North Korea and the Law on Anticipatory Self-Defense

Written by

Media reports over the last few weeks indicate that the already tense relationship between North Korea and the United States is getting worse. Now that North Korea is nearly ready to test an intercontinental ballistic missile, the United States has said that it will get more confrontational. U.S. Secretary of State Rex Tillerson even suggested that U.S. military action against North Korea is “on the table.” Such talk is sometimes part of a broader strategy to pressure other countries to negotiate, whether at the Security Council or elsewhere. But it can also be a precursor to war. And it comes at an acute time for the law on anticipatory self-defense.

As readers of this blog no doubt know, Article 51 of the UN Charter recognizes that states have an “inherent” right to use force in self-defense “if an armed attack occurs.” There is an ongoing debate about whether and, if so, when Article 51 permits states to use force to avert an attack that has not yet occurred. Claims for interpreting Article 51 expansively—to permit defensive force even if the attack is only speculative—have been made with respect to “rogue” states that are developing nuclear weapons. In this post, I situate the North Korea case within that debate and explain why the United States might find it to be a particularly challenging case in which to press its expansive claim.

I. The Law on Anticipatory Self-Defense

A. A Restrictive Position

 The majority view on anticipatory self-defense is probably a restrictive one: that anticipatory self-defense can be lawful only if an attack is truly “imminent”—as in, about to occur. Under this view, states may not use force unilaterally to nip in the bud latent threats or attacks that are still conjectural. They must instead address those situations using non-forcible means or by obtaining the UN Security Council’s authorization.

That position finds support in various authoritative texts, including texts that specifically address situations involving the acquisition of nuclear weapons. In 1981, Israel bombed an Iraqi nuclear reactor that seemed ready to produce weapons-grade uranium. The UN Security Council “strongly condemn[ed]” the operation as a “clear violation of the Charter.” In 2004, the UN Secretary-General’s High-Level Panel on Threats, Challenges and Change likewise asserted that anticipatory self-defense is lawful “as long as the threatened attack is imminent” (para. 188). The Panel distinguished those cases from situations in which “the threat in question is not imminent but still claimed to be real: for example, the acquisition, with allegedly hostile intent, of nuclear weapons-making capability” (para. 188). It explained that non-imminent threats ought to be addressed without force or through the Security Council.

Indeed, the Security Council has repeatedly taken steps to prevent specific states from acquiring nuclear weapons. The Council has issued numerous resolutions condemning or authorizing sanctions against North Korea for conduct relating to its nuclear program. Likewise, the Council authorized sanctions against Iran, until it agreed to subject its program to stricter international oversight. And after the 1991 Gulf War, the Council imposed intrusive measures on Iraq to prevent it from obtaining nuclear weapons. This practice shapes expectations about the proper locus of decisionmaking authority. It suggests that the decision to sanction a state that might be acquiring nuclear weapons falls, at least in the first instance, to the Security Council, not to states acting unilaterally.

The debate surrounding the 2003 Iraq war arguably also supports that view. Part of the justification for the war was that the use of force was necessary to contain Iraq’s nuclear program. The United States claimed to be acting pursuant to the Security Council’s authorization. But most states disagreed with that claim and vociferously condemned the war. They did not believe that the Council authorized the use of force or that force without the Council’s authorization was justifiable in this instance.

B. A Permissive Position

However, the restrictive position on anticipatory self-defense is being contested. As Michael Reisman and Andrea Armstrong showed in their 2006 article, several states have expressly claimed the right to use force in anticipatory self-defense, without limiting that right to truly imminent attacks. In December 2016, the United States drew on Sir Daniel Bethlehem’s piece in the American Journal of International Law to articulate this legal position:

When considering whether an armed attack is imminent . . ., the United States analyzes a variety of factors. These factors include ‘the nature and immediacy of the threat; the probability of an attack; whether the anticipated attack is part of a concerted pattern of continuing armed activity; the likely scale of the attack and the injury, loss, or damage likely to result therefrom in the absence of mitigating action; and the likelihood that there will be other opportunities to undertake effective action in self-defense that may be expected to cause less serious collateral injury, loss, or damage.

Notice that “imminence” here does not have its ordinary meaning. By the United States’ account, it would permit defensive force to prevent attacks that are still fairly conjectural.

