Are Functionalism’s Flaws Fatal?

Jan Klabbers’ two decades of “obsession” with functionalism as a theory of international organizations law (IOL) pays off in this article. It is a highly perceptive analysis of the evolution of functionalism, arguing that it made sense when created more than a century ago, but its explanatory and prescriptive power began to erode almost immediately as the nature and impact of international organizations (IOs) changed. He argues that functionalism helps to understand the relationship between IOs and member states, but not relations within an organization or between the organization and third parties. More radically, Klabbers challenges the functionalist assumption that all activities of IOs are beneficial or benign. Not only is this factually wrong, but it also serves to obfuscate the impact of IOs by presenting them as “neutral” “apolitical” and “purely technical” creatures (p. 18). This makes it difficult to hold them accountable for they wrongs they commit, like the cholera outbreak in Haiti, let alone for perpetuation and propping up “an unfair global structure” (blog).

Like much of his work, Klabbers turns a critical eye on the “problem-solving” approach to the study of international law and organizations, without losing sight of real world challenges the law and IOs struggle with and the pragmatic solutions they are struggling to find. His analysis of functionalism is an exercise in “reconstruction” (p. 22). In the hands of a lesser scholar, this can be a risky strategy because it can tempt one to (re)construct a straw man for the purpose of debunking it. Klabbers avoids that trap by a careful reading of seminal texts and a persuasive interpretation of how functionalist theory has informed the actual practice of IOL in the past and today.

In this comment, I will: a) extend Klabbers’ critique in one area (the weaknesses of principal-agent theory); b) elaborate on it in another (IO accountability and responsibility); and c) based on the above, conclude with some thoughts about the question he asks at the end: must functionalism be discarded or can it adapt?

I agree that principal-agent (PA) theory only takes us so far in trying to understand IOL. As Klabbers points out, it does not capture inter-organizational dynamics well, nor does it have much to say about an organization’s relationships with outsiders, like non-members (consider the AU’s relations with western powers), non-state actors (such as relations between UN peacekeepers and rebel forces), or individuals. Even more damning – and this is where Klabbers critique does not go far enough – PA theory does not adequately explain or prescribe for what it purports to cover: namely the organization’s relationship with its members. The delegation of law-making powers to an IO, for example, would seem to be fairly straightforward from the point of view of PA theory: the member states authorize an agent to engage in certain tasks that will produce law, with control mechanisms built in to ensure the agent does not exceed the authority granted. That framing is useful when the principal, agent, powers conferred and control mechanisms are easily identified. But often they are not: the “collective principal” may be multiple principals with diverging views; the agents may be several steps removed from the principal; the powers conferred can be hard to specify (how far can the notion of “implied powers” be taken?); and what counts as a control mechanisms can be unclear (does indirect judicial review count?)

Consider the UN Security Council. Can one reasonably assume that in creating the UN, the founding members expected the Security Council (SC) to establish transitional administrations (TA) with full governing powers, as it did in East Timor and Kosovo? Is this a matter of implied delegation by the member states of the organization? If so, are they acting as a collective principal or multiple principals, with different understandings of what they did? By sub-delegating law-making power to the TA, is the Council acting as a collective principal or multiple principals? Does the TA have discretion to sub-delegate by handing over some of its law-making power to local authorities, as it did in Timor by setting up co-governance arrangements? Are the various “agents” here (the Council and the TA) acting within their margin of discretion, or are these examples of unauthorized policy drift? Other than terminating the mandate, what control does the SC have over the transitional administration? What control does the rest of the UN membership have over the SC?

Staying with the theme of delegated law-making, the growing role of non-state actors (IO secretariats and NGOs) adds another layer of complexity. Sometimes these actors have a formal law-making role in international organizations, like the International Law Commission in the UN, or employee and employer representatives in the International Labour Organization. Elsewhere, the role of non-governmental actors is more informal. They are not delegated law-making powers per se, but can have a significant impact on how international law develops. In these cases of attenuated delegation, the principal (member states) and agent (fr example, NGOs contributing to the normative process) may be so far removed from each other that it is misleading to suggest they are in a relationship at all. Moreover, in constructivist theory, international bureaucracies and NGOs may themselves be “agents of socialization”, spreading norms that ultimately become internalized in national governments and societies. This turns principal-agent theory on its head, because the agents transform how the principals (states) define their interests and even identities.

