknowledging their agency, does not amount to giving up our agency and responsibility. It is agent(s)-to-agent(s), and the condition of colloquy is reciprocal.

My use of the term “colloquy” is meant to flag that however casual the interaction, there is something formal about this kind of moral talking. ¹⁴ Not all circumstances of interpersonal action call for colloquy. I do not have to engage you in prior discussion if I am giving you a small birthday present, though I may have to if what I intend to give you is something large (a new car), or demanding (a puppy). And I may have to act without consulting in an emergency (though there are some communicative structures in place for emergencies—medical alert bracelets, organ donor cards—that stand in for what we want to have said). But across a wide range of ordinary and extraordinary actions, we can and should talk in a duty-relevant way with those we will affect, before and during our acting. This involves more than exchanging reasons.

In order to speak and act so that our interactions are effectively reciprocal, we require common terms, established patterns of action, signals of understanding. We do not reach shared decision points as if we had just been dropped into the world and need to find our way through a thicket of reasons. Principles that identify actions as wrong or impermissible in the circumstances are necessary, but not sufficient. Responses at different times need to cohere with one another; how I engage with you now initiates something between us that may open some paths, foreclose others; we incur new responsibilities; we may become vulnerable. It would be unbearable/unthinkable/morally impossible if all of the different ways this sort of thing happened were always a surprise. Sometimes all that is required of us is that we avoid doing what a duty forbids (by its directive content)—not to take more than our share or violate a right or fail to show up as promised. At other times we determine what we need to do by attending to what makes the duty matter to both parties—making use of the connection between a duty’s principle and value. When we would do that, our internal careful deliberation may not be enough.

To get a sense of what this involves, let’s look at the promissory duty. ¹⁵ Directives about promise-making, promise-keeping, and promise-breaking are the content conditions (or rules) of promising. There is also what promising is for—its object or value. If we take the point of promising to be about securing cooperation without depending on altruism, that puts the focus on avoiding promise-breaking since defection defeats promising’s point. But if we think the point of promising is to enable us to give another limited authority over what we do, our attention would be on trust as well as security of performance: that both parties act in good faith; that defection plus compensation is not an equivalent; that the promisee will not take inappropriate advantage. If it is the first, we should be concerned with preparation and performance, and perhaps in anticipation, possible repair. That is because the value of promising on this interpretation is as an instrument in the service of cooperation for a goal. If it is the second, our concern is about maintaining the relationship that the promise brings about, or its fit with other relationships the parties to the promise may have. Here, because the value of the promise is internal to it, the transfer of authority has to be guided by the value, and so must be a respectful or autonomy preserving transfer. The lesson for us is that both interpretations introduce complex, though different, bundles of things to take care of. The value of promising in play can affect both initiation and completion conditions as well as a host of ancillary responsibilities. ¹⁶

Suppose I would make a promise to do something important for you. You are at risk in relying, but it is reasonable in the circumstances to rely. There is an outside chance that I will not be able to perform, for reasons that would excuse, or even justify my default. Deliberative strictures of the duty forbid making a promise the to-be promisor knows they cannot keep, but they do not forbid making a promise they might not be able to keep. Every promise is like that. And there is no fixed schedule of obligation and risk, though there are some exempting conditions built in which annul the default obligation: about disasters and threats to life, when trivial promises turn out to have enormous effects (the “ceteris paribus” clauses of normal moral literacy ¹⁷). One way of managing the situation is for me to run a cost–benefit analysis to determine whether the risk for you is reasonable, and if it is, conclude that it is neither impermissible nor wrong to go ahead with my making the promise. You can do your own analysis. Another way is for me to explicitly lay out my view of the situation. What is reasonable from your point of view may not be captured by my cost–benefit judgment. We might be pleasantly surprised at the convergence the colloquy creates. Whichever way it goes, in the promisor’s explicitly laying out the unlikely but real risk, the promisee is given the opportunity to share responsibility for the promise. Here there is an expansion of the duty’s requirement toward colloquy. But now notice: if that’s right, there seems to be something like an imperfect duty inside the initiation conditions of the perfect one. ¹⁸

