Recent Posts

Welcome to the Blogosphere, Lawfire!

by Kevin Jon Heller

Apparently, being named Charles and having vast military experience is all the rage in the blogosphere these days. Last week I mentioned Charles Blanchard’s new blog. And this week I want to spruik Charles Dunlop’s new(ish) blog, Lawfire. Charlie is a retired Major General in the US Air Force (where he served, inter alia, as Deputy Judge Advocate General) and currently serves as Executive Director of Duke Law School’s excellent Center on Law, Ethics and National Security. He is also Professor of Practice at Duke. His bio is here.

Charlie’s blog has been around for about two years. Recent posts discuss the relevance of social justice to the encryption debate, defend prioritizing victims of genocide in US immigration policy, and claim that Chelsea Manning’s commutation is actually likely to harm transgender soldiers.

I often disagree with Charlie about national-security and IHL issues. (I’m on Adil Haque’s side, for example, in the fantastic Just Security debate he and Charlie had last year concerning the new Law of War Manual’s treatment of human shields.) But Charlie’s blogging is unfailingly serious, thoughtful, and informative. If you haven’t already, you should add Lawfire to your newsreader.

You can find Lawfire here.

African Withdrawals Mask the Real Issue at the Assembly of States Parties 15

by Matt Brown

[Matt Brown holds an LL.M in Public International Law from Leiden University and is currently a Defence Intern for Jovica Stanišić before the MICT and worked as an intern for the Coalition for the International Criminal Court during the Assembly of States Parties. The views expressed here do not necessarily reflect those of the CICC, and he held no involvement in the budget negotiating process.]

The backdrop to the 15th Assembly of States Parties (ASP) has been dominated by the announced intent to withdraw from the International Criminal Court (ICC) by South Africa, Burundi and The Gambia. Prior to the ASP, there was almost exclusive attention on the withdrawal issue and whether discussion during the ASP would pave the way towards a reversal of these decisions to leave the Court. The importance of these withdrawals, and the need to evaluate the legitimacy of the arguments put forward by these States, cannot be underestimated, but almost under the radar, an issue of arguably equal importance has failed to garner the same detail of scrutiny and reaction. The financial strangulation of the Court by States Parties is if anything a far greater threat to the Court’s ability to fulfill its mandate than the withdrawals of three states.

To provide some context, the proposed budget put forward by the Court for 2016 was €153,328,200, (p187) which ultimately became an approved budget of €139,960,600* (p14). This demonstrates the gulf in stance, between the organs of the ICC and States, who although vocal in their support for the Court at the ASP podium, become remarkably restrained in contributing financially to the success of the institution. An initiative by eleven States, including Canada, Colombia France, Germany, Italy, Japan, and the United Kingdom to limit the Court’s budget, reflects how the Court’s biggest financial contributors are seeking to restrict the ability of the Court to expand its operations – cynically one might say into more politically sensitive situations, including Afghanistan.

This ‘dragging of the heels’ is likewise seen in comments by the Committee on Budget and Finance on 28 October who ‘noted with concern the large amount of outstanding contributions’ – €17.88 million (12.73 per cent) of the 2016 approved budget. In addition to that, outstanding contributions from previous years stood at €15.95 million, meaning ‘total outstanding contributions, including the regular budget, the Contingency Fund and interest on the host State loan, stood at €34.16 million as of 15 September 2016.’

A similar pattern has emerged this year, the Court for the 2017 budget proposed a figure of €150,238,000, (p7) but ultimately the approved budget came to only €144,587,300 (p1). Although this budget represents a rise of €4,626,700, (3.30 per cent) this overlooks the natural growth of the Court’s operations, and its increased operations as we move into 2017. Compared to 2016, the number of preliminary examinations rises to ten, with Honduras and Georgia being closed or moving to investigation, and new examinations opening in Burundi, the self-referral of the Gabonese Republic, and the reconsideration of The Comoros situation. The self-referral by the Gabon, falling after release of the proposed budget, highlights the strain that unexpected referrals can place upon the Court’s budget.

With respect to situations, in the proposed budget for 2016, the OTP outlined its expectation to be dealing with ‘twenty-two cases in eight situations’. For 2017, there is an increase to ten situations, (p10) which results in the continued increase in courtroom activity from 200 court hearings in 2015, to a projected 500 hearings for 2017 (p48). As the Court grows, the associated cost of its prior and current docket also creates a ‘snowball effect’. Time will be needed in 2017 for the progression of reparation proceedings in Lubanga and Katanga, alongside hearing the appeal in the Bemba trial, meaning that it is not just future investigations to consider in light of tightening resources, but also the progression of its existing caseload.

To provide some context on these figures and to illustrate the concern, the bi-annual budget for the ICTY in 2010-2011 was approximately €214,000,000 and the ICTR operated with a healthy bi-annual budget of €176,074,077 for the same period. Per year, these two institutions focusing on one ‘situation’ each worked with roughly two-thirds the budget of the ICC, an institution that from next year will be preoccupied with ten situations. International justice, however expensive it might be, was clearly not unaffordable to the major powers, who generously increased the ICTY budget from its humble beginnings of approximately €200,000.

