Reduction and Individualism

As I have said, I take it to be common ground that an institutional actional event is constituted by actions performed by individual agents. There are different ways to understand this.

I want to remain neutral about the grounding view, but agency individualists would reject it on the basis that, they hold, actions of institutional and other collective entities are not “over and above” the actions of relevant individuals. Accordingly, they offer a different interpretation of the thesis that collective entities are not independent agents. They understand it as a claim about reduction — a conceptual or semantic thesis to the effect that our concepts, or the meanings of our terms, ensure that, necessarily, any proposition about an action of a collective entity has exactly the same truth conditions as some corresponding proposition about actions of relevant individuals — where the latter does not contain any term referring to or quantifying over an action of a collective. Call this the “reductionist thesis.”²² The extreme individualist takes the reductionist thesis to be a basis for denying that collective entities are agents.

Kirk Ludwig is an example of an extreme agency individualist, and he has defended the reductionist thesis in detail (2016 and 2017). He says, “all collective action is a matter of indi­vidual agents contributing to bringing about some event or state.” And he seems to conclude on this basis that “we have no need for group agents in our understanding of ordinary discourse about collective actions, either in the case of informal groups or in the case of organizations” (Ludwig 2020: 79). But even if the reductionist thesis is correct, it does not follow that col­lective entities do not act and are not agents. Facts about bachelors can be reduced to facts about unmarried adult males. It would be a mistake to conclude from this that no person has ever been a bachelor. Reduction is not the same as elimination.

I believe that the thesis that institutional entities are capable of intentional action follows from the reductionist thesis when it is properly understood. The reductionist thesis provides non-mysterious truth conditions for propositions about collective action. It implies that, neces­sarily, propositions about the actions of institutional entities have the same truth conditions as corresponding propositions about the actions of relevant individuals. There is no reason to think that the truth conditions of the latter propositions cannot be satisfied. Individual persons can do things in their capacities as members of collective entities to bring about institutional actional events. When they do, it follows from the reductionist thesis that these collective entities act.

Ludwig claims that, given the reductionist thesis, “we have no need for group agents” (Ludwig 2020: 79). He might here be supposing that, given the reductionist thesis, it would be ontologically profligate to suppose there are collective agents.

Furthermore, as I have argued before, we can often best understand the moral situation of a person — the requirements she faces and her responsibility or lack of it for things that have happened — by reference to the moral situation of a collective entity she belongs to (Copp 2007a; Copp 2010; Copp 2012; see Lyons 2004; Wringe 2014a). In particular, the transitional duty can explain why a person who is a member of a collective entity has a moral obligation. The transitional duty is the requirement, if a collective entity is required to do something, that its members, and any relevant officers, do their part in bringing it about that the collective acts as required. It presupposes that collective entities do have moral obligations, which means that it presupposes that collective entities can be agents and can perform actions. On Ludwig’s view, the transitional duty is empty, since no collective entity is an agent. So on his view there is a puzzle. Why is it that people have duties in their roles as members or officers of collectives? What is it about their roles that gives them obligations if not that they are obligated to do their parts in bringing it about that the collective acts as required?

The moderate agency individualist agrees with me that the reductionist thesis does not entail or imply the extreme view that institutional and other collective entities are incapable of action. Let me now, then, turn from the issue whether institutional entities are agents, to the issue whether they can have moral obligations or bear moral responsibility. As I argued in the preceding section of this chapter, if institutional entities are agents properly so-called, then they are moral agents.

Seumas Miller is a clear case of a moral individualist.²³ He contends that in any case where it seems that an institutional entity has an obligation or bears responsibility, an “individualist ana­lysis” can be given according to which no collective has a moral property that is not possessed by its individual members. He further suggests that the availability of individualist analyses shows that, “properly speaking, no collective [is] possessed of any moral properties” at all (Miller 2007: 390).

One problem with Miller’s view is that, if we had an individualist analysis of the truth conditions of sentences of the form, “Collective C has a moral obligation,” this analysis would show the circumstances under which a collective C would have a moral obligation. It would not show that, properly speaking, no collective has any moral obligations. As I mentioned before, there is an analysis of facts about bachelors in terms of facts about unmarried adult males. It would be a mistake to conclude from the availability of this analysis that, properly speaking, no person has ever been a bachelor.

On Miller’s thesis that “no collective [is] possessed of any moral properties” at all, we could not use the transitional duty to explain why a person has a moral obligation. On Miller’s view, no collective has any duties, so we might wonder why people have duties in their roles as members or officers of collectives. What is it about their role that gives them obligations if not that they are obligated to do their part in bringing it about that the collective acts as required?