And there is more. Since promise-making both obligates the promisor and gives certain powers to the promisee, promising either creates or adjusts a moral relationship between the parties. Much attention has been paid to the completion condition of the promise (whether there must be sincere uptake), the wrong in negligent and false promises, excusing conditions, compensation and repair. The dramatis personae have assigned roles in a two-party relationship of obligation and performance. But suppose the parties, knowing the risks and conditions, recognize that there is a conversation to be had (perhaps more than one) between them, not just before they proceed, but also as things go on. A promise is not a vehicle set in motion where all the due care that anyone owes belongs to the initiation conditions and the performance of the promisor. Suppose the promisee comes to have information she could share that would greatly ease the promisor’s performance, or could take steps to increase the likelihood of success. It is not impermissible not to do these things. It is not a role violation. But to the extent one sees that the value of promising invites the colloquy (as a matter of due care), then the resulting shared responsibility for the promise can make such omissions morally wrong.

Colloquy can also bring an end to a duty. Suppose I recognize your need and provide help. I am (suppose reasonably) not wanting to continue to provide the same help tomorrow that I offered today, though I remain, from your point of view, in exactly the same position to be helpful. You have no right or claim that I continue to help or do more than I have already done. But I should nonetheless initiate a conversation about my limits, and perhaps, in light of that, help you find a next helper, though I am not obligated to do so, and might not have seen any reason to had I not had the conversation, which was the morally appropriate thing to do.

I have been focusing on the interpersonal implications of some duties—the way their compliance conditions can be actively trans-personal and temporally open-ended. These features flow fairly naturally from the dynamic and non-episodic nature of imperfect duties. The surprise is when this happens with perfect duties. We are used to a “one and done” model of perfect duty compliance. However, some of the data I have been canvassing suggests there is more to these duties than being square with their rule. Where we should do more, what we should do calls for an exercise of discretion: so again, something like an imperfect duty latent in the perfect one.

5. Duty and Deliberation

With this amplification of a perfect duty’s requirement, we take up residence in an unfamiliar moral landscape. Classical deontology offers a hard division between the strict action-prescriptions of perfect duties and agent-relative standards for imperfect duties. While the model is not tight—there are exceptions on the strict side, limits on agent-discretion on the other—the division into separate kinds of duty has seemed part of the moral bedrock. As we follow out the changed idea of what a duty is, we may find this bedrock to be procrustean.

Recall: In the theoretical framework I am using, the fundamental value that duty protects and promotes is the capacity for agency, more specifically, for moral agency—that we can act for moral reasons, take an interest in others’ interests, commit to projects that transcend our personal goals, assume and share responsibility for the shape of our own and others’ lives, regard ourselves as stewards of various non-human interests. These are sites of practical powers and vulnerabilities. Duties direct the use of our (and others’) powers as we take on projects and activities, including moral projects and activities, along with a system of alerts and repairs to the misuse of and threats to the capacities’ distinctive vulnerabilities. We can be injured, physically, emotionally, and mentally. We can be betrayed and oppressed. We become vulnerable through our loves—personal and political. We have ambitions; we are tempted by irrationalities. We require certain social or political institutions in order to affirm and secure our moral status. ¹⁹

The upshot is a set of duties that govern distinctive arenas of practical concern (property, bodily integrity, speech, friendship, etc.). ²⁰ Each duty generates a complex array of deliberative requirements in its area of authority with a distinctive moral architecture. The structural elements of a duty are: (1) vertical: going from strict requirements down to more open-ended directives, (2) hermeneutic: how things develop descending is affected by and affects higher-up elements, and (3) beholden to a coherence condition: a duty has a place in a system of duties.

Putting this structure to work will make its importance clearer. Let’s consider the moral status of the body. The body (natural and prosthetic) is a sine qua non of our agency, a condition of fundamental value’s being realized in choice and action, and therefore high on the list of things that warrant protection and on-going moral attention. ²¹ Some of the protections are narrow and stringent, blocking actions the performance of which are not consistent with the power to realize intentions in acting. ²² They will tend to have the form of perfect duties. We rightfully insist on others’ restraint. We require public protection of our bodies, authorizing coercion (as well as protection from abuse by public agencies that are allowed to use coercive force). This is the apex of the vertical moral order concerning our vulnerability to bodily injury and harm.