A balanced assessment of the issue, of course reveals that for the 1,479,301,700 spent by the ICC since 2002, there have ‘only’ been four convictions (eight if we include the additional Article 70 convictions). Weighing in at €369,825,425 per core crimes conviction, if we were to transpose that figure to the ICTY, then the 83 convictions would have cost the ICTY €30.61 billion, a figure which is approximately €28 billion above the actual cost (p13). A few rebuttals can be advanced however, to illustrate that this ‘price per conviction’ comparison masks some of the institutional differences between the ICC, ad-hoc tribunals and domestic proceedings.

First, tribunals including the ICTY, ICTR, SCSL and ECCC have the upfront and immediate outlay of establishing investigations for their respective situations, once established and having prosecuted those responsible for the alleged crimes, the costs of the institution can plateau and then recede, rather than increase, as seen in both the ICTY and ICTR and now their move to the MICT. In contrast, as the ICC has 124 States Parties, and is involved in ten different situations, the ‘start-up costs’ that these other Courts faced are borne by the ICC each time it begins work in a new situation, including field costs, interpretation and translation costs.

Second, neither the ICTY nor the ICTR provided any opportunity for victims to present their testimony before the Tribunal outside of serving as a witness. The much broader role for victims at the ICC, through the Legal Representative for Victims and the establishment of the Trust Fund for Victims which has itself seen a 15.3 per cent increase this year, illustrates that the ICC (for better or for worse) encompasses more than a narrow conception of the trial of the accused and their subsequent detention / acquittal.

Third, the ICC and the ICTY / ICTR emerged from two very different contexts, the 1990’s Tribunals were born with immediate jurisdiction granted by virtue of the Court’s establishment. The ICC, on the other hand, experiences a far more conservative approach to case selection, the processes of preliminary examination and complementarity means that the Court is not designed to have its success measured purely by the number of cases it completes, but also whether its presence can encourage effective domestic prosecutions, and some may argue serve as a deterrent. A simple data comparison of successful prosecutions to illustrate effectiveness therefore ignores that the ‘objective’ by which effectiveness is measured against, differs between the three organisations.

The budget issue is in essence not new, but it is vital to remind ourselves that the ‘colonial critique’ or so-called ‘African bias’ that received so much attention at the ASP is a multi-faceted issue. For the OTP to be able to expand the number and geographic spread of its investigations, there has to be the adequate funding to support it – thus it raises deep questions about the reluctance of certain States to increase the Court’s budget, at a time when the OTP appears to be preparing the ground for an investigation in Afghanistan. There is an available contingency fund, which allowed the OTP to open their 2016 investigation into Georgia (p10), a cost now incorporated into the 2017 budget; but at €7,000,000 (p169), the contingency fund is not designed to cover the costs of entire unplanned investigations.

In closing, and to use the United Kingdom as but just one example, the opening and closing lines of their general debate address highlights the point that States more than ever are unwilling to back up their rhetoric with adequate financial support.continued support for, and commitment to the International Criminal Court’, only to qualify this by expressing their commitment to ‘working with others to ensure the budget is as streamlined as possible.’

The withdrawals from the ICC were a set back for the Court’s desire to be universal, but for all the hyperbole of dialogue in the fight against impunity that dominated the ASP, the internal contradictions of States Parties own positions lays bare the fact, that whilst others might not be withdrawing, support for the Court is found not just at the ASP podium, but also in its financing. The wriggling away by States Parties from funding the Court raises even more interesting questions as to their justification – especially at a time when the Court appears destined to broaden its horizons.

*For clarity – all budget references include interest and the principal repayment (installments) for the host State loan for the permanent premises. – Other budget references may vary.

Brexit Symposium: UK Trade Negotiations with the EU

by Roger Alford

On November 7, 2016 I was privileged to organize a conference at the University of Notre Dame’s London Global Gateway on the topic of UK trade and Brexit. The conference had three sessions: (1) UK trade negotiations with the EU; (2) UK trade negotiations outside the EU; and (3) UK’s post-Brexit status within the WTO. The participants were all trade experts, including Lorand Bartels at Cambridge, Meredith Crowley at Cambridge, Piet Eeckhout at UCL, Jennifer Hillman at Georgetown, Rob Howse at NYU, Simon Lester of the CATO Institute, Sophie Robin-Olivier at Paris II Sorbonne, and yours truly.

Today I am linking to the first session that features Piet Eeckhout, Simon Lester, and Sophie Robin-Olivier. Piet Eeckhout focused on the High Court of Justice decision regarding Parliamentary oversight of the Prime Minister’s Article 50 withdrawal from the EU. Simon Lester focused on the possible meanings of the referendum and the likelihood of a “hard” or “soft” Brexit. Sophie Robin-Olivier focused on the linkage between the free movement of goods and persons, and the EU’s likely response to the UK’s attempts to decouple the issues. The discussion then addressed expert predictions of the likely result of UK trade negotiations with the EU. The consensus was that the EU has the stronger negotiating position and will not accept any free trade deal without free movement of persons. If the UK does not accept those terms, then the most likely result will be the UK trading with the EU under WTO rules.

