A panel of the DC Circuit recently held oral arguments in the case of Abd al Rahim al-Nashiri — a Saudi man accused of involvement in numerous terrorist plots and attacks against western ships, including the American destroyer USS Cole. Much of the discussion turned on the question of whether the United States and al-Qaeda were engaged in an armed conflict during al-Nashiri’s alleged conduct: the January 2000 attempted bombing of the USS The Sullivans; the October 2000 bombing of the USS Cole, most notably; and the 2002 bombing of the M/V Limburg.

In al-Nashiri, the defense seeks a finding that there was no armed conflict between the United States and al-Qaeda at the time of the Cole bombing; such a finding would preclude him being tried for that attack (or the attempt against The Sullivans) before a military commission. The reason is straightforward: Military commissions only have jurisdiction to prosecute violations of the law of armed conflict (LOAC), and the LOAC only applies during an armed conflict. Therefore, the military commission can only prosecute al-Nashiri if his alleged crimes were committed during an armed conflict. A determination that there was no armed conflict — and thus no military commission jurisdiction — when the USS Cole was bombed does not mean that al-Nashiri cannot be prosecuted for his alleged crimes; it simply means that he would have to be prosecuted in federal criminal court rather than by military commission.

Al-Nashiri has pressed this argument before, but neither of his previous attempts (the first before the military commission at the pre-trial stage, the second in the federal district court in the Western District of Washington) has been successful. Both courts declined to consider whether an armed conflict existed during the relevant time, finding instead that the question of whether and when an armed conflict existed is a factual one for the merits phase of al-Nashiri’s military commission trial.

However, these most recent oral arguments before the DC Circuit panel suggest that a US court may be considering, for the first time, whether the United States and al-Qaeda were in an armed conflict before 9/11. The question goes to the very heart of the debates about definitions and categories that underlie much of the post-9/11 legal discourse and challenges.

We tackle this question and address the surrounding debates in a recent article in the Michigan Journal of International Law. Although the Geneva Conventions and the associated International Committee of the Red Cross (ICRC) Commentary define the start of international armed conflict (IAC) — a conflict between two states — and provide guidance on identifying the existence of a non-international armed conflict (NIAC) — a conflict between a state and non-state group or between non-state groups — neither offers a methodology for isolating the start of NIAC. Rarely has a precise date for the start of a NIAC been articulated; rather, the predominant legal analysis generally focuses on identifying the existence of a NIAC itself during a given timeframe. As al-Nashiri’s petition demonstrates, however, identifying the start of this NIAC is essential to setting the temporal boundaries for conduct that can or cannot be charged in a military commission.

To explore this question, which until now has been seldom (if ever) addressed, we pose and examine three possible analytical options for identifying the start of a NIAC between a state and a non-state armed group:

1) The now-predominant Tadić analysis that looks primarily at intensity and organization to find the existence of a conflict;

2) The non-state armed group’s attack as the start of the conflict; or

3) The state’s declaration of the existence of a conflict as the triggering event.

The latter two options introduce, in essence, a unilateral trigger for a NIAC that is not part of any existing framework for the recognition of NIAC. Such a unilateral trigger also runs counter to the purposes behind the NIAC framework and threshold. In contrast, the Tadić formulation — protracted armed conflict between a state and an organized armed group, as identified by the intensity of the violence and the organization of the parties — matches well with the purposes of Common Article 3 and the intentions of the drafters of the Geneva Conventions in extending the protective and regulatory features of the LOAC to NIACs.

Examining the public record of hostilities between the United States and al-Qaeda through the lens of the Tadić approach, we argue that there was no NIAC between the United States and al-Qaeda before September 11, 2001. Indeed, we argue that the US/al-Qaeda NIAC actually began no earlier than late September 2001, when US forces in Afghanistan began engaging al-Qaeda and Taliban fighters, or October 7, 2001, the day US forces began bombing Afghanistan. In essence, after a number of isolated terrorist attacks, to which the United States responded predominantly with law enforcement measures, the United States eventually resorted to extensive military operations to defeat al-Qaeda and end its threat to the United States after the most egregious and consequential attack. As a result, neither 9/11 or the earlier terrorist attacks on their own could trigger the NIAC in the absence of this resort to prolonged military force by the United States in response.

The NIAC conflict recognition paradigm is built on the threshold that separates armed conflict from riots, banditry, and other acts of violence that do not rise to the level of armed conflict. The methodology focuses on the organization of the parties — do they manifest some degree of organization so as to be able to engage in an armed conflict and be identifiable as a party to the conflict? — and the intensity of hostilities between them — is the violence at a level beyond ordinary disturbances, leading to the use of the military, civilian casualties, displaced persons, or the involvement of the international community?

