Is the United Nations Charter law? Frequent violations of the charter and its uncertain impact on state practice, Jack Goldsmith points out in an April 16 post, lead many to wonder whether it really functions as law—but they wonder too much, he suggests. The question whether the charter is law is misleading or meaningless, or both. Uncertainties about the international law of war powers, he writes, differ little from uncertainties about the meaning and efficacy of the constitutional law of war powers. Goldsmith proffers no theory of law and takes no position on whether either the Constitution or the charter is law, though one might infer the point to be that if one thinks of the Constitution as law, one shouldn’t balk at thinking of the charter as law. His argument draws on the article he links to, “Law for States,” co-authored with Daryl Levinson.
I’m not persuaded. Stated at a sufficiently high level of abstraction, features of the two systems do appear similar. But delving more deeply into the particulars of their operation unveils profound dissimilarities. The most important difference lies not within the two systems themselves but externally, in the communities those systems aspire to govern.
“Law for States” focuses on the internal dynamics of both systems. It argues that “international laws norms are imprecise, contested, internally contradictory, overlapping, and subject to multiple interpretations and claims” because international law lacks centralized and hierarchical courts and a centralized law-making process. International law’s uncertainties, it argues, are caused by its internal institutional deficit.
In the most immediate sense, that’s true. But the important question is: What causes international law’s institutional deficit? The answer lies in the vastly different political conditions that prevail within the international community as compared to the United States. I noted a few in a Lawfare post last October:
Legal systems work well when actors within the system are relatively equal in power. Future dealings are expected. Trust is high. A consensus exists concerning foundational values. The price of non-cooperation is steep. Individual and collective interests align. Underlying social norms reinforce legal norms. Free riders and transgressors are easily spotted and penalized….Law’s force derives from underlying political conditions such as these. Absent such conditions, law goes dark.
Background conditions such as these shape a community’s institutional structures and, ultimately, determine whether its law generates compliance. Those exogenous conditions have been and still are dramatically different within the United States than in the rest of the world. And they give rise to crucial internal differences between the international and constitutional systems.
First, the mix of consent and coercion is different. In a consent-based system, actors within the system are thought to choose those rules they honor, whereas in a coercion-based system, they do not. Legal systems generally combine the two to varying degrees and fall at different points on a spectrum between poles of coercion and consent. International law is much closer to the consent side of the spectrum. Its formally legitimating norm continues to be the so-called freedom principle, which specifies that the rules of law binding on states emanate from their own free will, and that restrictions must be proven. A state therefore need not establish that international law permits a particular act; it is for others to establish that international law prohibits that act.
The foundational presumption of constitutional law is the opposite. Absent constitutional authorization, government action is thought to be prohibited. The president, Congress, and the courts all must point to a specific constitutional provision that authorizes what they want to do. They exercise only delegated power, the source of that power being the people. Nations, in contrast, are the source of their own power; no higher authority within the international system delegates power to them. In the domestic constitutional order, power flows to governments. In the international order, it flows from them. In the domestic constitutional order, the default principle is restriction. In the international order, it’s freedom.
Second, these inverted power flows, amplified by disparate background conditions, produce different processes for resolving disputes. At a superficial level those processes look similar—because the sources of authority that shape those processes look similar. The Constitution and the U.N. Charter both have a text. Each is backed by its Framers’ intent. Each is interpreted by courts. Each has generated patterns of practice.
But there is a critically important distinction: These sources interact differently within the two systems because they are weighted differently within the two systems. It’s impossible to measure precisely how much, but simple observation will suffice. Do the seven billion people in the world view the U.N. Charter with the same near-religious reverence with which the American people view the Constitution? Are the Framers of the U.N. Charter who met in San Francisco in 1945 viewed as the demi-gods that the American people believe met in Philadelphia in 1787? Are opinions of the International Court of Justice accorded the same respect by nations of the world that opinions of the Supreme Court are accorded by government officials and private citizens within the United States? Are clear violations of the U.N. Charter regarded with the same horror by nations as clear violations of the Constitution are by Americans?
Here’s how one constitutional law professor assessed the comparative bindingness of the war power rules within the two systems. In the run-up to the 2008 election, then-Sen. Barack Obama was asked whether he would wait for approval by the U.N. Security Council before committing U.S. troops to combat. “[W]e will never take military options off the table,” Obama replied. “And it is important that we don’t provide veto power to the United Nations or anyone else in acting in our interests.” Asked about the Constitution’s constraints, however, Obama’s answer was very different. “The president,” Obama wrote, “does not have power under the Constitution to unilaterally authorize a military attack in a situation that does not involve stopping an actual or imminent threat to the nation.”
Why might Obama have believed the two systems’ obligational force to be different? After all, the question in each case is, as Goldsmith says, the same: How much weight should be given to text and original meaning and how much to historical practice? The answer, again, is that because default principles concerning freedom of action differ within the two systems, and because the sources of authority in each system are weighted and interrelate differently, the level of uncertainty within each concerning their war-powers rules is also different. The net effect is that it’s easier to regard deviations as unlawful violations in the constitutional system than it is in the international system.