UPDATE: Summary of Session Two on UK Trade negotiations outside the EU is available here.

UN Apologizes for Role in Cholera Outbreak

by Kristen Boon

On December 1 in a meeting in the UN’s Trusteeship Council, the UN Secretary General apologized for not doing more in the UN Haiti Cholera affair, stating “”On behalf of the United Nations, I want to say very clearly: We apologize to the Haitian people … “we simply did not do enough with regard to the cholera outbreak and its spread in Haiti. We are profoundly sorry for our role.”   It also announced details of a material assistance package that will total some $200 million, provided sums can be raised.   A media report on the speech can be found here.  The webcast is currently available here.

 

This meeting was eagerly anticipated, as the culmination of the UN’s change of direction, which it signaled in August of this year.  After announcing that that the UN would provide some compensation in October, the UN announced a two-track approach involving better water sanitation (track one) and “material assistance” (track two) to the victims.  The details of this new approach were released in a new Secretary General report.

Of particular interest is the Material Assistance Package, which is described as follows: “Track 2 is the development of a package of material assistance and support to those Haitians most directly affected by cholera, centered on the victims and their families and communities.  Affected individuals and communities will participate in the development of the package.  This will inevitably be an imperfect exercise, fraught with practical and moral hazards, and it has been complicated by the impact of Hurricane Matthew.  The package is not likely to fully satisfy all those who have been calling for such a step, nor will it happen overnight.  However, the Secretary General has concluded that it is better to take this step than not to.”

The report indicates that much work remains to be done.   First, the funds for Track II ($200 million) need to be raised, and paragraphs 60 – 64 demonstrate there is no clear timeline. Second, the reports details two different approaches to assistance: community based or individual. The report notes the logistical difficulties of proceeding down this path, although it doesn’t eliminate it. Due to the absence of data on who the victims of cholera are and were, it seems likely that a community based approach will prevail.

Reactions to the announcement have been generally positive. In a press release, Brian Concannon, one of the lawyers for the victims and Executive Director of the Institute for Justice & Democracy in Haiti, stated “This marks a remarkable shift in the UN’s response, and is a major victory in the cholera victims’ six-year long struggle for compensation, cholera treatment and elimination, and an apology. Victims have demanded justice from the streets of Port-au-Prince to the courts of New York, and finally they are being heard.”   However, many have been quick to pick up on what the UN did not say: that it was responsible for introducing cholera into Haiti.  Philip Alston, Special Rapporteur on Extreme Poverty and author of a recent and very critical report on the UN’s actions, termed this a “half-apology” in an interview with The Guardian because the Secretary General omitted to apologize for the introduction of Cholera in the first place.  He declared this a “missed opportunity.”

It is significant from another perspective as well: if the UN had acknowledged its liability and accepted responsibility for the introduction of cholera in Haiti, the material assistance could have been presented as expenses of the Organization under Art. 17 of the UN Charter, which would have given the Secretary General the opportunity to request they be added to the regular budgets (such as the peacekeeping budgets) and assessed from Member States at the normal rate.

We will be posting other reactions to the UN announcement this week:  stay tuned!

 

Regionalization as a Blessing  or as a Curse? The EU and International Criminal Justice

by Carsten Stahn

[Carsten Stahn is Professor of International Criminal Law and Global Justice at Leiden University and Programme Director of the Grotius Centre for International Legal Studies. He is Co-Editor-in-Chief of the Leiden Journal of International Law, Executive Editor of the Criminal Law Forum and project leader of the Jus Post Bellum Project.]

Introduction

International Criminal Justice is a tipping point. It is a bit like a scene in Woody Allen’s “Match Point” movie. The ball is in the air. It has hit the net. But it cannot quite decide where to go. It may tip to the one corner of the field, or the other.

There has been a lot of support for international criminal justice in past decades. The EU has been crucial in this regard. But there is also backlash and critique. Some wonder what’s the point of international courts and tribunals. The international community is good at creating new treaties and new institutions. But it seems to  be less good at devoting long-term attention and resources that are necessary to ensure that they are effective. Once a new institution is created, there is a feeling the job is done and that is time to move on.

After the ICC withdrawals, the question of the justification of international criminal justice is more acute than ever. Some might claim that the withdrawals are a sign that international criminal justice becomes more effective since it targets power politics. Other might say that we need to go back to the drawing board and reflect more critically on the foundations of our assumptions. Both arguments appear to have a grain of truth.