In January 2017, the United Kingdom endorsed a very similar legal position. U.K. Attorney-General Jeremy Wright purported to limit the most expansive implications of the U.K. claim. He underscored that “[i]t is absolutely not the position of the UK Government that armed force may be used to prevent a threat from materialising in the first place.” But his language just begs the question of when a threat has materialized. In the context of nuclear weapons, the threat could be said to materialize—and to justify defensive action—once a state that has demonstrated a hostile intent comes close to acquiring nuclear weapons. At that point, the risk of a potentially devastating attack increases. And waiting for the threat to become more operational might deprive the defending state of a meaningful opportunity to protect itself, without significant death or destruction. The U.S. and U.K. legal position thus creates space to justify anticipatory actions against states like North Korea.

To be sure, most states have not expressly endorsed that position. But as Jacob Katz Cogan and I have shown, most have also stayed silent in the face of actions that reflect it. Two examples are particularly relevant to the North Korea case. First, in 2007, Israel reportedly attacked a partially constructed nuclear facility in Syria. Although the UN Security Council condemned a very similar operation in 1981, states were almost completely silent about the 2007 action. Second, media reports indicate that Israel and the United States repeatedly attacked Iran’s nuclear program, as the Security Council was pressuring it to accept more international oversight. These attacks varied in their severity, but at least some of them caused physical destruction or death. Again, the response was muted.

My point here is not that the permissive position on anticipatory self-defense is the best articulation of the law. My point is that the law on anticipatory self-defense is potentially in flux. It might already be shifting or might soon shift from the restrictive position toward the more permissive one. Moreover, for the time being, a state that uses force in anticipatory self-defense might be able to calibrate its action such that it falls in a legal grey zone—in which it foregoes the legitimizing effect of having the law on its side but also avoids the verbal or material blowback of a violation. The operation would not be widely accepted as lawful, but neither would it be widely treated as unlawful.

II. The Prospects for U.S. Strikes against North Korea

Given that the United States has itself advanced the permissive position on anticipatory self-defense, North Korea could present something of a test case. The United States might try to exploit the legal grey zone or press for its position on the law. Those moves are unlikely to succeed for at least three reasons. First, using force to curtail North Korea’s nuclear program would be operationally difficult and present a serious risk of an escalation in violence. Antony Blinken, a State Department official in the Obama administration, recently explained:

Much of North Korea’s nuclear complex is concealed underground, inside mountains or in places unknown to United States intelligence. Meanwhile, the country is making rapid progress with mobile missiles powered by solid rocket fuel that can be rolled out of hiding and prepared for launch in minutes.

Moreover, as Max Fisher put it, “[a]lmost any plan would bring a high risk of unintended escalation to all-out war, analysts believe.” Thus, low-level, one-off operations—of the sort that were used against Syria and Iran—appear to be infeasible against North Korea. This matters from a legal perspective because other countries might have a harder time averting their gaze if the anticipatory action is a major military operation or risks triggering a broader war. Indeed, part of the justification for permitting anticipatory force in this context is that the operation mitigates the damage that would occur if the nuclear threat becomes more operational. That justification is less convincing if the anticipatory action itself causes or leads to enormous damage.

Second, the Trump administration has shown little interest in, and has at times been outright hostile toward, international law and international institutions. Those who have a stake in preserving these arrangements and who worry about their deterioration during a Trump administration thus have reason to push back against the United States—to resist an operation that they might otherwise tolerate because they view it as evincing a blatant disregard for the law. The 2003 Iraq war might be instructive. Once the United States indicated that it was prepared to go to war, no matter whether it obtained a contemporaneous Security Council resolution authorizing force, other states (both on and off the Council) were more intent on isolating the United States and demonstrating their own commitment to the Council’s primacy in this area.

Third, the United States might now try to rally other states to support or tolerate a defensive operation. But this would require a serious diplomatic effort. The United States would likely have to persuade its allies that the alternatives to defensive force—including action through the Security Council—are either inadequate or infeasible. At the moment, the United States seems ill-prepared to undertake this effort. The U.S. State Department is, by many accounts, in disarray and sidelined from U.S. foreign policy decisionmaking. Moreover, Secretary Tillerson reportedly has isolated himself from career officials who might otherwise guide him in developing an effective diplomatic strategy.

III. Where Does This Leave Us?

To be sure, the United States might still use force to try to curb North Korea’s nuclear threat. This scenario would try the jus ad bellum’s resilience and the United States’ commitment to it. Though the United States has violated the jus ad bellum before, it has also consistently engaged with and demonstrated its overall support for the regime. I argued last month that there are reasons to believe that the Trump administration will be different.

