May On Shared Responsibility

Collective responsibility as a general term associates moral agency with group agents rather than individual agents. It refers to the causal responsibility and blameworthiness of groups, where moral responsibility is located in the group, as a collective, rather than in the individual members of the group.

Shared responsibility is a particular form of collective responsibility. The term refers to the concept of individual group members each being held partially responsible for harm that is caused by the group’s actions. We can think of the difference between group and shared respon­sibility in terms of distribution: group responsibility applies blame to the group as a whole, while shared responsibility applies blame to individual members of the group. Even if we accept the possibility of shared responsibility as collective responsibility, we might wonder if it is pos­sible to fairly distribute blame among the individual members of a group. And if it is possible, when and how could shared responsibility exist?

Larry May provided an early account of shared responsibility in which he defends the pos­sibility of such a distribution. In his 1992 book Sharing Responsibility he constructs an account of shared responsibility based on his view that the identity of individual moral agents is largely shaped by their communities. May acknowledges the connection between one’s attitudes and one’s behavior, but he does not see our attitudes as something over which we have sole control.

For May, the concept of shared responsibility allows us to expand the traditional concept of individual responsibility to acknowledge the social influence on our attitudes and the social nature of many acts. Accordingly, shared responsibility provides the tools for holding individuals responsible for things that were not directly caused by the individual. May’s view of individual responsibility as something that is both unique to an individual and inextricably tied to the col­lective also grounds his concept of shared responsibility, which is meant to “divide responsibility among the members of a group, rather than to hold each member fully responsible or to hold the whole group collectively responsible” (May 1992: 37—8).

An attribution of collective responsibility relies on the identification of something over and above the individual contributions of each member of a group; but, since shared responsibility is about the aggregation of individual responsibility, May argues that it can be used to under­stand responsibility within groups that are less cohesive. Thus, he sees his account as particularly useful for situations in which a person made a “choice that contributed to a harm for which that person was then at least partially responsible” (May 1992: 36).

There are several types of situations in which the need for the concept of shared responsi­bility arises, according to May. The first situation is when an individual makes a choice to con­spire with another to engage in joint action. The fact that a choice was made by the individual is what warrants some amount of individual responsibility, but the fact that the individual’s actions were not sufficient to cause the harm renders full responsibility inappropriate. Second, shared responsibility is appropriate when an individual takes a voluntary action, and this action is a necessary part of the harm, but the individual does not have any knowledge or intent of contributing to the harm that occurs. Again, it is the choice and its necessary contribution to the harm that warrants partial responsibility for the individual. Finally, and most controversially, May argues that there are situations in which an individual can be held partially responsible for choosing to maintain particular attitudes that increase the risk of harmful behavior on the part of others.

In each of these situations, there is more than one person who contributes to a harm. In the first scenario, there is a somewhat clearly constituted group, insofar as the two or more individuals have agreed to work together to facilitate the harm that occurs. In the second and third scenarios, however, the “group” is less cohesive. Some claim that a group cannot exist without a formal institutional structure, shared group intentions, and a commitment of the individual members of the group to the entity of the group as a whole (see, e.g., Gilbert 1996; 2014).

May distinguishes between a corporation as an example of a group with significant organ­izational structure,¹ and a mob as an example of a group with little (if any) organizational struc­ture, but which has come together for economic or political reasons. A mob, however, provides an opportunity for individuals to engage in actions that they would not be able to on their own, and May argues that social identification can play a role in creating shared attitudes and behaviors. He notes that:

[g]roup members influence one another by a certain kind of socialization, often not fully recognized by the group members, in which the values of the group member tend to be reshaped so as to conform to the norms of the majority of the members of the group. Mob members influence each other in such a way that people are encouraged to act in pursuit of goals and in ways they would never have allowed themselves if they had been uninfluenced by fellow mob members.

(May 1992: 78)

Take a group of individuals who have just witnessed an incident of unwarranted police bru­tality against a young man, a scenario I will refer to as Mob. In Mob, the individuals are outraged by the same event, and due to their shared outrage, begin yelling at the police officer and even­tually flip over his police car. If a single individual had witnessed the act of police brutality, she would have been equally outraged, but she may not have been encouraged (or emboldened) by the presence of others to speak out against the police officer, and she certainly would not have been able to flip over his car.

But what about a putative group that is even less cohesive, one that is created out of cir­cumstance and not shared interests? May uses a common example of a drowning child and a collection of bystanders on the shore (I’ll call this coordinated bystander scenario Beach) to demonstrate what he means by “capacity for joint action.” In the example, none of the indi­viduals who hear the drowning child are capable of saving her on their own, but if they fail to coordinate a rescue attempt, “it may be implausible to say that their failure to act constitutes collective inaction” because they do not necessarily have the time to coordinate amongst them­selves (May 1992: 111). May contrasts this example of a drowning emergency with a famine (which I will refer to as Famine), a situation in which there is more time for a putative group of individuals (or institutions) to develop the mechanisms necessary to coordinate group action. On May’s view, the putative group in Famine could be held collectively responsible for failing to coordinate and failing to prevent harm, as well as individually responsible for failing to prevent harm. Shared responsibility, however, does not rely on the existence of a group, since it is based on the aggregation of individual responsibility.