One crucial question is the role of regional organizations in international criminal justice. After the end of the Cold War, institutional development has quickly shifted from domestic to universal approaches. The role of regional institutions has long remained at the periphery.     Recently, much attention has been devoted to regionalization in the context of African critiques of international criminal justice, and the Malabo Protocol. While the Protocol has many problems (e.g., in relation to crime base, complementarity or immunities), there seems to be at least some support for the general assumption that regionalism can have benefits for international criminal justice enforcement.  Such advantages include geographical proximity to crimes, and the ability to reflect specific regional interests or priorities. In existing doctrine, some attention has been devoted to the role of regional human rights courts as “quasi-criminal jurisdictions”. But there are relatively few explorations of the role of the EU as actor in international criminal justice.

The EU counts undoubtedly among the strong champions of international criminal justice on the international plane. One of its unique strengths is that it has achieved some “unity in diversity”. All 28 EU member states are states to the ICC Statute. This allows EU members to act as a collective entity.

EU support for the project of international criminal justice cuts across institutions. The Council has enacted multiple legal instruments to strengthen national investigation and prosecution of mass atrocity crimes. Initiatives, such as the European Arrest Warrant, the European Evidence Warrant, the Framework Decision on the freezing of property and evidence, and the Framework Decision on the standing of victims in Criminal Proceedings, or the European Network of Contact Points are relatively unique on the international plane.

The EU was the first regional organization to sign an agreement on cooperation and assistance with the ICC in 2006. The European Parliament has been a strong supporter of international justice. Europeans have taken a lion’s share in the funding of international criminal justice. This is complemented by the important work of Eurojust, and of course, the broader network of the Council of Europe which extends beyond EU Member States.

The main achievements from a macro perspective are in my view two-fold.

First, European institutions have forged a certain alignment of normative preferences within the European legal space. This is an important achievement. Hardly any other regional bloc has gained a similar level of convergence, and approximation of national approaches.  Decisions under the Justice and Home Affairs Pillar have prompted various member States, like Belgium, Denmark, Germany, The Netherlands and Sweden to establish “specialized units” for the investigation and prosecution of international crimes.  EU member states score high in terms of implanting legislation. This suggests a positive correlation between EU membership and commitment to international criminal justice. The European Area of Freedom, Security and Justice remains imperfect. The EU can do better in terms of strategic coordination.  Some domestic jurisdictions feel that developments are so dynamic that it is difficult to keep up with the pace of transformation. But the degree of cooperation defies comparison.

Second, EU approaches have significantly contributed to “damage control” at the international level. Without the support of European countries, the ICC might have never seen the day in its current form. In the early years of the Court, the EU has taken a strong counter approach towards US policies towards international criminal justice. US approaches have navigated between objection under the Bush Administration to “smart power” approaches under Obama administration.

The EU has differed fundamentally. It has openly discarded US objections in a common position in 2003, while trying to foster a constructive partnership between the US and the ICC. The EU has defined guiding principles for bilateral non-surrender agreements under Art. 98 of the ICC Statute. Later, the EU members have been instrumental in securing Security referrals to the ICC in relation to Darfur and Libya, and supporting a Syria referral.

The EU approach may be characterized by three cardinal features: (i) “principled” pragmatism,  (ii) non-confrontational approaches, and (iii) a long-term vision towards international justice. These are virtues that are key to the success of international criminal justice.

In times like these, the EU serves more than ever as a fire brigade. Damage control is urgently needed. The voices of European members on the Security Council are crucial to avoid action that might hamper existing institutions. There is a need to speak up against unfair critique, and to counter false rhetoric. (more…)

Contextualizing the Debate on First Strikes

by Charles Kels

[Charles Kels is a major in the U.S. Air Force. His views do not reflect those of the Air Force or Department of Defense.]

The fascinating and edifying debate between Adil Haque (see here, here, here, and here) and, respectively, Deborah Pearlstein (see here), Jonathan Horowitz (see here and here), and Kevin Heller (see here and here) over the criteria for non-international armed conflict (NIAC) risks overlooking the proverbial elephant in the room: what to do about the phenomenon of states, notably the United States, invoking jus ad bellum principles to both justify and regulate the use of force?

One way of looking at Adil’s “first strike” proposal is as a solution to the problem of so-called “self-defense targeting” or “naked”/“robust” self-defense: it preserves the distinction between the jus belli branches by ensuring, in Daniel Bethlehem’s formulation, that “any use of force in self-defense [is] subject to applicable jus in bello principles governing the conduct of military operations.” Adil’s framework would, at least presumably, complicate the current White House’s efforts to distinguish between “the use of force in armed conflict or in the exercise of the nation’s inherent right of self-defense,” since the very fact of a military strike against certain non-state actors would automatically trigger a NIAC.

Of course, solving one problem can engender others. As Kevin notes, the lex lata is clearly on the side of Adil’s critics in insisting upon a substantive organization and intensity test for NIACs. Perhaps more to the point, the traditional critique of post-9/11 U.S. counterterrorism operations—particularly of the targeted killing variety—has been that a geographically unbounded conflict, whether framed as a war on terrorism or against specific terrorist groups, is essentially asserting the right to render the entire world a warzone. There is at least some (and maybe more than some) validity to this concern, as evidenced by the malleability of extralegal terms such as “areas of active hostilities” that the president can apparently turn on or off depending upon policy preferences.

