It seems normal to think that we have certain special obligations to those we are close to, and sometimes these obligations might be stronger than our general moral duties. They might lead us to make tough decisions, like Euthyphro deciding to testify against his murderous father. At the least, the law reflects our sympathies towards special obligations; spousal privilege, for instance, prevents spouses from condemning each other in court.  Nevertheless, these special obligations can be burdensome as it sometimes forces us to make such ultimatums between moral duties and relationships. Some (“voluntarists”) have suggested that this burden is unfair if they are not voluntarily taken; after all, this might interfere with our freedom to choose how we live our lives. Some, like Diane Jeske, insist that we in one way or another consent to special obligations. Others, like Samuel Scheffler, are insouciant to this problem; still others, like R. Jay Wallace, also bite the bullet, but are more sympathetic to the tensions and find clever ways around them. This paper argues that there is indeed an issue, and that the attempts at resolving this issue have been unsatisfactory. The most important feature of autonomy is letting us choose our means of achieving a good life; with this in mind, the freedom to cast off the burdens of special obligations is the crux of meeting the autonomy worry. I suggest a reconciliatory approach, focusing on terminating conditions, to resolve the qualms with special (“sui generis”) obligations and the voluntarist’s worry about autonomy.

It is important to distinguish “natural duties” and “special obligations” because some argue that they are different kinds of rules, whereas others want to say special obligations are a mere species of natural duties.[1] Natural duties are broadly understood as moral rules which apply and are owed to everybody. To illustrate, some may argue for the natural duty to report murderers to the police. This natural duty would apply to everyone, so if someone was in a position to report a murderer, they ought to; moreover, such a duty is understood to be owed to everyone; one would owe it to another in virtue of being human that they do not danger others by letting murders run rampant. This is contrasted with special obligations which are owed “to some limited class of persons.” (Jeske, 529) If I make a promise with you to meet you at the café, my obligation to keep my promise and to meet you is not in virtue of you being human; rather, it is in virtue of the promiser-promisee relation, which limits my obligations to the class of persons containing only you. A different sort of special obligation is between relations with intimates: mainly, friends and family. These sorts of special obligations can be more pressing and sometimes ask us to go against natural duties: perhaps Euthyphro has a special obligation to his father that he does not report his father for murder. Familial-based special obligations seem particularly different from promise-based special obligations because we do not voluntarily enter into familial relations as we do with promise relations.[2]

            In comes the voluntarist challenge. The thesis, according to Jeske, is that “the only way that we can acquire special obligations is through some voluntary action (or actions), a voluntary action (or actions) which we know or ought to know signals the assumption of obligations…” (Jeske, 532) Voluntary actions may well be tacit or subtle, but I leave that an open question (it is an in-house debate for voluntarists). What the force of the voluntarist challenge is that there is some consent to accepting conditions. For instance, a promise to meet you at the café comes with the stipulation that you do so if you make the promise. Voluntarists have issues with special obligations, which are not voluntarily accepted. 

            There is a concern from voluntarists about the state of autonomy within the bounds of special obligations. Familial relations are often thought to be a source of special obligations, that is, acting above and beyond what natural duties require; however, familial relations are not voluntarily entered into. These non-voluntary special obligations can get in the bearer’s ways insofar as “space to develop projects and plans and to sustain their personal commitments.” (Jeske, 537) For instance, a child might have a special obligation to take care of their elderly parent, which is an investment of time and resources that the child could direct to themselves; accordingly, it is unfair that the child has this burden without having voluntarily committing themselves to the (special obligation generating) familial relationship. Simmons contributes to this conversation by pointing out that “institutional obligations” (like institutions of family) only have force when they are “freely promised,” (Simmons, 28) so perhaps there is some voluntary aspect of the parent-to-child relation; but those imposed, like the child-to-parent relation, “have no moral force at all.” (Simmons, 28) Voluntarists further challenge special obligations by pointing to how non-voluntary special obligations limit autonomy.

            A strategy to meet the voluntarist challenge is what Jeske calls the “deflationary response” (Jeske, 533). The idea here is that since familial-based special obligations are not explicitly voluntary (i.e. fail the voluntarist challenge), they must not be special obligations but natural duties. Notably, most voluntarists concede to the non-voluntary nature of natural duties. (Simmons, 29-30) Perhaps our obligations to family are not in virtue of familial relations (which would make them special obligations), but families are special cases where we can exercise this general natural duty owed to everyone. Perhaps the proximity or roles within a family give us better opportunities to exercise our natural duty of charity. This is the approach of some reductionists – to “reduce” special obligations to more general natural duties.

