What Makes These Things Obligations?
Even if an account of this sort captures our intuitive attributions of group obligations, it leaves the question of why the fact that a group would φ if members cared appropriately would constitute an obligation on the part of the group to φ.
This question is particularly pressing as this account, like all the proposals in this family, seem to suggest that what looks like a straightforward attribution of an obligation to φ to a group is in fact something quite different and much more complex, such as an attribution of individual obligations to members of the group to have a certain disposition such that if all members had that disposition they would φ.Elsewhere, I have suggested that the relation postulated in the caring proposal constitutes obligation, because it is the very same relation that constitutes individual obligation, i.e., an obligation of a group of only one individual. If this is correct, there are no hidden structural differences other than the number of obligation bearers. To give a sense of this sort of reply, I will briefly recount the proposal in question and explain how it applies to the individual case, and why the relation constitutes obligation. This is the proposal:
obligation: A group has a moral obligation to φ if and only if the group’s φ-ing (i) is morally important and (ii) would be ensured, in a normal way, if members cared as can be morally demanded of them (Bjornsson 2014; Forthcoming; cf. Bjornsson and Brulde 2017).
As before, the concern here is with all-things-considered, not pro tanto obligations. The moral importance of the group’s φ-ing can be instrumental or non-instrumental. On some views, the only relevant moral importance is the importance of not acting on certain kinds of motives, but most acknowledge the moral importance of bringing about or not preventing certain kinds of actions and outcomes, regardless of specific motives.
Substantive accounts of obligations can differ with respect to what is morally important and what makes it so, but also with respect to what it can be morally demanded that people care about, and to what extent. On some narrow Kantian views, morality only requires that we care about acting from maxims that can be willed as universal law, and on some narrow utilitarian views, it only requires that we care about the surplus of pleasure over pain. But most will acknowledge demands to care about a variety of values, including staples of both deontological and consequentialist moral theories: telling the truth, keeping promises, not using others as mere means, human flourishing, happiness, relationships, knowledge, and so forth. Because obligation leaves open what can be morally demanded of members of groups, it is compatible with a variety of normative views, and because it leaves open what ultimately grounds such demands, it is also compatible with a variety of views about their nature and ultimate sources, including intuitionism, rationalism, contractualism, and constitutivism.6Implicitly, obligation accounts for the common idea that an obligation to φ presupposes an ability to φ: if all members of the group care about what makes φ-ing bad, this would ensure, in a normal way, that they would φ.7 What, though, is it for sufficient caring to ensure an outcome, and to ensure it in a normal way?
For X to ensure Y is for X to rule out all relevant possibilities alternative to Y, not necessarily for X to cause Y The lock on the door might ensure that no one enters even if no one tries to enter, and if the passengers on the ferry care about staying afloat, this might ensure that they don’t capsize the ferry even if the reason that the ferry remains steady is they are all sitting down to rest after a hard day’s work. For obligations that can only be discharged through coordinated action, sufficient caring will often ensure such action by prompting intentions to join forces if others are similarly prepared, and act on the perception of such preparedness among other members.
Even here, though, individuals might participate out of habit, group conformity, or economic interest, and morally required caring kick in only if the joint endeavor is threatened.8As noted, obligation requires that the way in which appropriate caring ensures the group’s φ-ing is relevantly normal: a matter of skill or ability rather than mere luck. This might require that members’ appropriate caring would block at least some relevant threats to the group’s φ-ing by at least marginally coordinated behavior, rather than by individual behavior completely insensitive to that of other members. It also rules out cases where group actions would ensure φ-ing only in ways deviating significantly from the intended:
Serendipity: The adults’ only chance of saving all the drowning children is by using a lifeboat mounted on a support structure. But an arm holding the boat in place is rusted stuck and it seems that the only way of releasing it is by hitting that arm with an iron rod lying nearby. Unfortunately, as it happens, doing so would fail to release the lifeboat. Fortunately, it would make a loud sound that would alert a coastguard boat that is anchored behind a rock and would come and rescue the children.
Suppose that it is the group’s duty to try to save the child, and to try to get the lifeboat unstuck using the iron rod, and that these are the group’s duties because of the importance of the child’s survival. Still, if the only way that they would bring about the child’s survival would be by accident, they do not plausibly have an obligation to save the child in this case. By contrast, if they had known about the nearby coastguard boat and would have employed the iron rod to catch their attention, it would have been their obligation to save the child by doing so.
A final aspect of obligation worth mentioning is the connection between obligations and blameworthiness. If a group fails to discharge an obligation, it must be because one or more members’ caring fell short at some point.
But if something morally bad happens as a result of people not caring about the values at stake as can be demanded of them, they are plausibly morally to blame for it: at least that seemed to be the case in Helplessness. (We return to the issue of collective blameworthiness in the next section.)The claim now is that obligation is an equally plausible account of obligations of individuals, understood as “groups” of one agent. The demands on caring are the same, and what is of moral importance is the same. What is different is merely that it is the appropriate caring of a single agent that ensures the φ-ing of that one agent. Moreover, the account applies as straightforwardly to the case where one agent can either save her own child or use a lifeboat to save all children, to individual obligations to φ that are discharged without the aim of φ-ing, and to a one-person version of Serendipity. But if obligation is an adequate account of both individual and group obligations, it is clear why group obligations to φ are obligations: they are constituted by the very same relation between the obligation bearer and the φ-ing as are individual obligations.
9.6