This is the problem that Jonathan highlights with his Trojan horse analogy: lowering the NIAC threshold triggers humanitarian protections, but also invites wartime rules for targeting and detention. In this view, Adil’s proposal threatens to undo the hard work of those in academia and elsewhere who have persistently rejected the notion of a transnational NIAC without territorial limitations, or what they perceive as powerful states playing fast and loose with the concept of armed conflict in order to inflict lethal force with relative impunity.

Adil’s response to this critique is as brilliant as it is unsettling. Essentially (and to oversimplify), he maintains that the only real consequence of applying international humanitarian law (IHL) to first strikes is to create war crimes accountability for flagrant violations. Otherwise, the co-application of human rights law (IHRL) operates to negate deadly force as a first resort except where states have formally derogated from applicable treaties and such derogating conduct proves necessary (necessity here can only be understood in the IHRL, rather than IHL, sense). Central to Adil’s argument is the notion that IHL does not authorize conduct which it fails to prohibit: comportment with IHL is a necessary, but not sufficient, condition of lawful killing in armed conflict.

It probably comes as no surprise that as a military practitioner, I have deep misgivings about Adil’s cramped reading of IHL (or put another way, his rendering of IHL-IHRL co-application in a manner that appears to swallow IHL in seemingly every case but pitched battle on Napoleonic terrain). His conception of IHL as solely constraining, and never enabling, seems to cut against the way law in general works. To borrow Adil’s driving analogy, is a speed limit of 75 miles per hour not, at least in some way, permission to drive 70?

More specifically, while Adil astutely emphasizes that IHL does not confer “affirmative” legal authorities on states, it does privilege and immunize certain conduct that would otherwise be illegal. This is the point that Deborah made in her colloquy with Adil: IHL provides the legal framework for status-based targeting in armed conflict, not necessarily because it grants the power to kill, but because it removes the presumption that killing is unlawful in virtually all cases besides self-defense. “War,” as Telford Taylor famously wrote, “consists largely of acts that would be criminal if performed in time of peace.” (To Adil’s point that combatant immunity is inapplicable in NIACs anyway, I would only proffer that the convergence of IAC/NIAC norms may increasingly cut against this, and that better minds than mine—notably Ian Henderson and Jen Ohlin—have taken this issue head-on.)

Indeed, what Adil terms the “mistaken view” of states is, in my estimation, the very key to IHL’s cogency and moral force as a self-contained system of licenses and limitations regulating armed conflict and “alleviating as much as possible the calamities of war.” In my own limited experience and understanding, it is hard to overstate the significance of IHL as not just a set of restrictions overlapping and complementing IHRL, but as a moral lodestar critical to defining what it means to be an honorable warrior. Military lawyers tend to perceive IHL less as a barrier or obstacle telling the client what it cannot do—although it serves this function as well—and more “as a prerequisite to the meaningful exercise of power” in the first place. As Geoffrey Corn recently reminded us, IHL “serves to mitigate the potential moral corrosion that is often produced by mortal combat,” in large part by “providing the warrior with a rational and morally grounded framework” undergirding their actions.

It is notable in this regard that many of the pivotal developments in IHL have “owed less to professors, statesmen, or humanitarians, than to soldiers” themselves, resulting in a set of rules both by and for warriors that reflects the accrued wisdom of history and hardship. Armed conflict denotes a specially-defined space wherein soldiers can act effectively, decisively, justly, and—yes—violently pursuant to a unique code of conduct.

This is why it was important to rebut Ryan Goodman when he admirably, but erroneously, argued for a duty to capture under IHL by offering novel conceptions of military necessity and humanity. Under Adil’s proposal, it is worth noting, whether or not IHL imposes a least-restrictive-means requirement wouldn’t terribly matter, since IHRL would sweep in to impose such a requirement in nearly all cases. The difference between Ryan and Adil’s contentions is mostly the locus of the duty to capture—IHL for Ryan and IHRL operating in parallel with IHL for Adil—but the practical effect of undermining clarity in status-based targeting would be similar.

Adil has done us a great service with his erudite discussion shaking up the sometimes stale debate over NIAC definitions. Personally, I find his proposal to lower the NIAC threshold preferable to any suggestion of raising or complicating the categorical IAC threshold, as erring on the side of some realistically effective regulation of violence seems preferable to the prospect of an enforcement vacuum. But we shouldn’t lose sight of the fact that IHL matters, and not just because it puts war crimes culpability on the table. Armed conflict brings into play both the aspects of IHL that human rights lawyers tend to like, and also those that they don’t. Eroding the boundaries between war and peace can’t help but yield this result.