The narrow constraints at the apex are not a formula for generating permissibility verdicts for a wide-open comparative consideration of options; they introduce deliberative boundaries for what the options can be. Some patterns of instrumental reasoning that we commonly use as we negotiate the physical world are off limits when the physical object is the human body. ²³ We begin deliberation knowing that what we can do to and with the bodies of persons is highly constrained. ²⁴ Consent matters, though it is not always dispositive. There will be duties to self that require us to protect our own bodies. My wanting you to do something that will impair my agential functioning is not a license for me (or you) to by-pass the duty. In the body-affecting things we would do to others, we must remain mindful of their authority and agency.

That it is a deliberative constraint that excludes options explains why the duty in its first stage looks like a conventional perfect duty imposing restrictions on actions. But because it is a deliberative constraint, that kind of restriction is a result, not a formal feature of the duty. Limits on action can be strict in some regions of acting where the body is affected, not in others. The variation belongs to the duty. Risk of harm can be unavoidable in the acquisition of some skills (or in the efficient movement through social space—even if it is bicycles, not cars). There will be questions about harm to others’ bodies and harm to self in sports and play and some consensual sexual practices. Sport and sexual freedom are defensible on many grounds, but what is at stake here is arguably about accessing bodily powers for the one, and ways of more profoundly inhabiting the body’s sensorium for the other. The arguments the duty’s deliberative framework draws on to permit or encourage some actions and constrain others will reflect these more domain-specific articulations of the agency values that gives the human body moral status, adding different deliberative directives for proceeding in these various practical spaces.

Since a duty protects and promotes a locus of fundamental value, the value will continue to shape content and form descending the duty’s vertical axis. If what is of value is vulnerable in contingent ways—as the body is to accident, say, or disease or age, or changing social role—then in addition to deliberative constraints on action-options, there will be deliberative directives to tend and promote, to repair, and in some cases to bring about conditions for flourishing. The same duty will then call for the adoption of an end in addition to setting deliberative constraints on actions. This is the region of health, of managing the appetites, of developing various abilities, and providing support, but also on the public side of morality, of health care and policies of environmental protection. There is discretion about the specific ways we (personally and collectively) reflect the underlying value as we take on these projects. The imperfect elements of the duty are distinct but not separable from the perfect ones, continuing the elaboration of the work of the same duty across a wide range of agential needs and activities.

Duties about communication and speech are another region where the interplay of the elements of the duties is vivid. We could have a perfect or nearly perfect constraint on lying and once we see why we should, extend the duty toward discretion. Speech conveys information, enables inquiry and cooperation; it is also a vehicle for solace, for creativity, and for fun. It is a power that puts us in a position to do all sorts of things, not all of them good. And not all of the abuses of speech involve lying. Its potential for causing harm is a reminder that not all occasions of speech are truth-apt. Consider the harms we may cause when we just speak our mind. “What do you think?” is not a moral command. We have discretion over what we reveal, and, assuming no prior commitment or specific obligation, we violate no duty in keeping our counsel, and have duty-based reasons (here is the imperfect element) to rein in our power to harm using words that express only some of our truths. The urgency of an imperfect element—to protect privacy, for example—can sometimes call for the violation of a perfect one—say, to keep a promise or tell the truth. There is no reason to assume ex ante that one element is always dominant, or that one is a limiting condition on acting for the other. ²⁵

What emerges from all this is a conception of a duty as an array of nested deliberative directives, where perfect and imperfect do not name kinds of duty but are adjectives that describe different modes or forms of a duty’s deliberative requirements (I have begun calling them perfect and imperfect elements). Which element presents will depend on what is at stake for a region of moral concern. The result is not, however, a formal division of moral space, but something more like a dynamic partition. We will get the (functional equivalent) of an impermissibility verdict where it is needed (lives and limbs not available for advantage calculations). But there can be situational adjustment when the threat to fundamental value can be addressed in different ways. It is wrong, verging on the impermissible, to place a person in a situation of high fear potential, even if the danger is manageable, their fear excessive, and the cost of their reticence is not trivial. However, there may be steps we could take that can move the deliberative needle. We can spend time with them, slow the events down, talk with them and “be there” in ways that make our encouraging sincere and authentic. Sometimes what is needed is providing the ex-ante moral equivalent of insurance protection, shifting or expanding the burdens of responsibility for a morally difficult action. Joint action can alter the choice situation, as can other resources that affect the moral division of labor. All of this would be under a duty’s deliberative direction.

