The philosophical literature on collective agency, collective responsibility, social epistemology, and social ontology is burgeoning.
Scholarly interest in ‘collective’ phenomena and theories reflects a persistent desire to tackle an old conundrum — the relationship between individual agents and the collectives they compose.
The problem of reconciling these two perspectives is not one for philosophy alone.1 There is a notable shift away from a strict individualism towards theories that reflect the fundamentally social, cooperative nature of human activities.Philosophical enquiry and theorizing has in many ways traditionally been too focused on the individual: the ideal of the autonomous, well-informed, moral agent has dominated ethics, philosophy of action and epistemology for a long time (and arguably still does). In contrast, in political philosophy the question of an inwardly and outwardly just society is often raised without spending enough time exploring what kind of ‘collective’ a society is and how to understand the relationship between this collective and the individuals composing it.
This chapter attempts to shed light on the issue of collective moral obligations, that is, obligations that individuals in loose groups (as opposed to group agents) may have together.2 I take moral obligations to be a basic feature of our moral repertoire. Put in the simplest possible terms, there are things we ought and things we ought not to do.3 For instance, we ought not to harm others without good reason and we ought to assist those in need. Further, we might want to distinguish between pro tanto and all-out obligations. Pro tanto obligations are demands on us that give us some reason to act, but which can be overridden by other, more important reasons. All-out obligations are those we ought to meet all-things-considered, taking all the different moral considerations into account.
My contention in this chapter is that our moral obligations (pro tanto and all-out) can sometimes be collective in nature.
By this I mean that moral obligations can jointly attach to two or more agents in that neither agent has that obligation on their own, but they — in some sense — share it or have it in common.4 I will explain this in more detail below.I believe that the notion of collective moral obligations fills a conceptual gap in philosophy. In a sense, one could say that moral philosophy has traditionally been concerned with the question ‘what ought I to do?,’5 while what we ought to do as communities has been the focus of political philosophy. But some of the things that we together can (and potentially ought to) do may be neither the political communities’ responsibility nor straightforward individual obligations. Furthermore, even where some desirable goal or action is primarily the political community’s responsibility, as a matter of fact, political agents (e.g. states and their institutions) often fall short of meeting their obligations. In either case, there may be groups of individual agents that can step up and produce the desired outcomes or perform the required actions. Examples include our joint ability to overcome collective action problems even in the absence of state action, such as closing (or reducing) the so-called global emissions gap (Blok, Hohne, van der Leun, & Harrison, 2012; Wynes & Nicholas, 2017). On a national scale, our ability to produce herd immunity against some infectious disease should be seen as collective. I will come back to these examples at the end of the chapter. My starting point will be a simpler, small-scale, real-world example of spontaneous collaboration between complete strangers:
Ten passers-by witness a car accident in which a motorcyclist gets trapped underneath a car, which has caught fire on one side. Somebody has to act very quickly to pull him out the other side and in order to do so the car will need to be ever so slightly lifted. None of the passers-by can lift the car on their own and pull the man out, but together they can (without taking any undue risks to their own health and safety).
As it happened, the people manage to lift the car and save the motorcyclist’s life.6For the sake of argument, let us assume that the following is the case: it is obvious to the witnesses of the accident that the man is in imminent danger and it is fairly clear what needs to be done to get him out of danger.
There are several scholars who argue that under circumstances such as these individual moral agents can be under a collective obligation (or have collective responsibility) to assist (Held, 1970; Isaacs, 2011; May, 1992; Miller, 2010; Pinkert, 2014; Schwenkenbecher, 2013, 2014, 2019; Wringe, 2005, 2010, 2016). Collective obligations, on their accounts, are distinct from and not reducible to individual obligations (to contribute to cooperative ventures, for instance).
In the following, I will distinguish different ways of spelling out such collective obligations. But before I do so, let me briefly talk about why anyone might think that we need the notion of collective obligations. One of the starting points of many debates on collective obligations is the observation that in cases like the one above, in order to produce the morally best outcome, or in order to perform the action most likely to secure that outcome, individual agents need to cooperate with one another and coordinate their individual actions. It takes more than one person’s effort to make a difference to the person in need.
More generally, there is a class of actions (and outcomes) that cannot be performed (or produced) by one person on their own. They require at least two people in order to be realized and no one individual agent can guarantee the success of the collective endeavor. These cases are characterized by ‘joint necessity.’7 Playing a duet is a joint necessity type of activity. By definition, it cannot be done by one person. Another example is ‘talking past one another.’ Joint necessity can be analytic (as in the two examples just given), where it is part of what it means to do x that x is done by at least two people.