Alternatively, the United States might forego military action against North Korea. In this event, North Korea’s nuclear program would still present a serious security threat. And again, the United States would be, at least for the moment, poorly positioned to lead a broad-based, multilateral initiative to contain that threat. Unless other countries step into the breach, the situation on the Korean Peninsula is likely to deteriorate even further.

Print Friendly, PDF & Email


No tags available

Leave a Comment

Comments for this post are closed


Jordan Paust says

March 28, 2017

It is difficult to accept that the US, UK, and Israel determine the content of treaty and customary norms. The better view is that general patterns of opinio juris must be sufficiently identified. Further, the limit in the text of Article 51 seems unavoidable.
In any event, the war against North Korea has not ended, whether or not an armistice can be operative after North Korean violations. The law of war paradigm provides different claims for justification. In the context of this state of war, North Korean nuclear weapons and ICBMs could become proper military targets even if an armistice is operative because their existence, plus North Koreas constant threats, fundamentally changes the situation. During a state of war, the US and South Korea cannot wait for nuclear weapons to be launched.

Marty Lederman says

March 28, 2017

Great, important post, Monica. I would only quibble about two things, either or both of which might be important.

First, I don't think the examples you offer, of two recent statements and two recent actions, demonstrate that "the law on anticipatory self-defense is potentially in flux" and "might already be shifting or might soon shift from the restrictive position toward the more permissive one"--or at least, not sufficiently in flux to come anywhere close to suggesting that a unilateral U.S. strike against North Korea would be lawful. The recent US and UK statements, to be sure, mention that the "the likely scale of the attack and the injury, loss, or damage likely to result therefrom" are potentially relevant factors. But they do not suggest that such factors could be sufficient to justify a use of force against a state that has demonstrated no concrete plans to actually engage in an armed attack against the acting states--let alone to strike before the state even has the capability of attacking. And the Syria 2007 and Stuxnet cases do not reflect anything about the law at all, because those actions were covert--i.e., no states claimed responsibility for them. This fact, if anything, suggests that such actions *do* breach the Charter.

There's no evidence, that is to say, that any states, including the US and UK, have changed their views from what they articulated in the case of the Entebbe operation in 1981.

Second, even if these statements and actions were less equivocal and more reflective of changing US/UK views than I think they are, I don't see how the statements and actions of a handful of states might change the meaning of Article 51 in a more permissive direction. Of course, whether Art. 51 is subject to "customary" development is part of a much broader, longstanding debate that I can't possibly address fully here.

Finally, just one other note of possible significance: In addition to the Charter questions, there's a serious question whether a strike of the sort you describe could be ordered by the President, without congressional authorization, *as a matter of U.S. domestic law.*

Jordan J Paust says

March 28, 2017

Marty: --re the US Const, -- with respect to the ongoing war, the President is, of course, C-I-C and can order the targetings. With respect to the self-defense paradigm, the President has a duty faithfully to execute the laws, which would include the US competence to engage in self-defense once an armed attack has begun per UN Charter art. 51.

Dapo Akande says

March 28, 2017

Dear Jordan,

In your two comments above, you seem to imply that a justification for the use of force against North Korea could be found in the claim that the US is still at war with North Korea, since 1950, despite the armistice to terminate hostilities in that war. In my view such a claim is legally incorrect (and of course dangerous). I wrote a post on this issue in July 2009 - The Korean War has Resumed !! (Or so we are told) where I argued that:

" . . . since the adoption of the UN Charter an indefinite armistice and a general close of military operations terminates hostilities and brings to an end rights which exist under the law of armed conflict (apart from cases where IHL applies beyond the termination of hostilities). For those rights to be resumed there will, in general, need to be a resumption of an armed conflict. This in turns requires the actual use of force by armed forces. A mere verbal termination of an armistice does not mean that there is an armed conflict. And of course, an actual use of force by one State against another must be justified under the jus ad bellum."

As I pointed out then, the view I expressed is supported by Security Council practice (or to be more accurate the practice of states in the Security Council, which is not contested by practice outside of it). See particularly, Resolution 95 of 1951, where the Security Council stated that:

“. . . since the armistice regime, which has been in existence for nearly two and a half years, is of a permanent character, neither party can reasonably assert that it is actively a belligerent or requires to exercise the right of visit, search and seizure for any legitimate purpose of self-defence."

Jordan J Paust says

March 28, 2017

Dapo: a lot has happened since 1951, which brings into question the continued existence of an old armistice. But since an armistice does not end a war, would it end an armed conflict as such as opposed to actual "hostilities"?
Dangerousness is a third concern that, in view of context and relevant policies at stake, may cut both ways (e.g., it is dangerous to allow North Korea to threaten the use of force, in violation of Article 2(4), and to acquire ICBMs armed with nuclear warheads).