Let us now, pulling from Sam Scheffler, outline the basic features of non-reductionism, and what can be said about the voluntarist challenge. We often speak of special obligations (“responsibilities”) by pointing to the relationship rather than some voluntary action, or in Scheffler’s words, “by citing the nature of our relationship to that person” rather than some “specific interaction.” (Scheffler, 189) To have a “special, valued relationship,” that is not “purely instrumental,” means that I “regard the person with whom I have the relationship as capable of making additional claims on me, beyond those that people in general can make.” (Scheffler, 196) They provide “presumptively decisive reasons for actions” (Scheffler, 196). These reasons are more basic (i.e. primitive or intuitive to normative practice) than any reasons arising from interactions. [3] Additional duties may subsequently arise in virtue of interactions (whatever agreed upon normative governing conditions), but they are grounded first and foremost in the relationship. For instance, non-reductive special obligations can be present when first meeting your spouse, but reductive special obligations (e.g. not to have any other partners) may arise from a promise to marry.

Scheffler offers seven conditions under his non-reductionist picture which relationships give us special obligations. These seven conditions are necessary for non-reductionism.[4] First, a relationship must be attributed with some “noninstrumental” value – in other words, an end in itself, which excludes the possibility of valuing a relationship noninstrumentally for some instrumental advantage (i.e. “reflexively instrumental”). (Sheffler, 198) Second, relationships must be “socially salient,” so the mere fact that “you happen to have the same number of letters in your last name as John Travolta” does not count as a relationship in this sense. (Scheffler, 198) Third, valuing relationships “means valuing the relation of each of us to the other.” (Scheffler, 199) Fourth, special obligations may be stronger or weaker depending on how much one values the relationship, and may be correspondingly outweighed by other considerations. (Scheffler, 199) Fifth, “the nonreductionist principle states a sufficient condition for special responsibilities, not a necessary condition,” so there is room for reductionistic special obligations (like promises). (Scheffler, 199) Sixth, it is possible that people have special obligations when they think they do not, and it is also possible that people do not have special obligations when they think they do. (Scheffler, 199) Seventh, reasons to value relationships can skew which relationships are actually valued because reasons for valuing relationships can affect conceptions of special obligations. (Sheffler, 200) These conditions outline the basic features of a non-reductionist picture.

The argument Scheffler offers for this sort of non-reductionism is that human beings value “our relations with each other,” and this valuing of relationships “is to see it as a source of reasons for action of a distinctive kind.” (Scheffler, 200) This argument is based on a sort of teleological premise. It says that a dimension of human psychology is aimed at interpersonal relationships and that this is an essential part of being human. “It seems that whenever people value an interpersonal relationship they are apt to see it as a source of values.” (Scheffler, 190) In brief, our need for relationships affects the way we make choices, and this can be appropriately construed as special obligations.

Scheffler pushes back on the idea that special obligations arise from “discrete interactions” (Scheffler, 190) within relationships, like promises or mutual benefit, which might satisfy the voluntarist challenge. Most voluntarists are not against the idea of special obligations per se but against the idea of non-reductionistic (non-voluntary) special obligations.[5]  Scheffler poses the voluntarist challenge as one as special obligations being “unfair” and “burdensome” for those who did “nothing voluntary to incur them.” (Scheffler, 192) This “requires a version of reductionism with respect to such responsibilities [or special obligations].” (Scheffler, 195) Again, a reductionist view of special obligations may satisfy the voluntarist challenge since they would be reduced to natural duties, but Scheffler is a staunch non-reductionist.[6]

Burdensome non-voluntary special obligations again give rise to the voluntarist’s worry about autonomy. Scheffler frames the problem as one having to do with our “social identities.” (Scheffler, 203) This includes “the extent that we choose our roles and relations, and decide how much significance they shall have in our lives, we shape our own identities.” (Scheffler, 203) The autonomy worry is this: “if our relations to other people can generate responsibilities to those people independently of our choices, then to that extent, the significance of our social relations is not up to us to determine” (Scheffler, 203) The ability to determine our social identities seems like something we do not want to give up. If the burdens of non-voluntary special obligations undercuts our autonomy in this manner, then it may be an unreasonable view to hold.

