In a surprise decision, a Pre-Trial Chamber at the International Criminal Court (ICC) has rejected the Prosecutor’s request, filed nearly 18 months ago, to open an investigation of war crimes and crimes against humanity in Afghanistan, including of allegations that U.S. forces and the CIA committed acts of torture there. In light of recent threats from both National Security Advisor John Bolton and Secretary of State Mike Pompeo to retaliate against the ICC if such an investigation proceeded, and the U.S government’s recent cancellation of the ICC Prosecutor’s visa to travel to the U.S., there is already a perception that the Court caved to U.S. pressure. An examination of the larger context of the case, however, and the ICC’s current challenges, suggests that this view is incomplete. In fact, this decision will likely come to be seen as the beginning of a broader effort by the judges and the Prosecutor to orient the Court’s very limited resources toward those investigations where there exists some meaningful prospect of success.

The Prosecutor’s Afghanistan request was before the judges because she sought to open the investigation on her own authority, or proprio motu. When a State Party to the Rome Statute or the Security Council refers a case to the ICC for investigation, the Prosecutor can open it without judicial approval. However, the Statute imposes the extra obligation of judicial authorization when the Prosecutor seeks to commence an investigation without such a referral, as an extra check on the Prosecutor’s exercise of discretion.

In their decision, the judges found that information (which is not necessarily the same as evidence) exists that war crimes and/or crimes against humanity were committed by the Taliban, Afghan government forces, and U.S. forces on the territory of a state party (Afghanistan) after May 1, 2003, the date of entrance into force of the Rome Statute following Afghanistan’s decision to join the Court. Further, the judges found that there exist no national proceedings concerning these alleged crimes, which would require the Court to stand down and defer. All of these findings point to a decision authorizing an investigation.

However, the Rome Statute also requires the judges, as a final step, to determine whether, “taking into account the gravity of the crime and the interests of victims, there are nonetheless substantial reasons to believe that an investigation would not serve the interests of justice.” And it is here, on this basis, that the judges decided to refuse authorization.

Ordinarily, this final provision has given neither the Prosecutor nor the judges much pause. The Prosecutor has explicitly stated that the political or diplomatic ramifications of a decision to investigate or prosecute would not be considered, and the jurisprudence has suggested that the “interests of justice” provision is narrowly designed to address “exceptional circumstances,” to use the language of the Prosecutor, where an investigation would be unwarranted.

In Friday’s decision, however, the judges have given the term new scope and meaning. They decided that the interests of justice also require them to assess whether the proposed investigation has any prospect of success:

An investigation can hardly be said to be in the interests of justice if the relevant circumstances are such as to make such investigation not feasible and inevitably doomed to failure. In making any investigation or prosecution only worth doing if and to the extent that it can be considered as genuinely instrumental to those objectives, the Statute reiterates the idea that the Court is not meant – or equipped – to address any and all scenarios where the most serious international crimes might have been committed; therefore, focusing on those scenarios where the prospects for successful and meaningful investigations are serious and substantive is key to its ultimate success.

With respect to Afghanistan, the judges found strong evidence to indicate that for the moment at least, an investigation there simply could not succeed. The judges noted that the Prosecutor waited eleven years before even requesting an investigation, largely because of difficulties in gaining cooperation for its preliminary assessments, and that a lack of both cooperation and security persist today:

suffice it to say that nothing in the present conjuncture gives any reason to believe such cooperation can be taken for granted. Indeed, the Prosecution acknowledges the difficulties in securing albeit minimal cooperation from the relevant authorities as one of the reasons explaining the unusual duration of the preliminary examination. The Chamber has noted the Prosecution’s submissions to the effect that even neutral, low-impact activities proved unfeasible. Accordingly, it seems reasonable to assume that these difficulties will prove even trickier in the context of an investigation proper.

Finally, launching an investigation that was likely to fail would come at a high cost. It would consume substantial resources of the Court, which are already extremely limited and stretched thin, and it would raise the expectations of victims. Their inevitable frustration when the investigation foundered would only undermine confidence in the Court generally.

Commentators are already suggesting that the Court gave into U.S. pressure, and certainly it would be naïve to suggest that the U.S.’s overtly hostile stance played no role in this outcome. But there is a bigger picture here, and that is that there is a growing consensus that the ICC is stretched too thin and that it should consider ways to focus its energies and resources in places where it could succeed. That kind of calculation is a familiar one to prosecutors who have to exercise discretion in allocating scarce resources in selecting which cases they pursue, and it is a calculation hereby exercised by the ICC judges themselves.

