(published in Just Security, Feb. 1, 2018)
There has been an important discussion in the last couple of weeks over the legality of possible limited strikes, part of a so-called “bloody nose” strategy, by the United States against North Korea. The main issue of that debate has been quite conclusively determined: such a strike would not be legal. And as Victor Cha, the White House’s pick, until recently, to be ambassador to South Korea, laid out in an op-ed this week, it would be deeply misguided as a policy choice. There remains more to be said, however, about a couple of interesting and potentially important questions regarding Japan’s position in relation to such strikes. One, which has been raised but not fully answered in the discussion, relates to whether Japan has already somehow consented in advance to U.S. action in collective self-defense of Japan. The second, which has not been explicitly addressed at all, is whether Japan defines “armed attack” for purposes of triggering the right of self-defense differently than does the U.S.
To recap briefly, the debate was set off by an essay in Lawfare by two West Point professors, Army Lt. Col. Shane Reeves and Army Capt. Robert Lawless, arguing that limited strikes on North Korea would be lawful. Virtually all elements of their argument were quite persuasively demolished by Kevin Jon Heller in Opinio Juris, and Michael Schmitt and Ryan Goodman here in Just Security.
A central premise of one strand of the Reeves/Lawless argument was that the test-firing of a North Korean intercontinental ballistic missile (ICBM) into the territory of Japan could constitute an armed attack, triggering an American right to use force as an exercise of collective self-defense. They were dismissive of the position of the International Court of Justice (ICJ) that collective self-defense requires a request from and consent of the country under attack, suggesting that the 1960 Treaty of Mutual Cooperation and Security between the U.S. and Japan (the “U.S.-Japan Security Treaty”), “may provide a basis” for American action. They went on to note that, in their view, Japan was in any event very unlikely to oppose American strikes in defense of Japan.
Treaty-Based Pre-Authorized Consent to Military Action?
This first question, of how the U.S.-Japan Security Treaty might alter the analysis was picked up and examined in an essay in Just Security by Aurel Sari and Hitoshi Nasu. In particular, they analyzed whether Art. V of the treaty might provide a form of standing request for assistance from Japan in the event of an armed attack. In Art. V each party “recognizes that an armed attack on either party in the territories under the administration of Japan would be dangerous to its own peace and safety and declares that it would act to meet the common danger.” Sari and Nasu concluded, based on both the language of the treaty itself, as well as that of diplomatic notes exchanged in 1960 (not 1966, as stated in their essay – available here in Japanese), that Art. V could be interpreted to provide the U.S. with a standing right to take action as an exercise of collective self-defense, without any need to consult the Japanese government further. They also note that “it could have far reaching consequences if the Japanese and U.S. governments were to take different views regarding the legality or strategic wisdom of taking military action against North Korea.”
I agree that it would have far-reaching consequences, but I do not think that the treaty can be interpreted to mean that Japan has provided advance consent, such that the U.S. need not consult Japan before using force in response to an armed attack on Japan. Heller, anticipating the Sari and Nasu argument, pointed out that Art. IV of the U.S.-Japan Security Treaty provides that the parties will consult together “whenever the security of Japan or international peace and security in the Far East is threatened.” Sari and Nasu, however, refute this by pointing out that the 1960 diplomatic notes reflect an agreement to interpret Art. VI of the treaty as requiring cooperation and joint consultation, but made no such mention of Art. V. Thus, they argue that this suggests that the drafters meant that no such consultation is required in relation to Art. V. But Sari and Nasu surely know that the far more extensive and recent Guidelines for Japan-U.S. Defense Cooperation, which govern the implementation of the treaty, provide extensively for just such consultation, particularly in the event of an armed attack on Japan.
