Show sidebar

My discrete but important disagreement with Harold Koh on the lawfulness of the strikes on Syria

Print submit to reddit

I argued last night that the President’s strikes against Syria appear to violate Article 2(4) of the United Nations Charter and that therefore it was unconstitutional for President Trump to order such strikes absent congressional authorization.  What’s more, all ten of the experts Ryan canvassed agree with me that the strikes were unlawful on either or both grounds (breach of the Charter and/or unconstitutional), if I’m reading them correctly.  (Accord:  Ashley DeeksRyan Goodman, and Deborah Pearlstein, too.  And here’s Marko Milanovic:  “I can’t imagine even a remotely plausible argument (let alone a persuasive one) as to why this act is not a breach of Article 2(4) of the Charter. . . .  [T]his is a situation in which the US government doesn’t have even a colourable argument that its conduct is lawful.”)

Swimming against the tide, however, my friend Harold Koh takes issue with this consensus view.  Harold argues that President Trump’s use of force might be lawful, under both the Charter and the Constitution.  In this post I’ll briefly explain why I’m not persuaded by Harold’s argument, attractive though it may be to those of us who think that morality and decency might sometimes cry out for humanitarian intervention in the face of the sorts of atrocities that Assad is inflicting on his own population (and therefore might justify Congress voting here to authorize a breach of the Charter in circumstances where there is some likelihood that the use of force might actually deter Assad from his butchery).

At the outset, however, it is important to clarify that the point of disagreement between me and Harold, although very important, is also quite narrow and discrete.  As I understand Harold’s post, he does not take issue with my primary point–one that often gets overlooked in such debates–which is that whatever the (famously contested) scope of the President’s Article II authority to use force without congressional authorization might be, that constitutional authority does not afford the President the power to cause the United States to breach its treaty obligation under Article 2(4) of the Charter, any more than it gives the President the authority to violate a statute constraining the use of force, such as the War Powers Resolution, the War Crimes Act, or the Torture Act. 

Where Harold disagrees with me, instead, is on the question of whether a humanitarian intervention such of this does, in fact, violate Article 2(4) of the UN Charter.  Invoking the “necessity” defense in Article 25 of the International Law Commission’s Draft Articles on State Responsibility, Harold argues that there is–or ought to be–“a narrow ‘affirmative defense’ to Article 2(4) of the U.N. Charter that would clarify the contours of an emerging lawful exception to a rigid rule that humanitarian intervention is ‘per se illegal.'”  Such an “affirmative defense” would hardly be unbounded:  Harold elaborates six fairly rigorous conditions that would have to be satisfied in order to invoke that defense, which I set out in a footnote below.*

Harold’s preconditions for a new “humanitarian intervention” exception, or affirmative defense, to Article 2(4) are extremely well-considered, very carefully crafted, and admirably constraining.  If the international community were one day to use Harold’s formulation as a template for a proposed amendment to the Charter, I would be deeply sympathetic to such an initiative (although of course I’d also want to carefully consider any counterarguments and alternative proposals, too–after all, as Legal Adviser Matheson reminded us with respect to Kosovo in 1999, “establishing new doctrines or precedents” that weaken the force of Article 2(4) might later “haunt” us, and our allies, if it the new exception is susceptible to being “misused by others”).

Nevertheless, I don’t find convincing Harold’s argument that there is any existing such affirmative defense that might justify President Trump’s actions yesterday, for at least four reasons.

First, and most fundamentally, Harold’s test is aspirational.  The Charter itself does not express, or in any way suggest, the existence of any such affirmative defense.  Nor am I aware of any other serious argument that such a test reflects current law, or even, for that matter, that it describes very many actions that states have taken–Kosovo being the single possible exception.  [Much more on this in a thorough 2013 post from Dapo Akande.]  Perhaps, as Harold hopefully suggests, “[w]e have reached a lawmaking moment, where international lawyers in and out of government need to discuss and define a narrow” affirmative defense.  Alas, however, such new law has not yet been made, and “international lawyers in and out of government” don’t have the power to make it–that’s something that would require the assent of the state parties to the Charter.

Second, Harold invokes the idea in the ILC’s Draft Articles on State Responsibility that there ought to be such an affirmative defense of “necessity” in cases where, for example, a state might take “certain forms of countermeasures to stop illegal acts by others.”  The ILC Articles are not a treaty, however, nor any other form of law that might supersede or amend the U.S.’s obligations under the Charter.  And even if they were, they’d be of no avail here–indeed, they tend to undermine Harold’s position in several respects:  For one thing, Article 25(1) of the Draft Articles itself would preclude a necessity defense in cases where the action in question would “seriously impair an essential interest of the State or States towards which the obligation exists”–and much as we all might be deeply unsympathetic to Syria’s interest in preserving its sovereignty as reflected in Charter Article 2(4) under the current circumstances, surely the U.S. strikes “seriously impair” that state interest.  Moreover, Article 25(2) of the ILC Articles would preclude such a necessity defense where “the international obligation in question excludes the possibility of invoking necessity”–and the Charter does so.**  Finally, and most specifically, Article 50(1)(a) of the ILC Articles specifically provides that “[c]ountermeasures shall not affect . . . the obligation to refrain from the threat or use of force as embodied in the Charter of the United Nations.”