Based on the analytical formulation used by international tribunals, the ICRC, national courts and governments — including the Department of Defense Law of War Manual — the sporadic violence between al-Qaeda and the United States did not cross the NIAC threshold before September 11, 2001. Although al-Qaeda likely manifested a sufficient degree of organization by the mid 1990s to be a party to an armed conflict, the level of violence simply did not approach the level of intensity generally understood to trigger a non-international armed conflict.

Before 9/11, there were only three violent incidents between the United States and al-Qaeda, occurring on three separate days spread out over two years. During this time, few, if any, of the indicators of a NIAC highlighted in international jurisprudence and the Commentary were present, such as: the spread and increase in armed clashes between the two sides, increase in government troops and use of heavy weapons, increase in destruction caused and numbers of casualties, civilians fleeing from combat zones, attempts to take and hold territory, efforts to conclude ceasefires. Al-Qaeda did repeatedly declare war on the United States, but non-state groups’ declarations of war have no effect in international law.

Moreover, the United States relied primarily on law enforcement tools rather than military tools in its response to these al-Qaeda attacks. With the exception of the cruise missiles launched at al-Qaeda training camps and an al-Qaeda-linked pharmaceutical plant after the 1998 embassy bombings, the United States did not otherwise engage al-Qaeda with military force or deploy regular military units until the weeks after 9/11. The United States sent civilian law enforcement agents to East Africa after the embassy bombings and to Yemen in the wake of the Cole bombing, and it indicted the alleged perpetrators of those attacks in civilian courts. In addition, the United States characterized itself as at peace throughout that period, including immediately after the Cole bombing.

The 9/11 attacks added significantly to the level of intensity and, for many, changed the calculus entirely. There is no doubting the seriousness of al-Qaeda’s attacks, which in total saw more than 3,000 people killed. But, when analyzed in accordance with the well-recognized paradigm for identifying a NIAC, the instances of violence were too infrequent and separated by quiescence too long and too established to constitute an armed conflict. Only after the United States inserted special forces into northern Afghanistan and then began bombing Afghanistan did US and al-Qaeda (and al-Qaeda-affiliated) fighters come into the sort of regular and protracted clashes that observers normally associate with an armed conflict. In our article, we examine in depth the harder question of whether the 9/11 attacks themselves started the conflict, but for the purposes of al-Nashiri’s case and his pre-9/11 conduct, that question is irrelevant.

Interestingly, the United States has consistently charged military commission defendants for crimes based on conduct stretching as far back as 1996, but it has not clearly articulated a precise date for the start of the armed conflict with al-Qaeda. The US prosecutorial position suggests that the United States is employing an entirely different conflict identification methodology — notwithstanding the United States’ apparent endorsement of Tadić and its progeny in the Law of War Manual and even military commission jury instructions.

The two most likely options for the government’s calculus, as noted above, involve a focus on the actions or declarations of only one of the parties in a situation of violence, essentially vesting one party with the ability to unilaterally trigger a NIAC. For example, one alternative conflict-identification methodology would enable states to trigger an armed conflict by characterizing a situation as an armed conflict — essentially invoking the authorities found in the LOAC simply by declaring itself to be “at war” with a non-state armed group. Or, a non-state armed group could trigger an armed conflict simply by attacking a state in a sufficiently severe or serious manner, even if the state does not respond with military force. However, both of these alternatives pose grave danger to the stability of the international system and the protection of individuals in situations of violence.

Past and current non-international armed conflicts demonstrate an extraordinary amount of nuance in the way NIACs develop. Importantly, however, the reluctance to recognize an armed conflict upon a single non-state group attack or a state’s unilateral pronouncement has an essential, protective purpose. A world in which a single anti-government violent act triggers an armed conflict and, with it, the government’s right under the LOAC to use force as a first resort against all persons it identifies as “the enemy” is a frightening prospect. Such a framework is a recipe for escalating violence and overly expansive government powers neither foreseen nor desired by the drafters of the Geneva Conventions. It is a framework that diminishes states’ responsibilities to protect human rights even while taking action to maintain law and order by turning any violence or disturbance into war.

Wrapping pre-9/11 events into the ongoing conflict with al-Qaeda may be an “easy” way to enlarge military commission jurisdiction and opportunities for military prosecution. But the comprehensive legal framework for conflict recognition rejects this “easy” answer of unilateral non-state or state action. Instead, it demands a nuanced and principled analysis of the totality of the circumstances to ensure that conflict recognition operates in accordance with the core purposes of the LOAC.

The opinions and views expressed are those of the authors alone. They do not represent the views of the US Department of State or the US Government.