Scheffler responds by saying that “it is clear that the capacity to determine one’s identity has its limits.” (204) Scheffler bites the bullet and says that it is just a brute fact that we cannot have absolute autonomy. Many factors which are integral in shaping our identities are not our choice. Just imagine being born in a different geographical area, with a different socio-economic status, and a different social milieu – it is not hard to imagine that we would be entirely different people with different ideologies, histories, or whatever constitutes our identities. Moreover, Scheffler responds, natural duties “apply to us whether or not we have agreed to them.” (Scheffler, 202) Voluntarists have been vociferous about special obligations being voluntary else they be reduced to natural duties, yet natural duties are not themselves voluntary either. They seem to take natural duties as a primitive truth, and surely any justification given for natural duties can be paralleled as a justification for special obligations. The reductionist approach simply appeals to the non-voluntary nature of natural duties, but what specifically justifies the non-voluntary nature of natural duties? Why have natural duties as a starting point and put the onus on special obligations? After all, says Scheffler, “No claims at all arise from relations that are degrading or demeaning, or which serve to undermine rather than to enhance human flourishing.” (Scheffler, 205)

R. J. Wallace is equally unconvinced by the reductionist approach (to special obligations) to meet the voluntarist challenge. Indeed, it is possible to reduce special obligations to natural duties of “reliance and trust,” (Wallace, 177) “vulnerability,” (Wallace, 178) or “reciprocity and gratitude,” (Wallace, 178) but this may not be the best explanation. Wallace is convinced by Sheffler’s thought that “we could have complete voluntary control over all important features of our identity and social connections.” (181)[7]

Wallace frames the worry about autonomy as one about “control.” (Wallace, 180) Wallace agrees with Scheffler that “complete voluntary control” might be a fantasy, but we still want “to be able to define for oneself the terms of one’s relationships to those whom one loves.” Wallace is a little more sympathetic to the autonomy worry than Scheffler. “If relationships of love are partly constituted by sui generis requirements, it would appear that there is little scope for the adult participants in them to negotiate the terms of those relationships for themselves.” Wallace calls non-reductionistic obligations “sui generis” obligations; in other words, obligations of an entirely different kind than natural duties. A robust set of sui generis obligations (“requirements”) push a crude set of obligations on our relationships. The issue for Wallace is the autonomy to shape the nuances of the dynamics of our various relationships.

Wallace’s response to the argument from autonomy: “it is partly up to individuals to determine for themselves the exact contours of the obligations they fall under in so far as they participate with each other in relationships of love.” [8] (Wallace, 189) His contention is that having space to shape some of the “contours” of our obligations is sufficient to preserve autonomy. He continues: “Relationships of love have a reflexive aspect, being constituted in part by the conceptions that are shared by the parties to them; this gives those parties scope for autonomous self-determination within the context of relationships that are nevertheless sources of sui generis obligation.”[9] (Wallace, 189) Negotiation must be “a process of mutual accommodation,” (Wallace, 190) so escaping duties on a whim thereby is not possible. Nevertheless, the contouring is supposed to save autonomy.

Jeske has different way of meeting the autonomy worry. She pushes the intimacy criterion which suggests that “continued participation” (Jeske, 540) is a necessary condition for special obligation generating relationships. Such relationships are maintained through a “mutual character,” (Jeske, 541) where the relationship becomes a shared project (e.g. interests, goals, history, reciprocity), and such “genuine intimacy cannot be coerced.” (Jeske, 541) Since it cannot be coerced, and we are always consenting to the relationship, autonomy is preserved. [10]

Let us further clarify how the autonomy worry is being met with a distinction of voluntaristic reductionism and non-voluntaristic reductionism. The voluntaristic reductionist meets the autonomy worry by reducing special obligations to features which arise from voluntary interactions. This approach says that autonomy is preserved because there is always some sort of consent to relations which give rise to special obligations. This is similar to Jeske’s approach.[11] In contrast, there is non-voluntaristic reductionism (what Jeske calls the “deflationary response”) which says that special obligations can be collapsed into more general natural duties. The voluntarist accepts the non-voluntary nature of natural duties, so there is no worry with special obligations if they derive from natural duties.

Wallace approach is to pushes back on non-voluntaristic reductionism. “What theoretical economies are achieved by reducing duties of love to more generic moral obligations, given the standing of love relationships as independent domains of non-derivative normativity?” (Wallace, 188) He is committed to a non-reductive account of sui generis obligations, and he admits that such a view, on the face of it, has a worry with autonomy. Rather than ignoring it (like Scheffler), he deals with this worry by saying that our contouring is enough to settle our worries.