Or joint necessity can be circumstantial, where as a matter of fact (rather than as a matter of principle), something cannot be done by one person, for instance, if it takes two or more people to lift a heavy table (or a car, for that matter).We can further distinguish between strict and wide joint necessity.8 For strict joint necessity to apply, the number of available contributors to a collective outcome equals the number of contributors minimally necessary to produce it. What it means to be an available contributor would depend on the outcome in question. For the motorbike accident described above, it would mean anyone close enough to see what is happening and able to make some kind of contribution. For strict joint necessity, the success of the joint venture is counterfactually dependent on each available contributor playing their part. It is entirely within my power to stymie any efforts of our duet playing and the same applies to you.
Wide joint necessity applies where there are more available contributors to a joint outcome than minimally necessary. There are many large-scale examples of wide joint necessity such as producing herd immunity (against a certain infectious disease), bridging the emissions gap (UNEP, 2017), or producing a referendum outcome in favor of marriage equality. In order for herd immunity against a particular communicable disease to be achieved it is not necessary that everyone who can safely be vaccinated be in fact vaccinated. Depending on the disease, the figure may be around 90 percent. What this means is that, in contrast to strict joint necessity cases, my unilateral defection in a wide joint necessity case does not guarantee collective failure and neither does yours. This might lead someone to the conclusion that therefore individual obligations to contribute to such goods are always less stringent. But I think this would be the wrong conclusion to draw, as I shall show below.
In many joint necessity scenarios, something morally important is at stake.
Especially where lives are in imminent danger, people tend to share the intuition that those who could help ought to do so, e.g. those witnessing the motorbike accident ought to assist the trapped person. But this common intuition may create a dilemma, because, no individual can guarantee the success of the joint endeavor (or produce the desired collective good).9 That is, individually, none of these passers-by can assist the trapped motorcyclist. Hence, it cannot be any individual agent’s obligation to rescue him. They can only help jointly. So, whose obligation is it? We might want to say that it is the obligation of all of them together. But what does that mean? Is it the ‘group’ of passers-by that has the obligation to assist? Or is there ‘merely’ an obligation on each of us to do our best given others’ actions? The answer is not straightforward. In my view, this impasse is regularly felt when we try to make moral decisions: it is the pull between the individualist option (to do something that is under one’s individual control only) and the collectivist option (where the success of one’s actions often depends on others’ contributions).Scholars have chosen different routes to answer the question about the locus of moral obligation in joint necessity cases. Roughly, they can be divided into two groups, which I will call ‘conservatives’ and ‘revisionists.’ Revisionist scholars will usually introduce new moral vocabulary and concepts to fill what they believe to be a gap in traditional moral theorizing where joint necessity is concerned. Many argue that there is some kind of group-level obligation (or responsibility) that applies to loose collections of individuals such as the passers-by in scenarios like our exemplary case (Held, 1970; Isaacs, 2011; May, 1992; Wringe, 2010). Other revisionists, including myself, speak of individuals holding joint obligations (Miller, 2010; Pinkert, 2014; Schwenkenbecher, 2013, 2014, 2019) or sharing obligations (Bjornsson, 2014).
Conservative scholars, in contrast, do not see the need for new conceptual tools, but attempt to resolve collective action puzzles in a way that is maximally continuous with existing theory. They tend to argue that joint necessity cases give rise to (perhaps slightly more complex than usual) contributory duties only. According to Parfit, for instance, each of the individual passers-by simply has an obligation to contribute if she thinks that enough others contribute to get the joint endeavor off the ground (Parfit, 1984). Collins and Lawford Smith would argue that each ought to take steps towards forming a group that can then act as an agent (Collins, 2013; Lawford-Smith, 2015).I will not discuss the merits of these different types of approaches in any detail here, since I have done so elsewhere (Schwenkenbecher, 2018). The obvious downside of the conservative approach is that the obligation to produce the collective good (or to realize the joint endeavor) is not allocated. In our example, then, there is no obligation to free the trapped motorcyclist, even though individual agents have obligations to contribute.10 Holly Lawford-Smith acknowledges this problem for the conservative approach, but bites the bullet because she thinks the advantages of this approach still outweigh its disadvantages (2015). In contrast, the obvious downside of the first type of revisionist approach is this: it seems to be built on the assumption that there is a (novel) entity, a group agent of sorts, that can not only act on the problem at hand, but has a sufficient level of unity such that it can hold a moral obligation (or be held morally responsible). Revisionist scholars have tried to avoid this kind of criticism by arguing that being an agent is not a necessary condition for being the bearer of a moral obligation as far as groups (or collections of agents) are concerned (Wringe, 2010).
However, my aim here is not to give an overview of the literature but to instead flesh out my own (revisionist) approach to collective obligations and show how it applies in a range of cases.11 This approach, while revisionist, avoids the objection sketched above by refraining from postulating a group agent (even a putative or potential one) and instead conceiving of the obligations to assist in joint necessity cases as shared or joint.’
18.1