Hamdan, the ICRC Commentary to Protocol II, and Conflict Party Structure: A Reply

[Editors’ note: We welcome the following response to Marko Milanovic’s initial guest post regarding the Hamdan decision. Ryan Goodman is Rita E. Hauser Professor of Human Rights and Humanitarian Law at Harvard Law School; Derek Jinks is Professor at the University of Texas Law School; and Anne-Marie Slaughter is Dean of the Woodrow Wilson School of Public and International Affairs and the Bert G. Kerstetter ’66 University Professor of Politics and International Affairs at Princeton University.]

We write to respond to Marko Milanovic’s puzzling post concerning citations to the ICRC Commentary to Additional Protocol II (the Commentary) by the Supreme Court in Hamdan and in an amicus brief we co-wrote in that case. In this post, we address only Milanovic’s criticism of our (and the Court’s) citation to the Commentary. In that brief, we argued–in part–that the distinction between non-international armed conflicts and international armed conflicts is the party structure of the conflict. This is a modest–and, at some level, uncontroversial–proposition. We cited the Commentary to support this point. It is possible to read the Commentary, as Milanovic does, to define a non-international conflict in terms of party structure plus a territoriality requirement. We disagree with that reading, as we argued to the Court and will describe below. But regardless, the Commentary unquestionably supports the part of our argument that we cited it for.

To situate our discussion appropriately, allow us to recap quickly. In support of our claim that non-international armed conflicts are conflicts of a state/non-state party structure, we cited, inter alia, to the Commentary—quoting the following language:

“a non-international armed conflict is distinct from an international armed conflict because of the legal status of the entities opposing each other”

Milanovic points out that the sentence continues, after a colon, as follows:

the parties to the conflict are not sovereign States, but the government of a single State in conflict with one or more armed factions within its territory.”

[Note that Milanovic mistakenly identifies only the clause beginning with the word “but” as the omitted text. We (and the Court) omitted the entire independent clause following the colon (beginning with the words “the parties”).]

Milanovic argues that the omitted language somehow qualifies (or perhaps even completely subverts) our claim that the Commentary endorses our interpretation of “international” as a party structure concept. He further argues that the omitted clause is “completely contrary” to the Court’s ultimate position. We couldn’t disagree more—with respect to both points.

Before addressing these points, we first want to express our regret over the inaccurate pincite for the quote. Since the page of the Commentary we cite (p. 1351) also supports the specific point we advance in the brief, we strongly suspect that, in trimming a double page reference to a single reference, we dropped the wrong pincite for the specific quote provided. Of course, this is neither here nor there–the inaccurate pincite was sloppy. We want to underscore, though, that the page we cite also provides strong support for our point about party structure. And, more importantly, the page of the Commentary that our brief does cite–in a section entitled “[t]he distinction between international and non-international conflicts”–provides further strong evidence that Milanovic’s over-reading of the omitted language is plainly inaccurate. Indeed, the Commentary supports our ultimate argument–and the Court’s ultimate conclusion.

First, the Commentary supports, unequivocally, the limited point for which we cited it–that international conflicts involve a specific party structure and non-international conflicts involve a different party structure. Notably we did not cite this language in the Commentary as dispositive of the ultimate question—-whether Common Article 3 applies to some or all trans-territorial conflicts. It would have been odd–and unpersuasive–to do so since this source is not the Commentary to Common Article 3 itself (and, of course, the United States is not party to Protocol II). We cited it only for the first step of our argument: the party-structure point.

Milanovic might respond that even if the concept of non-international armed conflict presupposes a certain party structure, it may well be that Common Article 3 (and Protocol II, for that matter) requires something more than a qualifying party structure. It may be that the field of application is limited to conflicts exhibiting both a particular party structure and a limited territorial scope. The text of Common Article 3 is instructive—defining its field of application as “armed conflicts not of an international character occurring in the territory of one of the High Contracting Parties.” [Italics added]. Both our amicus brief and the Court take this idea seriously–analyzing the meaning of this apparent territorial limit in light of the text, structure, and history of the Conventions. We do not want to rehash these arguments here, but rather to emphasize that we went on to make them in the brief itself as additional parts of our overall argument. Thus it is Milanovic who is distorting our (and the Court’s) argument by taking one aspect and presenting it as our entire claim.