Something about Jeske’s view does not seem to reflect normative practice. Let us take a closer look at her “intimacy” criterion. According to Jeske, the terminating conditions for a relationship should also be based on intimacy. A lack of intimacy excuses us from special obligations. This seems save autonomy, but at the cost of undermining the force of special obligations since intimacy is a mutual practice. If one arbitrarily ceases in maintaining intimacy and flippantly drops a relationship, it might be bad that one drop a relationship that promotes flourishing and satisfies human social need, but it would ultimately be a permitted act (there is no natural or special duty against this). The nature of intimacy seems entirely arbitrary. This is also what makes it attractive – arbitrariness, being the arbiter of our choice, is a feature of autonomy. 

Moreover, Jeske seems unsympathetic to the pull of having relationships. There is some sense in which even harmful relationships satisfy our intrinsic need (as social creatures) for relationships. Her view of relationships seems slightly crude in that relationships are only of value if there is intimacy (i.e. reasons to continue the relationship). For instance, an abusive relationship would have absolutely no value on Jeske’s view. On the other hand, on Scheffler’s view, perhaps it has some value in the mere fact that it is a relationship (this is not to say that this value is not significantly outweighed by the abusive nature of this relationship). It seems like relationships are merely instrumental to promoting mutual flourishing and not itself a non-instrumental end.[12] Jeske’s account preserves autonomy at the cost of ignoring an essential part of human psychology (that we value relationships). This cost makes her view untenable.

Nevertheless, the take away from Jeske’s approach is that the force of the autonomy worry is a matter of consent to duties. Wallace’s suggestion, that shaping the contours of a relationship satisfies autonomy, does not address the main issue of burdensome obligations which were not consented to. I also think Wallace agrees that this is an issue, and that he thinks it wrong to ignore it as Scheffler does. Jeske deals with this bigger issue at the cost of ignoring normative practices (i.e. sui generis obligations); in contrast, Wallace captures these normative practices without addressing the bigger issue, the “real” force of the autonomy worry.

Before looking at a possible way to reconcile these issues, let us take a close look at the worry with autonomy. First, I think there is a challenge with the autonomy of personal projects. Sui generis obligations often get in the way of our own personal projects, but this seems like a less of an issue, and we can bite the bullet like Scheffler. After all, we would have to abandon natural duties too because an amoral outlook is much better for pursuing a projects. Second, I think there is a challenge with the autonomy of being a good moral agent. This might be the bigger concern because it is a part of our normative practice that we value being good moral agents, but if sui generis obligations ask us to go against this, we might feel it is a heavy burden to our autonomy. It might come down to a value theory conflict of the good of relationships and the good of being a good moral agent. A sufficient condition for autonomy is allowing one to choose between the goods necessary for flourishing.

This view of autonomy is fairly intuitive. When pressed to give a list of things one values the most, people will come up with a wide variety of things (e.g. cars, clothes, family, love, etc.). Nevertheless, I think they can be narrowed down to a few fundamental categories of goods. These are non-instrumental goods – for instance, chocolate might be an instrumental good insofar as achieving the non-instrumental good of pleasure. It is an open question what the exhaustive list of non-instrumental goods consists in, but two things that (I think) would appear on everybody’s list are relationships and virtue (i.e. being a good moral agent).[13] Such goods are essential for a good life and promoting Eudaimonia (i.e. flourishing). It is not clear how one ought to order these eudaimonistic goods; still, how one orders these goods affects (ipso facto) how one acts, and these choices culminate to what sort of life one lives. This autonomy to order one’s eudaimonistic goods is what must be protected.

Wallace’s account fails in this respect. Wallace follows Scheffler in saying that special obligations may be outweighed by natural duties, but this impartial calculus leaves no room to exercise autonomy. Let’s go back to the Euthyphro father’s example. Suppose an essential feature of the father-son relationship is that there is a sui generis obligation not to report one another if one commits murder. If natural duties outweigh this sui generis obligation, then he must turn his father in. This impartial calculus leaves no room for Euthyphro’s autonomy. What if he valued the goods of relationships and placed them ordinally higher than the goods of virtue? This seems burdensome and restricts his autonomy.