The Corrosive Risks of Lawless Leadership

by Geoffrey Corn

[Geoffrey S. Corn is Professor of Law at South Texas College of Law Houston in Houston Texas. Prior to joining the South Texas College of Law Houston faculty in 2005, Professor Corn served in the U.S. Army for 21 years as an officer, retiring in the rank of Lieutenant Colonel. He served a subsequent year as the Army’s senior civilian law of war expert advisor. ]

Leadership is the sine qua non of great military forces. And the reason why is clear: because what military leaders demand from subordinates is a complete willingness to go into harm’s way, put life on the line to achieve the collective mission, and to employ deadly force on command.

There is no easy answer to the question of what attributes that define great leaders. Even if there were a list of objective criteria, the intangible qualities – qualities subordinates and enemies alike recognize – probably defy definition. There is, however, one indispensable attribute: trust. The bond of trust between leader and subordinate – a bond based on commitment to the core values that define the nation for which the military fights – is essential to genuine combat effectiveness.

For the American soldier, integrity is the glue that cements the essential bond of trust between leader and subordinate. Integrity is a core value of the U.S. Army, and demands that soldiers and leaders, “Do what’s right, legally and morally . . . a quality you develop by adhering to moral principles.” Respect for law and a genuine commitment to moral interests advanced by the law are therefore central to truly effective leadership and the development of combat effective units.

Respect for law is therefore central to the maintenance of good order and discipline in the military unit, and in that sense is a genuine, “force multiplier” for the unit commander. But true good order and discipline requires more than blind obedience to superior orders. It requires trust – trust that the leaders issuing those orders adhere to the same core value of integrity as the subordinate tasked to obey them. Indeed, it is a myth that military leaders develop units willing to endure the challenges, dangers, and deprivations of war through despotic or dictatorial leadership detached from commitment to law and morality. Indeed, any leader who thinks obedience is ensured by leveraging fear for the consequences of disobedience does not understand the time-tested link between quality leadership and combat effectiveness. Ultimately, the fear of what the subordinate is asked to do will almost always outmatch the fear of sanction for refusing to do it; subordinates confront those fears and the risks of combat that produce them not because they are afraid of their leaders, but because they trust them.

In the U.S. military, commitment to values plays a critical role in building the mutual trust and respect essential to good order and discipline and genuine combat effectiveness. Doctrinal publications that address leadership are replete with this message, emphasizing that men and women who volunteer to serve the armed forces of our great democracy expect their leaders to embrace the values that define our nation. This reflects a recognition that true motivation flows from building genuine trust between the leader and the led; trust that flows from mutual respect. This is what lays the foundation for developing a commitment to a common cause; a passionate desire to make the maximum individual contribution to the collective effort. Indeed, as an officer candidate, the first piece of, “required knowledge” I and my fellow candidates had to commit to memory and recall on demand emphasized this central truism of military leadership. It was General Schoffield’s Definition of Discipline, which warns that, “The discipline which makes the soldiers of a free country reliable in battle is not to be gained by harsh or tyrannical treatment. On the contrary, such treatment is far more likely to destroy than to make an army.”

Prior to the U.S. presidential election, numerous highly respected experts expressed significant concerns about Donald Trump’s commitment to rule of law in relation to military and national security policy development and execution. These concerns have not abated since his victory, and while the signs are still cryptic, many observers fear he, and those he has appointed to senior positions within his administration, will seek to revert back to the type of law avoidance that in many ways defined U.S. national security policy in the aftermath of the tragic terrorist attacks of September 11th, 2001.

This would be a tragic error, and some commentators, including me, have already warned of the adverse strategic consequences of such backsliding. But these potential adverse consequences are not the only reason to be opposed to the dismissal of international legal obligations related to the conduct of hostilities and treatment of detainees. Another, and perhaps even more profound reason, is the corrosive effect on our armed forces such an approach will produce.

The legal regulation of armed conflict advances a range of important interests. The most commonly referenced is the mitigation of human suffering resulting from the effort to limit the destructive and harmful effects of war to only those that are justified by military necessity. However, there is another interest that is often overlooked, an interest directly linked to effective leadership and good order and discipline: providing the warrior with a rational and morally grounded framework that contributes to mitigating the moral hazard resulting from the brutal reality of war. This is no trivial or inconsequential benefit. Men and women who have been subordinate to command directives related to warfare understand, perhaps uniquely, the consequences of unleashing mortal combat power. Indeed, the very essence of military duty is the obligation to employ deadly combat power on order; to kill on demand.

The law of armed conflict serves to mitigate the potential moral corrosion that is often produced by mortal combat. This benefit of legal compliance is perhaps best articulated by Telford Taylor, World War II Army intelligence officer, and subsequently a principle prosecutor of high ranking Nazi war criminals in his book, Nuremberg and Vietnam: An American Tragedy,:

Another, and to my mind, even more important basis of the laws of war is that they are necessary to diminish the corrosive effect of mortal combat on the participants . . .

Unless troops are trained and required to draw the distinction between military and non-military killings, and to retain such respect for the value of life that unnecessary death and destruction will continue to repel them, they may lose the sense for that distinction for the rest of their lives . . .