The ICC presently has eleven investigations open in ten different countries (there are two in Central African Republic), though of course some are more active than others. Thrice before the Prosecutor has opened investigations relying on her own authority, in Kenya, Georgia, and Burundi. In Kenya the Prosecutor brought charges against six individuals, and all of the cases failed. In Georgia, the ICC had a preliminary examination open for eight years and has had a formal investigation open for more than three. Not a single case has been brought to date. The Burundi investigation was opened last year, but meanwhile Burundi has left the Court, and the prospects of success there are dim. Two situations were referred to the ICC by the UN Security Council, Sudan and Libya. Neither investigation received much support from the country or region at issue, nor from the Security Council itself. Not a single case from those two investigations has succeeded in The Hague.

The lesson here is that international criminal investigations are likely to succeed only if there exists support either from the country at issue, or from other states or relevant actors that can ensure that cooperation will be forthcoming. These are practical realities that courts and prosecutors often face. The ICC judges grappled with these realities more openly than we’re often accustomed. The International Criminal Tribunal for the former Yugoslavia succeeded in large part because both the U.S. and the European Union exerted considerable pressure on both Croatia and Serbia to cooperate with the tribunal. Moreover, successful investigations likely require sustained focus for a prolonged period of time, allowing the court to bring smaller cases to start, building to more high-level prosecutions. That approach has been impossible because the Court is presently stretched across so many situations. As a result, cases from the DRC, the Central African Republic, and Côte d’Ivoire have failed, some in part at least because of thin investigations by the Prosecution. The Court’s overall record, with only three convictions for core international crimes in 17 years, represents a serious and urgent challenge to its legitimacy.

These realities have caused ICC actors and observers to consider how the Court might concentrate its work to produce more successful investigations and prosecutions. When the second Prosecutor, Fatou Bensouda, took over in 2012, she announced that her office would conduct more in-depth investigations, and in 2016 she published a policy paper stating that in making decisions about case prioritization, she would consider factors such the available “international cooperation and judicial assistance” and “the Office’s capacity to effectively conduct the necessary investigations within a reasonable period of time, including the security situation in the area where the Office is planning to operate.” In other words, the Office of the Prosecutor has accepted that it must consider the reality of the circumstances on the ground, and the prospect of success, in making choices about where to conduct its investigations.

Now, in this Afghanistan decision, the judges have joined this discussion about the Court’s future and how to reorient it toward successful investigations and prosecutions. What is critical to this decision is that while the U.S. was the most vocal in its opposition to the Court’s investigation, none of the potential targets were supportive of the Court’s efforts. The vast majority of the crimes alleged in the Prosecutor’s request to investigate were committed by the Taliban first and Afghan government forces second; neither was cooperating with the ICC’s investigation. What’s more, the security situation in Afghanistan continues to impair the Court’s ability to operate on the ground. Few observers contend that an ICC investigation in Afghanistan would be successful, though even trying to mount an investigation would require significant resources. While the belligerent stance of the U.S. is part of this picture of non-cooperation, and likely figured in the judges’ decision, a larger concern about sending the Court on yet another failed mission no doubt also weighed.

But we don’t even know that for sure – whether the U.S. threats against the Court and its staff helped drive this decision. The judges may have even considered the perception of caving to those threats a significant cost to the Court and not a path they would otherwise take. These other pragmatic concerns involving the futility and drain of resources to proceed with investigations appear to have prevailed.

The judges’ decision raises a lot of difficult institutional and legal questions. If it is true that the Court can only succeed where it has international support, doesn’t that mean that it will often be unable to proceed against powerful states, whether they are States Parties or non-States Parties? Probably yes. But that’s a reality that the judges did not create by their decision, but merely acknowledged. And as a fledgling institution, it might be necessary for the Court to build and establish itself with small steps, focusing on those situations where international support will allow it to succeed. Opening up investigations against powerful actors or in those situations where there is lacking international support might produce short-term legitimacy gains – signaling that justice will be applied equally – but at very high long-term legitimacy costs when the Court ultimately fails.

To be sure, there are questions about whether under the ICC Statute and prudentially, the judges should be making these discretionary decisions or whether they should be left to the Prosecutor. I argued previously that the judges could authorize the Afghanistan investigation and then the Prosecutor could de-prioritize parts or all of it if she determined that success was unlikely. There are strong arguments for why this would be a preferable course, and if the Prosecutor is able to appeal this decision and chooses to do so, that might be an alternative outcome. But for the moment the important point is that while it will be tempting to view the judges’ decision as simply a bend to the bullying power of the U.S., there is a larger and more complicated story here about charting a successful path for the ICC going forward.