“Law for States” does not compare (or even examine) the actual practice each system generates. But look, quickly, at what what’s at issue. Proponents of broad presidential war powers sometimes refer to a list of over two hundred instances in which force was ordered by presidents without congressional approval. In fact, in few of those instances did the use of force pose a threat of retaliation against potential U.S. military targets beyond the battlefield or significant danger to non-combatants in the United States or elsewhere. The overall risk to the nation incurred in these incidents was almost always negligible. Moreover, the argument that Congress has implicitly approved through acquiescence is dubious. Congressional inaction intended as silence looks, procedurally, the same as congressional inaction intended as disapproval: In neither case is any measure officially adopted. Collective action problems make it difficult to get 535 members to deal coherently with such challenges. So there are reasons to weight constitutional war powers practice lightly relative to other sources such as original intent.
In contrast, there are reasons to weight international war powers practice more heavily as against other sources. Goldsmith points out that we don’t know how many cross-border interventions would have occurred but for the charter, which is true (though it’s hard to identify instances of forbearance by the U.S. government; its typical response to obstructive charter constraints was more likely Madeleine Albright’s, responding to British legal objections to NATO’s proposed Kosovo campaign—“Get new lawyers”). It’s important to recognize the limits of our uncertainty, however, and to stick to hard evidence. Evidence of compliance with the charter is speculative; disconfirming evidence, on the other hand—evidence of violation—proves to a certainty that the rule has not caused the state behavior it requires. And that evidence is voluminous and indisputable. The charter’s rules governing use of force have been violated hundreds of times by many states over many years.
Thus the question arises whether the U.N. Charter is still law. Goldsmith writes that argumentation, interpretation, and practice have altered the original meaning of Article 51 of the charter, which, he suggests, no longer requires an actual armed attack as a predicate for the use of defensive force; the mere anticipation of an armed attack will now suffice. That view seems commonsensical. But some prominent commentators have questioned it. Scholars such as Hersh Lauterpacht, Louis Henkin, Ian Brownlie, and Phillip Jessup appeared to believe that the use of defensive force was permissible only in response to an actual armed attack. The ICJ has never embraced the “imminence” argument. I’m aware of no pattern of practice in which states desiring to use defensive force in response to a probable attack have delayed doing so until the attack became imminent. Rather, the stain of violation smears over the U.N. Charter’s entire canvas, leaving a veritable Jackson Pollock of defacement that defies categorization.
That, I suspect, is why Goldsmith rejects arguments that the charter permits humanitarian intervention by states without Security Council approval. The putative historical precedents for it—or for any adaptive exception—don’t line up. Goldsmith writes that it’s all in one’s theory of a document’s meaning and change; what, then, is the theory under which the erosive force of massive, incongruent, unclassifiable practice has stopped at the armed attack requirement of Article 51? Why did the speeding sled of dissolution halt abruptly and precisely three-fourths of the way down the hill but not at the bottom, killing Article 2(4)? If incongruent practice can change the charter’s war power rules a little, it can change them a lot, and if it can change them a lot, it can eviscerate those rules altogether. There is no conceptual difference between a process that creates a slightly different rule to substitute for a former rule and a process that creates a laissez-faire rule to substitute for a former rule—especially when the default rule is the freedom principle.
It’s unclear why Goldsmith would join issue on whether the charter permits humanitarian intervention or the use of defensive force in response to a non-imminent threat. “Law for States” explicitly declines to address whether the charter is law. If one is unwilling to sign onto the proposition that the charter’s use-of-force rules are legally binding and that violations are unlawful, why get embroiled in efforts to parse the rules’ meaning? It’s not necessary to propound a theory of law to know what’s not law. “The validity of the law,” Hans Kelsen wrote, “presupposes a minimum efficacy of the law.” If Kelsen is right, the more parsimonious approach is to assess the efficacy of the charter’s war powers rules and judge whether they are valid law.
Traditional international law doctrine provides little useful guidance in telling us precisely when an emerging norm has become binding or an existing rule becomes non-binding. Whether the issue is a rule’s birth or death, however, the transformative process is the same. At some point, the balance tips; at some point, the rule changes; at some point, obligation shifts. Working rules become paper rules. At that point, opposite interpretations become equally reasonable—and a legal instrument that means anything also means nothing. That’s the point at which the U.N. Charter has arrived. Dwelling at length on its meaning is now, in Michael Walzer’s words, “utopian quibbling.” Some understandings of the charter may still in some regions be law; it is entirely possible, for example, that different use-of-force rules now prevail in Europe than in Africa or the Middle East. Some understandings of the charter may still, in some circumstances, prompt compliance as social norms; foreign policy decision-makers sometimes are influenced by sub-legal norms. At the margins, on either side of the tipping point, the functional difference between law and non-law may be virtually nil. But if states intended to make the charter’s use-of-force rules universally binding, they would have put in place a global incentive structure to induce compliance with those rules. They have not. States distrust concentrations of state power. They see the risks posed by a universal system that provides no escape from lawfully centralized coercion as greater than the risks of a system that lacks centralized coercive enforcement mechanisms—mechanisms of the sort that exist within the U.S. constitutional system. The political conditions that underpin the international system do not yet support such mechanisms.
Would it have been misleading or meaningless to have asked, in 1945, whether the Kellogg-Briand Pact was still law? I wouldn’t have thought so. Knowing whether a law works, and if not, why not, is the first step to improving it. That process is delayed by believing that a plainly ineffectual treaty is still law.