Dapo Akande says

March 28, 2017

Dear Jordan,

Even if an armistice does not end a war, the fundamental legal change that occurred in 1945 means that a state that uses force on the territory of another state needs to justify that use of force under the jus ad bellum. The fact that there was or even is a war is not such a justification under the jus ad bellum and specifically under the UN Charter. Otherwise, that state acts in violation of the prohibition of the use of force under Art. 2(4). The jus in bello does not in and of itself justify a resort to force. Once hostilities have ceased, the state wishing to resume them will need to rely either on self-defence or a Security Council resolution. Hopefully, the US does not take the view that the resolution authorising force in 1950 still authorises a use of force today. If not, then we are back to self-defence and to the issues that Monica discusses.

Jordan Paust says

March 29, 2017

You recognize one possible claim: the UN SC authorization never ended (coupled with the point that the Korean War never ended).
Further, Article 2(4)'s text only prohibits three types of threats and uses of force.
In any event, isn't this an interesting situation?

Monica Hakimi says

March 29, 2017

Thanks for all of these comments.

Dapo and Jordan have largely exhausted the question of whether the United States may attack North Korea on the ground that their war has never ended. I think Dapo has the better argument here, but I don't have much to add to his comments.

Marty asks a fair question -- one that I intended to leave open in my post: is the law on anticipatory self-defense changing? I'm willing to say only that it might be changing and that, for now, there seems to be a "grey zone" within which some anticipatory actions are neither widely accepted as lawful nor treated as unlawful.

Marty seems unwilling to go even that far. He says that the forcible actions against Syria and Iran are irrelevant to the law on self-defense because they were taken covertly. If anything, he says, their covert nature suggests that they were illegal.

I disagree that those actions are irrelevant to the law. First, they received considerable media coverage and were widely attributed to Israel and the United States. (There is even some reason to believe that one or both of those states purposefully leaked information about the operations to the press.) Although Israel and the United States have not publicly justified their specific operations against Syria and/or Iran, both states have advanced expansive claims on anticipatory self-defense. So, it's not clear that they consider the operations to be unlawful.

Second and related, Marty sees no evidence that the US/UK position on anticipatory self-defense would extend to a case like North Korea -- i.e., to permit defensive force against a state that is developing nuclear weapons but has not yet demonstrated concrete plans to engage in an armed attack. I'm not sure Marty is correct. As I said in my post, the US/UK claim certainly creates room to justify a defensive action in this kind of case. Moreover, as Reisman and Armstrong showed in their article, expansive claims on anticipatory self-defense are often made in the context of WMD. Here, for example, is a 2002 US presidential statement: "Because deterrence may not succeed, and because of the potentially devastating consequences of WMD use against our forces and civilian populations, U.S. military forces and appropriate civilian agencies must have the capability to defend themselves against WMD-armed adversaries, including in appropriate cases through preemptive measures." Again, I am not contending that the US/UK position would clearly permit defensive force against North Korea. But it's wrong to suggest that the US/UK position is clearly more constrained.

Third, the Syria and Iran incidents are legally relevant because they demonstrate that states, on the whole, are willing to tolerate some anticipatory actions that they are not (yet) willing to justify with generalizable legal rules. In other words, states have chosen not to apply or enforce what, in their view, might be the best position on the law.

Finally, that dynamic might (but does not necessarily) portend a change in the law. A similar dynamic was evident about 10-15 years ago in the law on defensive force against non-state actors. There, a few states advanced legal claims that were more permissive than the mainstream view. These states repeatedly acted consistently with their own claims. In many early cases, they acted covertly and avoided repercussion. And over time, the law has clearly shifted in their favor -- though it still might not be as permissive as they contend.

Jordan Paust says

March 29, 2017

Monica: quoting the Bush doctrine on preemptive self-defense, which was widely discredited in the literature and by the SC when raised earlier by Israel, as opposed to minority views on anticipatory self-defense, surely triggers Dapo's scary test. If that becomes the Trump doctrine, .....

Marty Lederman says

March 29, 2017

Monica: Thanks so much for that thoughtful response. I agree with much of it.