As Francis Lieber put the matter in his 1863 Army regulations: “Men who take up arms against one another in public war do not cease on this account to be moral beings . . .

Law, or more importantly fidelity to the law, enables this effect, as well as the derivative benefit of building the bond of trust between superior and subordinate. Commanders and other leaders in a military unit are looked to by subordinates to make decisions that advance the objective of mission accomplishment. But there is also an expectation that they will do so consistently with institutional core values, which means subordinates expect that the tasks they are assigned, no matter how brutal or violent, are aligned with their legal obligations.

Responsible commanders understand this, and build units that are prepared to unleash combat power violently and decisively. But that willingness is built on a foundation of trust: that the orders they are executing are consistent with the core values of their military institution and with their nation. Accordingly, commanders must recognize that subordinates simply assume that when, where, and how they use that power must align with the law of armed conflict. As I discussed in an article written last year, international law has historically linked the qualification to engage in hostilities – the “privilege” of belligerency – with operating under responsible command. If all “responsible” meant in this equation were developing subordinates who obey orders, then units like the Japanese forces in Nanking or the German Einsatzgruppen would be icons of responsible commands. But we know they are not; that they are icons of leadership failure. Why? Because their obedience was disconnected from the imperatives of law and morality.

Any high level decision that compels, or even encourages disconnection from the legal underpinnings of truly responsible leadership risks corroding the bond of trust and confidence between leader and led. Even where a short term tactical or operational advantage is perceived from such policies, this must be outweighed by the long term negative consequences to good order and discipline, and to the moral integrity of the men and women who accept the burden of service. Leaders bear an obligation to protect from the hazards of war, not only physical, but also moral. As James McDonough expressed so prophetically in his seminal memoir of small unit leadership in Vietnam,

I had to do more than keep them alive. I had to preserve their human dignity. I was making them kill, forcing them to commit the most uncivilized of acts, but at the same time I had to keep them civilized. That was my duty as their leader . . . War gives the appearance of condoning almost everything, but men must live with their actions for a long time afterward. A leader has to help them understand that there are lines they must not cross. He is their link to normalcy, to order, to humanity. If the leader loses his own sense of propriety or shrinks from his duty, anything will be allowed . . . War is, at its very core, the absence of order; and the absence of order leads very easily to the absence of morality, unless the leader can preserve each of them in its place.

It is imperative that all military leaders, including our commander in chief, understand this vital relationship between law and leadership. This is why U.S. commitment to law must be manifested in every aspect of our military operations, and projected globally as a defining aspect of our global identity. Disconnecting leadership from the time-tested and credible legal foundation provided by the law of armed conflict risks corrosion of the essential bond of trust between leader and led. The decisions made by our president, and the tone those decision project, will reverberate through the force with intense magnitude. As a desktop leadership reminder that I inherited from a former commander emphasizes, “leader actions are exaggerated and repeated.” What needs to be exaggerated and repeated is commitment to law, and to the integrity and honor that such commitment manifests. And that must start with our President.

A Brief Rejoinder to Haque on the ICRC’s Interpretation of NIAC

by Kevin Jon Heller

My thanks to Adil Haque for his response to my post. Adil and I rarely disagree in any profound way about IHL, so it’s enjoyable to spar with him about whether a first-strike by government forces against an organized armed group automatically creates a NIAC — thus triggering IHL — or whether a certain intensity of hostilities between the two is required.

I will have more to say about Adil’s response soon, but I wanted to quickly address one particular implication in his post: namely, that the ICRC’s Commentary on AP II supports his claim that a single military operation by government forces or by an organized armed group is sufficient to trigger a NIAC because it is more than a “sporadic act of violence.” Here is what he writes:

In my view, a military operation by State armed forces that meets with no armed response and may never be repeated is not a “sporadic act of violence” within the meaning of APII 1(2). On this point, I follow the ICRC Commentrary to APII, which negatively defines “isolated and sporadic acts of violence, as opposed to military operations carried out by armed forces or armed groups.” APII 1(2) describes disturbances and tensions created by disorganized or unarmed groups, criminal gangs, and individuals. APII 1(2) does not describe “acts of violence against the adversary in offence or defence” (that is, attacks as defined by API).

With the exception of ambiguous quotes like the one above, there is little support in the ICRC’s Commentary on AP II or in any of the ICRC’s commentaries for Adil’s position. The ICRC clearly believes that any kind of NIAC — AP II or Common Article 3 — requires adequately intense hostilities.

Let’s start with the AP II Commentary Adil cites. The Commentary opens its discussion of AP II by emphasizing (p. 1343) that CA3 and AP II have the same structure — and that neither applies in the absence of sufficiently intense hostilities (emphasis mine):

The content and scope of all of these articles will be analysed in the respective comments on them. Before doing this it seems useful to have a closer look at the basic pattern of Part I, which reveals the similarity of the ideas which inspired Protocol II and common Article 3. To understand the scope of the Protocol one should indeed always bear in mind the fact that this instrument supplements and develops common Article 3; it is an extension of it, and is based on the same structure.l Their common characteristics find expression, explicitly or implicitly, in Part I. These can be summarized as follows…

The threshold where Protocol II becomes applicable is determined by the criteria expressed in Article 1 (Material field of application), which means that it is intended to apply only to conflicts of a certain degree of intensity.