Ludwig also advocates a kind of moral individualism. He argues for what he calls the “Factor Model” of collective responsibility according to which “any claim that a group is morally responsible for something must be resolved into a distribution of moral responsibility to its members, with none left over for the group per se” (Ludwig 2020: 79; Ludwig 2007).

I have already discussed Ludwig’s argument that collective entities are not agents, so let me here consider his argument from the “idleness worry.” This is not a good argument. The most it shows is that if a collective is responsible for something, some relevant persons must have moral reason to alter their behavior. It does not follow that there is no responsibility “left over for the group per se.” Moreover, the argument is compatible with the collectivist model I proposed earlier in this chapter. The collectivist model postulates the transitional duty, which guarantees that relevant people have a pro tanto duty or reason to do the things that would constitute a collective’s doing what it is morally required to do.

In earlier work, I argued for a version of the “Collective Moral Autonomy Thesis” or “CMA thesis” (Copp 2007a; Copp 2012). On the version I defended, it is conceptually possible for a collective to bear all-things-considered moral responsibility, or to have an all-things-considered moral obligation, without any relevant member being all-things-considered morally responsible in the matter, or having a relevantly related all-things-considered moral obligation.

One of my arguments for my version of the CMA thesis is an argument from moral com­plexity. To illustrate the point, I gave an example in which a government has an all-things- considered moral obligation to do something. The Prime Minister is the person charged with carrying out the government’s obligation. Unfortunately, she has a significant conflicting personal duty to her daughter, and in the circumstances, it outweighs her transitional duty to do what she would have to do in order for the government to fulfill its obligation. Hence the Prime Minister does not have an all-things-considered moral obligation to do what she would have to do in order for the government to fulfill its obligation. In the example, no-one else does either (Copp 1976; Copp 2007a; Copp 2012).

Ludwig has several responses to the example, and I have responded to them elsewhere (Copp 2012). It is perhaps worth mentioning that one of his responses turns on the “idleness worry” (Ludwig 2007: 424—426). If the Prime Minister does not have an all-things-considered moral obligation to bring it about that the government fulfills its duty, then, Ludwig says, she has no moral reason to do this, and it is idle to hold that the government has the all-things-considered obligation in question. But given the transitional duty, the Prime Minister has a pro tanto obli­gation to bring it about that the government fulfills its duty, which entails that she has a reason to do this. In the example, this reason is outweighed, but it still exists. It is a reason for her to regret what she does, even though she does what she is morally obligated all-things-considered to do. This is the moral residue of the transitional duty. The government’s all-things-considered obligation also has significance. The fact that it was not fulfilled can mean that the government is obligated to make amends.

Miller claims that, in my example, the Prime Minister must have a relevant duty “qua Prime Minister” to do what she would have to do in order for the government to fulfill its obligation (Miller 2007: 400; Miller 2010). I agree. Given that there is the transitional duty, and given her role in the government, the Prime Minister does have a pro tanto duty to do what she would have to do in order for the government to fulfill its obligation. Miller claims that this is what she is morally obligated all-things-considered to do qua Prime Minister (Miller 2007: 400; see also Miller 2010; Forrester 2010). The point, however, is that, to decide what to do, the Prime Minister must balance the transitional duty she has, qua Prime Minister, with the duty she has to her daughter, qua mother, in order to arrive at a conclusion as to what she (the person) is morally obligated to do simpliciter, all-things-considered. We each have a variety of roles. In deciding what to do, we need to take into account the duties we have in these roles. But, as moral persons, we need to reach a decision about the balance among these duties. In the example, the person who is Prime Minister is not morally obligated all-things-considered to do what she would have to do in order for the government to fulfill its all-things-considered obligation.

I believe that the example of the Prime Minister and her daughter shows that my version of the CMA thesis is true. This means that we should reject the strong reductionist thesis according to which our concepts, or the meanings of our terms, ensure that, necessarily, propositions about the all-things-considered obligations or responsibility of collective entities have the same truth conditions as corresponding propositions about the all-things-considered obligations or responsibility of relevant individuals. Despite this, of course, one might think that facts about the obligations or responsibility of collective entities are grounded in facts about the obligations or responsibility of relevant individuals (Bjornsson 2020: 128, 134; see Wringe 2014a). I cannot pursue this idea here, but I doubt it is right. I think it is too simple.

7.8