I nonetheless admit that certain smaller cases can be covered by Wallace’s contouring view. That is, the “weak” special obligations (negotiated governing normative conditions of a relationship) might also serve as terminating conditions when they remain neutral to the “strong” special obligations. For instance, perhaps turning one’s father in to the authorities violates “strong” special obligations, so suppose one does not turn him into the authorities. The father being a murderer may nevertheless violate “weak” special obligations in that there was some agreement made that they’d be morally upstanding characters. By violating the “weak” special obligations, there would be grounds for terminating the relationship. Still, this does not address the burden of going against natural duties to turn murderers into the authority; this, I think, is a significant challenge to autonomy. 

My suggestion is that we take a closer look at terminating conditions. The aim is a picture with sui generis obligations which also manages to give us sufficient autonomy. Perhaps even when the calculus obviously weighs on the side of natural duties (Euthyphro turning his father in), we may permissibly choose to side with sui generis obligations. Our justification would be on the basis placing higher value on the eudaimonistic value of relationships (rather than virtue). After all, our choices should be construed teleologically in terms of eudaimonstic goods. Here, we have the freedom to order our eudaimonstic goods and act accordingly.

On this picture, it was Euthyphro’s choice to turn in his father. I think Scheffler’s approach is right to include non-reductive sui generis obligations; however, I think Wallace is right in his sympathies to the voluntarist’s worry about autonomy, but his approach was not sufficient to save autonomy. Jeske’s approach dealt with the worry in the right kind of way, but at the cost of ignoring an essential part of human psychology – namely, the need for relations. My suggestion of terminating conditions proposes a way of sufficiently preserving autonomy while acknowledging the features of human psychology. It takes a eudaimonistic approach and argues that the core of autonomy is the ability to choose how one lives a good life, which stems from the ability to choose eudaimonistic goods. I further suggest that the conflicts with special obligations and natural duties can be framed within a value theory context; in essence, a conflict between the ordering of the eudaimonstic value of relationships and the value of virtue. By and large, this is one non-reductionistic reply to the voluntarist challenge.

Sources

Diane Jeske, “Families, Friends, and Special Obligations”

Samuel Scheffler, “Relationships and Responsibilities”

A. John Simmons,”External Justifications and Institutional Roles”

R. J. Wallace, “The Duties of Love”

---

[1] The distinction between “duties” and “obligations” is a subtle and important one, but it has no significance in this paper and can be used interchangeably.

[2] Friendship relations do not have this problem and meet the voluntarist challenge; according to Jeske, they are voluntary “mutual project” (539)

[3] As outlined by Scheffler’s fifth condition. (Scheffler, 199)

[4] This paper will henceforth use this as the model for non-reductionism.

[5] Generally, special obligations arising from promise relations satisfy the voluntarist challenge.

[6] A second issue (neutral to the voluntarist challenge) to special obligations is that it may “confer unfair advantages on their bearers.” (Scheffler, 192) This goes back to the example of not turning your murderous father in to the authorities, and your father doing the same for you. It also hints at the tension between natural duties and special obligations, especially if natural duties ask us to be impartial.

[7] Wallace further adds that “reductionism promises to liberate the realm of loving relationships from the shackles of normative constraints.” (Wallace, 181) If we have certain natural duties already (like trust, vulnerability, or reciprocity) then adding more duties (non-reductionistic sui generis duties) seem to be excessive. He mentions the “phenomenology of agency within the relationship of love,” (Wallace, 182) suggesting our experience of love is such that our actions “out of a sense of duty or obligation.” (Wallace, 182) This would result in an “excessively ‘moralized’ conception of relationships of love.” (Wallace, 182)

[8] E.g. sexual fidelity between life partners

[9] Sui generis obligations meaning obligations arising from (and which we can appeal to) the “specialness” of certain relationships.

[10] It is important to note that this is a diachronic model for continuing conditions; in other words, intimacy is not “chosen in a single discrete act,” (Jeske, 541) like a promise or contract. As such, special obligations which we did not consent to, like some familial relations, can be terminated on the grounds of intimacy, thus further preserving autonomy.

[11] Although, she does not in principle have to deny sui generis special obligations.

[12] Perhaps there is a way to interpret Jeske’s intimacy view as sui generis, but this approach seems rather ad hoc. Perhaps she would be satisfied with Wallace’s model, but this is an open question

[13] Other things might include knowledge, pleasure, or fulfillment, but the idea is that there is some general category of eudaimonistic goods. This is an expansion of Scheffler’s thought that we all value relationships.