Third, even if–contrary to what I’ve written above–there were a stronger argument for Harold’s claim that international law currently recognizes an affirmative defense to Article 2(4) for certain humanitarian interventions, that is decidedly not the view of the United States.  As Harold laments in his Houston lecture, in 1999 not only did the United States government not invoke a “humanitarian intervention” rationale to justify what appeared to be a breach of the Charter in Kosovo–it affirmatively announced (through the State Department Legal Adviser) that the “United States . . . had not accepted the doctrine of humanitarian intervention as an independent legal basis for military action that was not justified by self-defense or the authorization of the Security Council.”  Michael J. Matheson, Justification for the NATO Air Campaign in Kosovo, 94 Amer. Soc’y Int’l L. Proc. 301, 301 (2000).  (Harold writes that “[a]t the time,” when he was “acting not as a government lawyer, but as a human rights policy official within the U.S. government,” he “thought it outrageous that the U.S. government would fail to state a legal rationale to justify its use of force.”  Surely it’s not outrageous, however, for a government to refrain from stating a legal rationale for a use of force in a case, such as Kosovo, where the lawyers and relevant officials had in fact concluded–correctly, in my view–that the use of force, however “legitimate” (Matheson), was not lawful.)

Moreover, I would be very, very surprised if the Trump Administration were to deviate from this longstanding view of the United States and suddenly adopt Harold’s view of a new international law “affirmative defense” to Article 2(4) for humanitarian interventions.  That’s simply not going to happen.  And therefore Trump will not be able to use Harold’s argument as a grounds for establishing that he did not breach the Charter–and, in turn, he will be precluded from invoking that argument as a grounds for establishing the constitutionality of last night’s breach of the Charter without congressional authorization.

Fourth, and finally, even if Harold’s view of the affirmative defense to Article 2(4) were correct, it does not appear that President Trump has satisfied all of the necessary conditions.  In particular, the strikes last night–in contrast to the U.S.’s use of force in Kosovo in 1999–were not “collective, e.g., involving the General Assembly’s Uniting for Peace Resolution or regional arrangements under U.N. Charter Chapter VIII.”  [Marko Milanovic adds:  “And even if there was a customary humanitarian intervention exception from the prohibition on the use of force (and there isn’t), its requirements would clearly not be met in this instance.  Hundreds of thousands of people have died in Syria even without the use of chemical weapons, and thousands of people will continue to die even if the Assad regime never uses such weapons again.  There is, in other words, nothing legally or morally unique about the use of chemical weapons as opposed to other war crimes and crimes against humanity in Syria which did not (and will not) provoke an interventionist response.”]

Thus, even on Harold’s own aspirational view of a new law of permissible humanitarian intervention without Security Council approval, the use of force in Syria breached the Charter (and thus, according to my argument, also violated the Constitution for lack of congressional authorization).


* According to Harold, there would or should be an affirmative defense to what would otherwise be a breach of Article 2(4) if and only the following six conditions for humanitarian intervention are satisfied:

(1) If a humanitarian crisis creates consequences significantly disruptive of international order—including proliferation of chemical weapons, massive refugee outflows, and events destabilizing to regional peace and security—that would likely soon create an imminent threat to the acting nations (which would give rise to an urgent need to act in individual and collective self-defense under U.N. Charter Article 51)

(2) a Security Council resolution were not available because of persistent veto; and the group of nations that had persistently sought Security Council action had exhausted all other remedies reasonably available under the circumstances, they would not violate U.N. Charter Article 2(4) if they used

(3) limited force for genuinely humanitarian purposes that was necessary and proportionate to address the imminent threat, would demonstrably improve the humanitarian situation, and would terminate as soon as the threat is abated.

In particular, these nations’ claim that their actions were not wrongful would be strengthened if they could demonstrate:

(4) that the action was collective, e.g., involving the General Assembly’s Uniting for Peace Resolution or regional arrangements under U.N. Charter Chapter VIII;

(5) that collective action would prevent the use of a per se illegal means by the territorial state, e.g., deployment of banned chemical weapons; or

(6) would help to avoid a per se illegal end, e.g., genocide, war crimes, crimes against humanity, or an avertable humanitarian disaster, such as the widespread slaughter of innocent civilians, for example, another Halabja or Srebrenica.

To be credible, the legal analysis of any particular situation would need to substantiate each of these factors with persuasive factual evidence of: (1) Disruptive Consequences likely to lead to Imminent Threat; (2) Exhaustion; (3) Limited, Necessary, Proportionate, and Humanitarian Use of Force; (4) Collective Action; (5) Illegal Means; and (6) Avoidance of Illegal Ends.

** The ILC Commentary on draft Article 25 is to similar effect:  “As embodied in article 25, the plea of necessity is not intended to cover conduct which is in principle regulated by the primary obligations. This has a particular importance in relation to the rules relating to the use of force in international relations and to the question of “military necessity”.  . . . The question whether measures of forcible humanitarian intervention, not sanctioned pursuant to Chapters VII or VIII of the Charter of the United Nations, may be lawful under modern international law is not covered by article 25.4.

Image: U.S. Navy

About the Author

is a professor at the Georgetown University Law Center. You can follow him on Twitter (