Mob, Beach, and Famine bring up another important feature of shared responsibility, which is the manner of dividing responsibility. May explains how this happens as follows:

Dividing responsibility for a harm is also different from assigning to each of several people full responsibility for a harm. Some or all members of a group may be assigned less than full responsibility for a harm in cases of divided or shared responsibility.

(May 1992: 38)

Individual responsibility is to be shared among the various actors, and is to be divided fairly, but not necessarily equally. May notes that “some persons in [a] collective may be able, because of their various leadership skills, to be more effective in bringing these people to form a group able to act intentionally” (May 1992: 111). He defends a view of distributing responsibility that mirrors his ontological view of groups, which acknowledges an individual’s choices, as well as the influence and potential that comes from the group. A purely individualist account of respon­sibility would focus solely on the individual’s intentional decisions, and it wouldn’t matter to the assessment of responsibility whether or not there were other actors or a putative group. May’s view, on the other hand, takes into account that an individual may be influenced by others, and “thus it sometimes makes sense to view the individual’s responsibility in a group as different from what it would be if there were no group” (May 1992: 112—13).

In Mob, it may be that one particular individual suggested that the group work together to flip over the police officer’s car. Each individual member of the mob would be assigned some level of responsibility for the damage done to the police officer’s car,² and the ringleader would receive a greater distribution of responsibility than the individual who stood back and cheered while the others flipped the car. On the shore in Beach, a CEO is likely more responsible than a teenager for failing to coordinate the pair of former Olympic swimmers to work together to save the drowning child. And in Famine, it may be that relief organizations with experience responding to large-scale crises receive more blame for the continued harm than a new nation­state with few resources.

It is important to note that in both Beach and Famine, it is collective inaction, rather than action, that results in harm. For the remainder of this chapter, I will focus on what seems to be the most crucial contribution of May’s view of shared responsibility, which is his defense of individual responsibility for collective inaction. Some thinkers argue that while we can be held responsible for harm that we have intentionally caused, we may not be responsible for intervening where we are not the cause of harm (see, e.g., Kamm 1986; Sartorio 2005). As May argues, “[i]naction leads to serious harm in the world, just as certainly as intentional, active wrongdoing. Yet inaction, especially collective inaction, presents difficult problems for theories of responsibility” (May 1992: 105—6). But assuming we can make a normative claim that there are situations in which individuals or institutions are obligated to act to prevent harm, we must be able to identify which failures to act should be singled out as warranting blame, and how the responsibility for failures to act should be distributed among individual members of a group (or putative group).

As shown in Beach and Famine above, assessing shared responsibility for collective inaction requires an analysis of whether the collective had the capacity to act as a group. As May notes, assessment of this counterfactual, “Could the collection of people have avoided inaction?” proves much more challenging than an assessment of action, especially in the case of putative groups with no history of collective decision-making (May 1992: 109). If the putative group could not have reasonably been expected to act as a group, then a “collective” failure to act would only be a failure of each individual to act. But if the individuals could have acted other­wise, then they would have been individually responsible for harms resulting from the failure to constitute themselves as a group, and corresponding inaction. The kinds of omissions that a group might be responsible for include those harms that the group has done something to con­tribute to already, those harms about which they have created expectations that they would act to prevent, and those harms that they are uniquely situated to prevent. As for putative groups, May largely relies on a determination of what the rest of society sees as reasonable to expect, in light of existing norms.

Individual responsibility for action is more easily distributed, based on actual behavior, including acts of leadership or persuasion, as shown by the ringleader in Mob. Both individual and collective responsibility for inaction, according to May, is focused on capacity. Thus, an individual who could have used leadership skills to prevent harm should have a greater share of responsibility than those who were only capable of following orders. Yet May does not see assessment of moral responsibility for a harm as an endeavor to distribute a fixed number of “equal shares based on the number of people present in the group” (May 1992: 113). Rather, he claims that individuals should be required to do as much as they are capable of doing, which is based on an assessment of an individual’s capacity, and which does not change in relation to the capacities of the other individuals in the putative group. This method of distribution also allows for a distinction between an individual who directly causes harm and individuals who fail to act, while still attributing some amount of blame to the indirect actors.

A final aspect of May’s view of shared responsibility and collective inaction worth considering is how to hold individuals responsible for inaction. May claims that another diffe­rence between shared responsibility for collective action and shared responsibility for collective inaction is the corresponding moral weight. For May, the partial, shared responsibility is of a different kind than the guilt associated with full responsibility for directly causing harm. He suggests that shame is a more appropriate tool than guilt, since it is “directly related to a person’s conception of herself or himself, rather than to explicit behavior (which is what guilt most commonly attaches to)” (May 1992: 120). His claim seems to be both descriptive and normative, as May sees an individual’s failure to act as part of a collective as something that generates the moral feeling of shame, rather than guilt, because it relates to the association between one’s own identity and the identity of the group. I will return to this aspect of May’s argument in the next section, where I explore criticisms of May’s account of shared responsi­bility for collective inaction.

15.2