6. The Consilience of Duties

In filling out the structure of a duty, attention also needs to be paid to the ways a duty has horizontal connections to other duties, either as part of a system of duties or, within the system, to other duties that are non-accidentally adjacent. To have a term for this, I call it “the consilience of duties.” ²⁶ Beneficence and respect live in each other’s houses because of the way each draws on agency values. Being the victim of disrespect diminishes a person’s ability to adopt and pursue independent ends, and if what it is to respect someone is to acknowledge the value of their agency, not in the abstract but in their life activity, that is not consistent with indifference to their needs. Public welfare systems that demean recipients (long waits, excessive and intrusive forms, inadequate explanations) behave in morally unacceptable ways despite providing a real benefit. ²⁷ They do not do one thing right and get another thing wrong. One of the tasks of due care is to attend to the intersection of duties, to manage consilience, making sure that the manner of serving one duty does not set up failure at another, as well as proactively using the occasion of acting for one duty to meet the demands of another.

When more than one duty applies at the same time the usual worry is about conflicts of duty—where only one of the duties that applies can, as it happens, be met. We then need to know which duty gets priority, to think about what happens with the unmet duty, and to consider whether, in theory or in practice, the conflict indicates a flaw in the system of duties. What the possible consilience of duties tells us is that when duties co-apply, they can together direct an agent’s deliberation about what to do. They can do this not as is the case when two duties just happen to be satisfied with one doing (such duties are circumstantially concurrent), but when the manner of satisfying one duty is implicated and transformed by the other. ²⁸ The resulting consilience can deepen or modify our understanding of what is at stake in the deliberative protections both duties provide.

Thinking about consilience of duties as the positive flip side of conflicts of duties suggests that the focus in the conflict cases on modal contradiction and duty counting (an “ought implies can” question) may miss a deeper issue. ²⁹ The problem when there is conflict is not that two duties apply and cannot both be met. All sorts of things stand in the way of satisfying a duty that are consistent with its application. A more important issue is whether the underlying values of the conflicting duties are incompatible. That not every advance in freedom is an advance in equality is something we might well rue. If, however, we discovered that the duties that protect these values were not consilient, that our best hope was their (unlikely) lack of conflict, we would be locked into a dismal moral space of trade-offs. Were they instead consilient, there is moral progress to be had in developing an understanding of the duties in such a way that the value of each can be interpreted or elaborated in light of the value of the other. They could then be put to work together, complementing and (hopefully) strengthening each other, though possibly on a path that neither by itself would have selected. ³⁰

The practical upshot of the consilience of duties can sometimes yield constructive convergence in the single action of a single agent, but it can also be realized in a series of connected actions done by the same or different agents. The after-an-extreme-natural-event beneficent hospitality I extend to a neighbor is consilient with public agency responses that come on-line over time (both belong to duties that apply). My hospitality may be open-ended, but where there are suitable public services, it should have an end, and part of what I offer should be helping to prepare my guests for moving on (spending extra time retrieving or replacing documents, for example). Likewise, public actors might reasonably expedite services for those who are not recipients of private beneficence in the knowledge that the common value of the two duties being realized together makes room for different choices, both in the immediate aftermath and in the later efforts to restore individual and community well-being. ³¹

It is not always easy to tell which structural element of a set of duties is in play. Here is a real case. It turned out that providing financial counseling in the setting of a free healthcare clinic increased use of the healthcare resources (more pre-natal visits, immunizations, etc.). ³² It could be a value-added phenomenon, making the clinic visits more worthwhile than health care on its own. A two-fer. But listening to what subjects said, it appears to be a consilient phenomenon, where increasing participants’ understanding of the multivalent determinants of poverty through providing support for economic independence and health care more fully respected and therefore engaged their agency.

Consilience is just one striking moral item encountered in working out this idea of a value-centered duty-based deontology. It is not proof of concept, but it is encouraging. If we follow out the idea, I expect we will need to forego the thought that a duty is fully knowable; it may only reveal itself in stages as we act under its guidance and reckon with its place in a system of duties. It may turn out that we are responsible for more—more than we first thought or more than the last time we acted—given the ways duties connect with and affect each other through our and others’ acting. On the other hand, we may not have to bear the totality of our moral responsibility alone as we negotiate the complex topography of our duties. And then there is the talking. Despite what it often looks like in philosophers’ examples, neither moral deliberation nor moral action is something we do best on our own—or silently. Those we interact with are often nearby, willing, even eager, to have a say in how what we do affects them. In our moral theories, as in our lives, we should welcome that.