This post is the latest installment of our “Monday Reflections” feature, in which a different Just Security editor examines the big stories from the previous week or looks ahead to key developments on the horizon.

Last month, the Prosecutor of the International Criminal Court (ICC) announced that the sole person accused in the Mali situation, Ahmad Al Faqi Al Mahdi, had expressed his intention to plead guilty to the charges before him. If the Trial Chamber accepts his plea, it will be the first time that an accused person has pleaded guilty at the ICC. How this first plea unfolds could affect the viability of future pleas at the ICC, and therefore both the prosecution and the judges should consider how best to begin developing an institutional practice that is conducive to guilty pleas.

First, a bit of background. As I wrote previously, Al Mahdi, a brigade commander in Ansar Dine in Mali (a militant Islamist group associated with al-Qaeda in the Islamic Maghreb), was charged and arrested last September for committing the war crime of intentionally attacking religious and historical monuments in Timbuktu in 2012. During Al Mahdi’s confirmation hearing, a proceeding to determine whether there is sufficient evidence to commit the case to trial, he indicated that he wished to plead guilty. “The testimony, the information provided, reflect the truth,” he declared. Al Mahdi’s guilty plea can be finalized only after it is entered before the Trial Chamber, which must determine first if the plea is informed and voluntary and if there exist sufficient facts to support it.

What is not yet clear is if Al Mahdi is pleading “straight up” or is doing so pursuant to a “plea agreement” with the Prosecutor. The accused certainly has strong incentives to plead guilty. He is on video committing and admitting to the crimes. But it also appears that although he will admit the crimes, he will, according to his lawyer, “explain himself and … defend himself as to the merits” and that there will be “no concessions.” At the same time, redactions in the transcript of the confirmation hearing around the discussion of the guilty plea suggest that there may be more to the story than we currently know, including an agreement with the prosecution regarding the admission of guilt.

Whether or not there is a plea deal with the prosecution, both the prosecution and the judges should reflect now on how to build a practice around admissions of guilt and plea agreements that creates a space for guilty pleas while advancing the interests of the Court and justice.

While in the US we may have debates about the pervasiveness of of plea bargaining, the concept itself is largely uncontroversial. Not so in the international criminal sphere, which draws together lawyers and practices from different justice systems around the world, many of which do not have plea bargaining or permit it only for low-level crimes. Judges and lawyers from these systems are more likely to view plea bargaining with suspicion, if not downright hostility.

And caution is certainly warranted. International criminal tribunals generally target senior commanders and leaders for war crimes, crimes against humanity, and genocide. Justice often requires the most thorough prosecution of these cases and the victims and affected communities expect a full accounting. But as I have argued elsewhere, admissions of guilt and plea bargaining can also help fulfill the goals of international criminal justice, particularly in light of the few resources and thin political support currently afforded the project. A clear admission of guilt by the accused can bring finality to the proceedings that is less subject to contestation by supporters of the accused, and that can (in some cases) provide important evidence for other cases. Victims may also derive more satisfaction from seeing a perpetrator accept responsibility than from enduring months and years of the defendant’s denials in trial.

And the reality is that plea bargaining is already an established feature of the international criminal justice system. It was deployed at the ad hoc tribunals — in particular at the International Criminal Tribunal for the former Yugoslavia (ICTY) and the International Tribunal for Rwanda (ICTR) — and is expressly permitted by the Rome Statute of the International Criminal Court.

But if plea bargaining is ever to succeed, a supportive practice must be developed at the Court, and the early cases will be critical in this respect. As is true in many common law countries, including the US, the guilty plea process at the ICC gives the judges the ultimate authority to accept a guilty plea and impose sentence. Therefore, the prosecution and defense can enter into a plea agreement, but they cannot bind the judges.

Because of this division of roles, there must be some shared understanding between the prosecution and defense on the one hand, and the judges on the other, about the proper scope of plea agreements and guilty pleas. If accused persons enter into plea agreements with the prosecution, they will need to have some confidence that they will be sentenced roughly in accord with the recommendation set forth in the agreement. Otherwise, they will have little incentive to reach an agreement. There is evidence that at the ad hoc tribunals, plea bargains slowed and even stopped after judges in some cases imposed sentences that far exceeded those recommended by the parties.

So how can a plea bargaining practice be developed? It starts with the prosecution. First, the prosecution should adopt a transparent policy with respect to admissions of guilt and plea bargains, as it has on other important subjects. The policy should set forth the factors that the office will consider when deciding whether to enter into a plea agreement, and how it will view issues like the bargaining of charges, sentence recommendations, cooperation, and the role and input of the victims. Such a policy will give the judges confidence that the prosecution is approaching this topic with thought and deliberation, and will allow them to assess how individual cases fit into the prosecution’s overall approach.

Second, the prosecution should proceed with care in its early cases. It should ensure that its agreements can be defended before the judges and the public, and should take time to explain its rationales for entering into any agreement. It is in the prosecution’s interest to develop trust in its approach to plea bargaining from the beginning, both with the judges and outside communities.

But the judges too have a responsibility. They should approach plea bargains voluntarily reached by the parties with considerable deference. Together the parties will have a greater familiarity with the facts and merits of the case, allowing them to make well-reasoned assessments of the culpability of the accused. The simple fact of an agreement between adversarial parties should give the judges some confidence in the terms of the deal. And the judges must accept that a plea agreement entails a compromise that brings benefits to the Court and the project of justice, but to the accused as well, who will ordinarily receive a lower sentence for accepting responsibility. Even if the judges believe that a different sentence might be warranted in a particular case, they should be inclined to give force to the terms of a plea agreement as long as they find them to be reasonable and defensible.

To be clear, a plea agreement is not appropriate in every case, and the judges should not simply rubber-stamp them. Nor will it be the case that even the most reasonable plea agreements will please everyone. There will always be critics. That comes with the territory. But there should be no question that plea agreements can advance the goals of the Court in some cases, and if plea agreements are ever to succeed, it will be essential that both the prosecution the judges are attentive to developing an institutional practice that is conducive to guilty pleas and creates a shared understanding of when they can move forward.