1. In particular, I agree that the *Bush Administration* was clearly trying to assert such a right of "purely," strongly anticipatory use of force in 2002, in the NSS and the OLC opinion on Iraq. My point however, is that the Obama Administration did not come anywhere close to that--indeed, as I read what it wrote and said, it was an implicit and important rebuke of the unjustifiable Bush theory. See my post at

2. I agree that the Syria and Stuxnet incidents demonstrate that (acting and unaffected) states are *sometimes* "willing to tolerate some anticipatory actions that they are not (yet) willing to justify with generalizable legal rules," and "not to apply or enforce what, in their view, might be the best position on the law." However, the fact that those actions, if the reports are correct, were covert--and that no states on Earth made any argument in support of their legality--demonstrates to me a broad understanding that they likely were *not* legal under the Charter. (One place where I might have a slight disagreement with you is that I don't think the fact that "they received considerable media coverage and were widely attributed to Israel and the United States" bears on their legality--and certainly I don't think it supports their legality.)

The only place where I have a fundamental disagreement is on your last point--the purported analogy to whether the Art. 51 "inherent" right of self-defense can be triggered by armed attacks by nonstate actors that are in no way attributable to the host state. As you probably know, I am very dubious of the oft-repeated assumption that "the mainstream view" was that such attacks fell outside Article 51. Of course, many *publicists* and academics said so in the period between, say, 1960 and 1980. But virtually no states did so. And I think the view expressly or implicitly embraced by many states lately (e.g., w/r/t use of force in Syria against ISIL) is fully consistent with the "inherent" and historic right of self-defence in Article 51, going back at least to the Caroline incident itself. That is to say, I resist the idea that the Charter once prohibited the use of force against such NSAs, but that the rule was (or could be) altered by subsequent custom. In fact, attacks by such NSAs were always encompassed by Article 51. As far as I know, no state has argued otherwise (e.g., no state has argued that what 2(4) once prohibited it no longer does).

Adil Haque says

March 30, 2017

Hi folks,

Thanks for a stimulating discussion. I don't have a substantive comment, but I thought I'd quickly reframe the issue in a way that new readers (particularly students) might find helpful.

No State has claimed a legal right to use preventive force in violation of the UN Charter. Therefore, what's at issue is simply the interpretation of the Charter itself. According to ordinary rules of treaty interpretation, we should ask:

1. Did the Charter *always* permit preventive force?

That would be surprising.

2. Has there been a "subsequent agreement between the parties regarding the interpretation of the treaty" permitting preventive force?

Evidently not.

3. Has there been "subsequent practice in the application of the treaty which establishes the agreement of the parties regarding its interpretation" permitting preventive force?

Nope. There may have been subsequent practice, but none that establishes agreement.

4. Have there been changes to other "relevant rules of international law applicable in the relations between the parties"? Has the customary jus ad bellum changed in ways that should now inform the interpretation of the Charter?

Like Marty, I find this possibility somewhat baffling, since it seems to require violations of the Charter to generate new customary law to then inform the interpretation of the Charter.

That's it. I hope some readers find it helpful.

All the best,


Tamás Hoffmann says

April 1, 2017

I find it quite interesting though not at all surprising that Prof. Hakimi only distinguishes between two approaches: a "restrictive" and a "permissive" position on anticipatory self-defence, both accepting the legality of anticipatory self-defence.

This view seems to be a majority position among Anglo-Saxon international lawyers, however, it seems disingenious that this post does not even consider the possibility - which actually seems to be the majority position among the non-Anglo-Saxon international lawyers - that anticipatory self-defence is not part of the lawful use of force. The leading authorities on the use of force such as Ian Brownlie, Christine Gray, Yoram Dinstein or Olivier Corten have all rejected the legality of anticipatory self-defence (though Dinstein introduced the concept of intercepting self-defence) and so did the majority of states as well. The concept does not really seem to be compatible with the plain text of Art. 51 ("if an armed attack occurs", "en cas d'un aggression armée") and while it was indeed referred to in the High-Level Panel Report, interestingly enough it was not included in the 2005 World Summit Outcome (a fact that the post somehow fails to mention).

All in all, it is quite puzzling that the author starts from the assumption that anticipatory self-defence is legal given the overwhelming contrary evidence. It follows the grand old tradition of American hegemonic international law which somehow presumes that the practice of the US (and a few allies) automatically translates into international law.

Jordan Paust says

April 3, 2017

Tamas: you are correct.
Yet, despite even viable legal claims, we are likely to see the massive destruction of the military capacity of North Korea, massive even if it only starts with targetings of nuclear weapons sites and support facilities (because a predictable response must be met immediately with widespread desyruction of military capacility). Apparently on China can ultimately change this outcome, but can they??