Are groups of individuals agents? We can say that nations go to war, companies implement their plans, teams lose or win games, and so on. But do groups really act? Or is it that only individuals do, and that their aggregate actions are attributed to groups only figuratively? That is, when we say that such and such football team won a game, what we really mean is that certain individuals coordinated their actions to follow a plan each of them agreed upon. A different understanding of group agency points out that individuals perform actions as members of a group only because they stand in a certain relation to one another by virtue of their membership; hence, in some sense, the existence of the group transcends the aggregate existence of its members. To choose between the two views of group agency, one needs to determine whether groups can be said to exist in relative independence from their members.
An issue related to this inquiry is whether groups can be moral agents – agents capable of making choices and exercising freedom and thus capable of being held accountable for their actions. Group agency is subject to a lively academic debate with important practical implications crucial for global justice. In order to identify the norms of moral order that underlie global justice and assign rights and duties correctly, we need to know which among various collectives present in the international arena, such as religious, ethnic and national groups, states, militant organizations, parties, NGOs, corporations, and others, can be recognized as moral agents in their own right with rights and responsibilities. For example, states in which governments do not express the will of the people certainly act as group agents, but their political power is not legitimate. Or, the official expression of group interests of an oppressive cultural group may not represent the true interests of all group members. To see what the group may be entitled to, we need to investigate its mode of existence. A theory of group agency provides a much needed background for this inquiry. Thus, the relevance of the notion of group agency to global justice lies in the notion’s potential to provide a substantive basis for correctly identifying subjects for normative treatment, in particular, moral group rights.
There are different accounts of what sets of individuals constitute collective agents and of what evidence can be given in support of their existence. The presence of an institutional organization and a decision procedure makes a collective capable of purposeful action over time and gives it identity over time independent of particular membership, as Peter French would point out. For example, corporations can be full moral persons with rights and responsibilities; because individuals within the corporation act according to the status within the group’s power structure, the structure incorporates individual intentions by subordinating and synthesizing them into a corporate decision (French 1979).
A less formal structure may suffice for a group agent to come into existence. One may say, with Christopher McMachon, that a group of cooperatively disposed people that lacks an institutional structure but has made the choice of a cooperative scheme becomes a group agent (McMachon 2001). The least demanding view of group agency requires neither formal institutional nor procedural organization to say that a group is an agent. Groups that lack explicit decision-making structures but have enough cohesion to engage in collective action can be held responsible as groups. For example, family members are motivated to act for the good of other members of their group (Feinberg 1968). Hence, we may say that a group exists when its members partake in a cooperative relation that extends through time and influences their actions. The presence of some or all the following features creates a group agent: a formal institutional organization; an established decision-making procedure or cooperative scheme; and a certain relation that enables individuals to engage in group actions based on a common interest. Those who admit that groups exist in relative independence fromtheirmembers take a realist stance with respect to group agency, and can be called “non-reductionists” – that is, what a group is cannot be reduced to the sum of characteristics of its individual members (note that on this account members standing in a certain relation to one another are more than just a sum of individuals, they are individuals plus the relation). Those who object to realism concerning groups can be called “reductionists” – they claim that a group can be reduced to the aggregate of its individual members. Hence, realists need to offer some proof of the existence of group agency. Philip Pettit provides evidence that collective reasoning yields results different from the summation of the results of individual reasoning: when individuals decide as a group, the outcome is often different from the one obtained by majority vote based on the outcomes of individual decisions (Pettit 2001). This proves that collective agents are discontinuous with the individuals who compose them. David Copp offers a normative argument for the existence of group agents: in certain situations the actions of individuals in their official capacity on behalf of collectives can be rational and morally innocent while the outcome of their actions is morally faulty.
Since moral fault has to be assigned and individuals are blameless, the fault resides at the collective level. The collective is at fault and thus must have acted; therefore, it is an agent. Reductionists can object that groups cannot be agents because they do not have self-awareness as individuals do. One may reply that although individuals have first-person phenomenological access that collectives lack, this access does not need to be required for agency. Carol Rovane argues that persons, individual or collective, endure over time only insofar as they have commitments to unifying projects that give them renewed reasons for action at different moments of their lives (Rovane 1998). A group person is constituted by interpersonal affairs and a single person is constituted by intrapersonal affairs that pertain to a unifying project. Group members cannot control one another’s actions directly, but neither can they do it in their relations to their future selves. To act today on an intention one had yesterday, one has to first remember that she had this intention. By recognizing the existence of the intention, she recognizes that she participates in a certain continuous project.
A similar action of recognition occurs in interpersonal affairs in which individual interactions contribute to the execution of the unifying project. Hence, that the collectives do not have a self as individuals do doesn’t preclude them from being agents. Consequently, collectives can be held morally responsible based on the kinds of projects they engage in and can have rights. Overall, at present the question of group agency offers more questions than definitive answers, but it is of crucial importance for global justice to continue developing theories of group agency.