The most recently revised version of the Guidelines, from 2015, in Part IV, Section C, under the heading “Actions in Response to an Armed Attack against Japan,” provides that “bilateral actions in response to an armed attack against Japan remain a core aspect of Japan-U.S. security defense cooperation.” Under sub-section 2, titled “When an Armed Attack Against Japan Occurs,” the Guidelines provide in part that:
If an armed attack against Japan occurs despite diplomatic efforts and deterrence, Japan and the United States will cooperate to repel promptly the attack and deter any further attack to return peace and security to Japan. . . . Japan will maintain primary responsibility for defending the citizens and territory of Japan. . . . [t]he Self-Defense Forces will have the primary responsibility to conduct defensive operations in Japan and its surrounding waters and airspace, as well as its air and maritime approaches. The United States will coordinate closely with Japan and provide appropriate support.” [emphasis added]
I would argue that this undermines any argument that the U.S. is entitled by Art. V of the treaty to act without consulting Japan. At the very least, in light of this language in the Guidelines, it would take a lot more evidence than Sari and Nasu present to come close to concluding Japan gave up its sovereign right to determine whether the U.S. can use force in its name. Sari and Nasu try to bolster their evidence by resorting to the object and purpose of the treaty, but the Guidelines were negotiated to implement and realize the objects and purpose of the treaty. For their part, Reeves and Lawless assert the treaty as a basis for their conclusion without grappling with the complications at all.
Japanese Definition of Armed Attack
Turning to the second major question, Sari and Nasu do not deal with the issue of Japan’s declaration of armed attack. They explain that the ICJ position is that the conditions precedent to the legitimate use of force in the exercise of collective self-defense are that, first, the victim state itself must declare that it has been the subject of an armed attack, and second, that the victim state must request assistance. But while the essays of both Sari and Nasu, and Reeves and Lawless, suggest that the U.S.-Japan Security Treaty might somehow satisfy the second condition in advance, neither essay addresses the first condition. What happens if the U.S. characterizes an event as an armed attack, but Japan does not?
Schmitt and Goodman, in their detailed destruction of the Reeves/Lawless argument, explain that while the U.S. takes the anomalous view that any use of force constitutes an “armed attack,” most of the rest of the world accepts the ICJ’s position that there is a considerable gap in the scale and gravity of violence that constitutes an armed attack, and the threshold for what constitutes a use of force. The U.S. might view a test-missile fired into foreign territory as an armed attack, but the rest of the world would not. Heller, in his response, explained, relying on Tom Ruys’ seminal text on the subject, why an unarmed test missile fired into Japanese territory would not satisfy the widely accepted standard for what constitutes an armed attack. (Retired Maj. Gen. Charlie Dunlap joined the debate by pointing out that the Reeves and Lawless hypothetical might involve the test-missile landing in a populated territory rather than the Japanese territorial sea—but I do not think that changes the analysis).
But here is a key question: Is Japan with the U.S. on the definition of armed attack, or with the rest of the world?
Until relatively recently, the answer to that question would have been fairly straightforward: the rest of the world. Japan’s Cabinet Legislation Bureau (CLB), loosely analogous to the Office of Legal Counsel in the U.S. Department of Justice (and modeled on the French Conseil d’Êtat), has traditionally hewed to ICJ interpretations of international law (more on the CLB in English here). The answers to questions on issues of self-defense in the Diet (legislature) by successive prime ministers, foreign ministers, and defense ministers have all reflected a similar position (in Japanese, see Shigenobu Tamura, Japan’s National Security Legal System). But the situation has been complicated somewhat by the recent attempt by Prime Minister Shinzo Abe’s government to “reinterpret” Article 9–the war-renouncing provision–of the Constitution of Japan.
Paragraph one of Article 9 of the Japanese constitution renounces war and the threat or use of force. It incorporated language from the Kellogg-Briand Pact and Article 2(4) of the U.N. Charter. From soon after the return of Japan’s full sovereignty in 1952, Article 9 has been interpreted as meaning that while Japan may use the minimum force necessary for purposes of individual self-defense, it is prohibited from using force for either collective self-defense or for collective security operations authorized by the U.N. Security Council. It has operated for some 70 years to effectively constrain Japanese policy—Japan has not been involved in any armed conflict since the end of World War II.