Turning to accountability, Klabbers rightly notes that formal mechanisms are rare and hard to activate given the functional immunity of IOs. This too is a problem for PA theory, because accountability mechanisms are the devices principals use to control the agent – to avoid “drift”, “slippage” or “shirking”. Since the formal mechanisms are weak, Klabbers observes a “general move towards accountability in whatever form” (p. 75) and alternative “responsibility practices” (p. 76). He may be right that these often do not depend on law, but I would argue that they are especially powerful when tied to a sense of legal obligation, as opposed to more elusive and contestable normative standards.

Indirect judicial review is one such mechanism – as the Kadi case on the Taliban/Al-Qaeda sanctions list illustrates. While it is hard to specify the UN’s obligation to individuals placed on the list, and the European Court of Justice (ECJ) does not have the power of judicial review over the UN Security Council, the listing regime was challenged on legal grounds, not simply because it seemed unfair or unwise. By striking down EU regulations adopted pursuant to an SC resolution, implementation of the resolution in Europe became virtually impossible. That the SC felt the effect is clear: by later enhancing procedural rights for those placed on the list, it indirectly responded to the ECJ’s indirect judicial review.

Ruth Grant and Robert Keohane identify seven types of accountability mechanisms that may apply to international organizations, from the supervisory and fiscal accountability that states wield over an organization, to the hierarchical authority the executive head wields over individual staff members, to peer accountability whereby one organization evaluates another. The seventh is “public reputational accountability.” It is widely accepted that reputational concerns affect state behavior, for interest-based and normative reasons. Do they affect IO behavior? It is easy to see how leading member states may worry about their reputation for actions taken (or not taken) in an IO, though collective decision-making dilutes that. It is also easy to see how the executive head of an organization may worry. For both, consider the Rwanda genocide. But what about IOs as corporate entities? Are they affected by reputational concerns?

The deliberations leading to the intervention in Iraq in March 2003 suggest they are. The best explanation for why the doomed second Security Council resolution explicitly authorizing military action was not formally tabled is that at least some members of the Council were worried about the institution’s credibility. This was certainly true of France, which quietly urged the US to withdraw the draft, knowing that if the latter went to war regardless, a failed vote would do more harm to the Council than no vote at all. It is probably true of the US as well, which, according to Richard Haas, director of policy-planning in the State Department at the time, withdrew the draft resolution in order “to preserve the Council’s potential viability in the future.” Both the US and France – the most bitter protagonists in the debate – wanted to preserve some credibility for the Council so it could help to deal with the aftermath of the Iraq intervention and with crises elsewhere.

Both judicial review and reputational factors are at play in the Haiti cholera case. Leaving aside the possibility that the UN may simply have been wrong to claim immunity – that this was a private rather than public law dispute obliging the UN to “make provisions for appropriate means of settlement” — the matter has been brought to a number of courts in the US. The decision of a New York court that the UN’s immunity stands is being appealed; two other cases are pending. Even if the plaintiffs never succeed, the cases add to what Klabbers himself calls – “the justified outcry over the outbreak of cholera…for which the UN is widely considered responsible” (p. 76). The UN may have dug too deep a hole to reverse course in the Haiti case, but the Secretary-General will likely think twice if a similar situation arises in the future – not only because s/he would worry about his/her personal reputation, but because the credibility of UN peacekeeping is at stake.

Klabbers concludes by asking whether functionalism can adapt or must it be discarded altogether. He suggests that the alternatives to functionalism – constitutionalism, global administrative law and the ILC Articles on IO Responsibility – can’t easily be reconciled with it. He claims that “to save functionalism” IOs must become more modest in how they define the “global common good” and more accountable to constituencies other than member states. It would seem that constitutionalism is a bridge too far: it cannot be reconciled with functionalism since it assumes a constitutional order that supersedes the authority of states. But global administrative law, with its emphasis on transparency, reason-giving and participation, along with the other accountability mechanisms described above, are more reconcilable. These derive from the practices of IOs, where member states are still the central players, and they have gained traction because they are acquiesced to by member states. In that sense are consistent with functionalism. But they are more inclusive in how they operate. Perhaps we are heading towards a looser form of functionalism that owes less to principal-agent theory and more to a kind of discursive accountability that involves the interaction of governments, IOs, NGOs, experts and the many stakeholders who are affected by the activities of IOs – with national courts chiming in from time to time. Law and a sense of legal obligation are not irrelevant in this process, it is simply that the legal norms are given effect in a diffuse manner that depends more on reputational factors than formal dispute settlement mechanisms.