Are “Coalitions of the Willing” Moral Agents?

| August 2014
Facebook Twitter Email
Print Friendly, PDF & Email

In “Coalitions of the Willing and Responsibilities to Protect: Informal Associations, Enhanced Capacities, and Shared Moral Burdens,”1  Toni Erskine argues that ‘coalitions of the willing’(CWs) lack moral agency, and are therefore ineligible for obligation, blame, and punishment.

I disagree. Here is an example. Suppose the United States, Saudi Arabia, and Turkey formed a coalition to intervene militarily in Syria, and the coalition partners needed to reach a consensus on which officer would lead operations. Imagine the following preferences:

United StatesSaudi ArabiaTurkey
First preferenceGen. WashingtonGen. RiyadhGen. Ankara
Second preferenceGen. TexasGen. TexasGen. Texas
Third preferenceGen. Riyadh or Gen. AnkaraGen. Washington or Gen. AnkaraGen. Washington or Gen. Riyadh

Negotiations occur and reasons are given. No party changes their preferences, but consensus is needed. So each partner begrudgingly compromises. The coalition agrees Gen. Texas will be in charge. Unfortunately, he gives poorly thought-through orders, resulting in numerous combatant and non-combatant deaths.

Where do we place blame? Obviously, some blame belongs with Gen. Texas himself. Additionally, though, the three states put Gen. Texas in charge and (suppose) his conduct was reasonably foreseeable. However, it was not the three individual states, taken separately, that put him in charge—it was the three states together. Each state can truthfully say “I would have preferred Gen. Washington / Riyadh / Ankara. If I had decided, Gen. Texas would not have led operations. Unfortunately, I did not decide. The group did. I compromised to achieve group consensus. Given the others’ positions, it was clearly either Gen. Texas or nothing.” It seems it is only by viewing the three states as a group—and as making decisions as a group—that we can make sense of the decision to appoint Gen. Texas. A natural assessment is that the coalition made the decision, and is to blame for it.

This assessment has implications for our assessment of appropriate punishment. Punishing the coalition itself might mean reforming or dismantling the coalition, or ‘naming and shaming’ the coalition in its own right. But punishing the coalition’s members—all Erskine could require—would mean something different: naming and shaming the individual states, say. The latter, state-targeted punishments would be inappropriate, given the extent to which the agency of members was constrained by the need for the coalition to form a unified decision. This thought experiment suggests that viewing CWs as moral agents allows us to make sense of their decisions, their blame, and their punishments.

Why, then, does Erskine deny CWs group agency (and thus group blame and punishment)? There are five conditions at play, each of which Erskine believes is necessary for group agency. According to Erskine, CWs fail to meet four of the five conditions. Each of these four assessments hinges on her judgment that CWs have merely an “informal organizational structure and deliberative capacity” (p. 138), which “involves negotiation, bargaining, and consensus-building among the various constituents in the absence of existing organizational structures and decision-making procedures” and “the inability to translate calls for collective action into decisions that are binding upon the group as a whole if such proposals depart from the positions of some of its members” (p. 123).

There are several ideas at play here. I have just suggested that CWs can make decisions that “depart from the positions of some of its members”: if the appointment of Gen. Texas isn’t such a decision, I am not sure what is. Perhaps, then, the issue is that CW partners aren’t bound by these group decisions. If ‘bound’ means ‘legally required to abide,’ then it is true that CW partners are unbound. True, but unhelpful: our aim in giving an account of collective moral agency in international relations should be to ultimately guide the law; to instruct the law on which agents can coherently be held to account and which cannot. If we ultimately defer to the law on this question, then the law can hold to account whoever it likes. On the other hand, if ‘bound’ means ‘committed to abide,’ then CW partners are as committed as we can reasonably expect any member of any group agent to be (which is to say, fairly robustly committed, but not entirely unconditionally committed). The unboundedness of members is not the problem.

Erskine’s dispute, rather, lies with the way in which our hypothetical coalition’s group-level decision was reached: it was reached by “informal,” rather than “formal,” means. Erskine does not provide distinguishing features between formal and informal decision-making.2 One possible feature is functionality: perhaps formal mechanisms are such that from the same inputs, one always gets the same output. Negotiation isn’t a ‘function’ in this sense: the outcomes of the same negotiation mechanism might vary across cases, even if the mechanism has the same inputs (initial individual opinions) each time. Yet if something as strict as functionality is required for a decision-making entity to have agency, then it looks doubtful that states are agents. The precise wording of states’ laws cannot be predicted simply by knowing what each legislator prefers the wording to be. These facts underdetermine the law’s wording, since the mechanisms for moving from legislators’ preferences to legislature decisions are insufficiently fine-grained in states. In fact, many states’ decisions are made by negotiation, just as CWs’ are. Yet Erskine insists states do have agency-conferring decision-making apparatuses (p. 120).

I suggest, then, that we should agree that group decision-making procedures are crucial to group agency, but be more permissive than Erskine about how “formal” these procedures need to be.3 This would have the upshot that states and (at least some) CWs have agency, including our hypothetical Syria coalition. To deny that we can “coherently talk about assigning duties or apportioning blame to the coalition itself” (p. 125) is to afford ourselves an unduly limiting picture of the moral world.



  1. See Toni Erskine, “Coalitions of the Willing and Responsibilities to Protect: Informal Associations, Enhanced Capacities, and Shared Moral Burdens,” Ethics & International Affairs, 28 no. 1  (spring 2014), pp. 115-145.
  2. On “informal” decision-making, Erskine points us to Thomas Christiansen and Christine Neuhold, eds., International Handbook on Informal Governance (Cheltenham, UK: Edward Elgar), but the contributors to this volume use numerous and varied conceptions of “informal,” as the volume’s editors note (Christiansen and Neuhold, “Introduction,” in International Handbook on Informal Governance, p. 3).
  3. I give a more precise proposal in “Collectives’ Duties and Collectivization Duties,” Australian Journal of Philosophy  91 (2013), pp. 213–48, at pp. 234–238.
Facebook Twitter Email

Category: Blog, Issue 28.1, Online Exclusive

Comments are closed.