In July, 2014, however, the Abe government purported to “reinterpret” Article 9 unilaterally via a Cabinet Decision. It then proposed revisions to a number of national security laws, which would have the effect of implementing the “reinterpretation.” The vast majority of constitutional scholars in Japan rejected the legitimacy of the “reinterpretation,” on both process and substance grounds, on which I’ve written elsewhere. It remains an open question whether the “reinterpretation” will become a de facto amendment. But for our purposes, the “reinterpretation” and revised national security laws have the potential to implicate Japan’s relationship with the jus ad bellumregime in the meantime. Without getting too deeply into the weeds (something I do more of in a shorter article on the “reinterpretation”), the Cabinet Decision, the revised national security legislation, and various statements made by cabinet ministers, have all served to blur the boundaries of when Japan may use force.
The Cabinet Decision itself purports to interpret Article 9 as permitting Japan to use force in response to infringements of Japanese sovereignty that do not amount to an armed attack. This contemplates, in particular, dealing with Chinese or Korean encroachments on disputed islands. It has given rise to a significant debate in Japan over the legality, under both Japanese and international law, of Japanese forces using force in so-called “gray zone” situations, which have been defined (rather anomalously) as constituting a state somewhere between that of armed conflict and that of peacetime. Nasu himself has written on the challenges that this domestic law conception of gray zone operations poses for Japanese compliance with jus ad bellum.
In addition, there have been statements by Prime Minister Abe and several other cabinet ministers that tend to confuse the view of how the government itself understands the conditions precedent for using force in self-defense. It has been suggested, for instance, that Japan could rely on the new conception of collective self-defense in the “reinterpretation” (which is defined somewhat differently from the jus ad bellum concept) to justify Japan engaging in minesweeping operations in Iranian territorial waters in the event that Iran attempted to block the Straits of Hormuz. While relying on the justification of collective self-defense for this position, there has been no effort to explain how the mining of the Straits would constitute an armed attack, on Japan or any other country (there are, of course, other international law grounds for such minesweeping, but these were never invoked). In short, the developments around the “reinterpretation” have created some uncertainty regarding the current government position on the criteria for the use of force, and even whether those criteria would comply with the jus ad bellum regime. Hence the complication in answering the key question.
Notwithstanding the extent of such uncertainty, however, there is recent evidence that Japan continues to define the concept of armed attack in line with the established ICJ interpretation. Indeed, there is evidence that relates precisely to an armed attack in the form of a missile strike on Japan. In the annual white paper on national defense, “Defense of Japan 2017,” in a section on responding to various threats (Part III, Chap. 1, Sec. 2), under the heading “Japan’s Ballistic Missile Defense,” a distinction is made between ballistic missile strikes that are determined to constitute an armed attack against Japan, and those that are not: “If ballistic missiles are launched towards Japan, and the situation is not acknowledged as an armed attack, the Minister of Defense can order the SDF units to take measures to destroy the ballistic missiles…”
The section similarly addresses attacks in various other forms, including even attacks with nuclear, biological, and chemical weapons, and distinguishes between those that constitute an armed attack, and those that do not. This would certainly suggest that, unlike the U.S., Japan continues to recognize and respect the existence of a difference in scale and gravity between a use of force in violation of Art. 2(4) of the Charter, and an armed attack triggering the right of self-defense. Indeed, this is consistent with one possible understanding of the “reinterpretation”, in that a limited response to “gray zone” situations characterized by a low-level use of force or intervention, can be distinguished from responses to true armed attacks in the form of a full-fledged use of force as authorized by Art. 51 of the U.N. Charter.
To zoom back out, the answer to these questions regarding Japan’s position are potentially quite significant. The argument that the U.S. can unilaterally conduct strikes against North Korea in reliance upon a collective self-defense argument whether by treaty or custom, loses its foundation if Japan neither recognizes some incident as comprising an armed attack, nor consents to, far less requests, American assistance in response. I tend to agree with the observations of Heller, Goodman, Schmitt, Sari and Nasu, that there are many situations in which Japan would not likely agree to American strikes in the name of defending Japan—Japan, after all, stands to suffer greatly in the event of an escalating conflict set off by limited strikes on North Korea. And the answers to these questions would suggest that the U.S. would be precluded from acting unless, and until, Japan acknowledged it had been victim of an armed attack, and requested assistance in responding to that attack.