Further, Milanovic might suggest that the omitted clause of the Commentary to Protocol II supports the view that the definition of non-international armed conflict depends on party-structure plus a territoriality requirement. The omitted clause does not support the territorial “plus” any more, or in any different way, than the treaty text itself. Per Article 1, Protocol II applies to non-international armed conflicts “which take place in the territory” of a state (This is similar to the text of Common Article 3 referring to conflicts “occurring in the territory of one” state).

The key issue is whether this wording is implicitly exclusionary or implicitly inclusionary; in other words, whether the asserted requirement that a conflict between a state and non-state actors occur within the territory of a state excludes armed conflicts that take place in the territory of more than one state. In our amicus brief, we analyze that question in five lengthy paragraphs beginning with the sentence immediately following the quotation from the Commentary. We discuss at length the correct interpretation of clauses—whether in the Commentaries or the text of the Conventions—that refer to the territory of a state party.

Formulations like the “but clause” of the omitted language simply stand for one of two propositions, depending on the exact wording and context: (i) Common Article 3 applies to conflicts occurring in the territory of at least one state party to the Geneva Conventions or (ii) Common Article 3 applies even to conflicts that occur wholly within a single state—even if the conflict involves only one state and, at times, even if it involves no states. To conclude otherwise would produce a number of inexplicable and untenable gaps in the Conventions. The clause animating Milanovic’s post is no different than these others. It plainly falls within the category of applying even to conflicts that occur wholly within the territory of a single state.

Our analysis in the brief explains this point in far more detail than we will here, but we want to mention just a couple of additional points about the omitted language. Recall the wording of the omitted clause: “the parties to the conflict are not sovereign States, but the government of a single State in conflict with one or more armed factions within its territory.” Neither the “sovereign States” formulation or the “but clause” is, as a textual or logical matter, exclusionary. Conflicts between sovereign states are “international” even if non-state armed factions fight on one or both sides of the conflict. And conflicts between states are, obviously, “international” even if they occur entirely in the territory of one state. These non-international elements do not deprive the conflict of its “international” character precisely because the party structure of the conflict is irreducibly international. The “legal status of the opposing entities,” to repeat the language of the Commentary, drives the characterization of the conflict. Likewise, the “but clause” provides only a simplified, stylized example of a non-international party structure—without positing any exclusionary condition. For example, the “but clause” references conflicts between “a single State” and non-state armed factions, but it is beyond obvious that conflicts involving more than “a single state” are non-international so long as the sovereign states fight on the same side of the conflict.

If we had had the space or inclination to analyze whether the Commentary supports our ultimate position (and the ultimate conclusion of the Court), we could have done so. Indeed, the Commentary to Protocol II provides strong—and, in our view, overwhelming—support, along several axes, for the range of claims we make in the brief (at p. 1319, et seq. and 1351 et seq). The Commentary makes clear, again on the very page we cite (p. 1351), that “all armed conflicts which are not covered by Article 1 of Protocol I” are covered by Protocol II and/or Common Article 3. That is, there is no category of armed conflict (e.g., otherwise non-international armed conflicts which are trans-territorial) not covered by one of these provisions.

In sum, conflicts that exhibit a non-international party structure do not lose their non-international character simply because the fighting extends beyond the borders of a single state. The heart of our argument is that it is the legal status of the opposing parties that is crucial—by the plain language quoted from the Commentary and everything that follows therein. The territorial purview of the fighting has nothing to do with the legal status of the armed group, just as the territorial purview of the fighting has nothing to do with the legal status of sovereign states.

Milanovic is correct that we (and the Court) provide the wrong pincite for the quoted language. His further claims regarding the omission of language from that quote are not only uncharitable (in view of the full analysis of the brief and the opinion), but also misguided.