Later, the Commentary discusses (p. 1355) what AP II means by “internal disturbances,” taking the position that such disturbances include situations in which military operations by government forces — even against an organized armed group — do not lead to sufficiently intense hostilities (emphasis mine):

[T]he ICRC gave the following description of internal disturbances during the first session of the Conference of Government Experts in 1971:

“This involves situations in which there is no non-international armed conflict as such, but there exists a confrontation within the country, which is characterized by a certain seriousness or duration and which involves acts of violence. These latter can assume various forms, all the way from the spontaneous generation of acts of revolt to the struggle between more or less organized groups and the authorities in power. In these situations, which do not necessarily degenerate into open struggle, the authorities in power call upon extensive police forces, or even armed forces, to restore internal order. The high number of victims has made necessary the application of a minimum of humanitarian rules.”

[snip]

In short, as stated above, there are internal disturbances, without being an armed conflict, when the State uses armed force to maintain order; there are internal tensions, without being internal disturbances, when force is used as a preventive measure to maintain respect for law and order.

Finally the Commentaries specifically point out (p. 1356) that such “internal disturbances” do not create a NIAC and do not trigger IHL:

Internal disturbances and tensions are not at present within the field of application of international humanitarian law; the ICRC has carried out activities in this field on an ad hoc basis. However, this does not mean that there is no international legal protection applicable to such situations, as they are covered by universal and regional human rights instruments. 31 It is not within the scope of this commentary, however, to go into that subject.

The ICRC’s position on CA3 and AP II NIACs — as requiring hostilities of a certain intensity, and thus as not being triggered by “first strikes” — is an old one. Here is what Pictet said (p. 49) in the ICRC’s 1952 Commentary on the First Geneva Convention (emphasis mine):

[I]t was suggested that the term “conflict” should be defined or, which would come to the same thing, that a certain number of conditions for the application of the Convention should be enumerated. The idea was finally abandoned — wisely, we think. Nevertheless, these different conditions, although in no way obligatory, constitute convenient criteria, and we therefore think it well to give a list of those contained in the various amendments discussed; they are as follows:

(1) That the Party in revolt against the de jure Government possesses an organized military force, an authority responsible for its acts, acting within a determinate territory and having the means of respecting and ensuring respect for the Convention.

(2) That the legal Government is obliged to have recourse to the regular military forces against insurgents organized as military and in possession of a part of the national territory.

This statement only implicitly endorsed an intensity requirement, so Pictet clarified that the was talking about actual hostilities between government forces and an organized armed group in the ICRC’s 1960 Commentary on the Third Geneva Convention (p. 37):

Speaking generally, it must be recognized that the conflicts referred to in Article 3 are armed conflicts, with armed forces on either side engaged in hostilities –conflicts, in short, which are in many respects similar to an international war, but take place within the confines of a single country.

And just in case that statement remained ambiguous (“both” would have been more precise than “either”), the ICRC clarified in its 2016 Commentary on the First Geneva Convention that of the various indicia of NIAC that Pictet discusses, intensity is one of the two most important ones (emphasis mine):

387  A situation of violence that crosses the threshold of an ‘armed conflict not of an international character’ is a situation in which organized Parties confront one another with violence of a certain degree of intensity. It is a determination made based on the facts.

421  Over time, of the criteria enumerated in the Pictet Commentaries, two are now widely acknowledged as being the most relevant in assessing the existence of a non-international armed conflict: that the violence needs to have reached a certain intensity and that it must be between at least two organized Parties/armed groups. The existence of a non-international armed conflict thus needs to be assessed according to these specific criteria.

422  The wording of common Article 3 gives some rudimentary guidance on its threshold of application: what is required is an ‘armed’ ‘conflict’ not of an international character, in which ‘Part[ies] to the conflict’ are involved. This indicates that for common Article 3 to apply, a situation of violence must have reached a certain level of intensity, characterized by recourse to arms by non-State armed groups that are capable of being Parties to an armed conflict.

According to the ICRC, in short, all NIACs require adequately intense hostilities. The difference between a CA3 NIAC and an AP II NIAC is one of degree rather than kind.

The ICRC Commentaries are only as good as the analysis they contain, so Adil is obviously free to defend an interpretation of Art. 1 of AP II and of Common Article 3 that reads the intensity requirement out of NIAC. In doing so, however, he is clearly breaking with the ICRC.

Second Thoughts on First Strikes: A Reply to Heller

by Adil Ahmad Haque

[Adil Ahmad Haque (@AdHaque110) is a Professor of Law and Judge Jon O. Newman Scholar at Rutgers Law School.  His first book, Law and Morality at War, will be published by Oxford University Press in January.]

My sincere thanks to Kevin Heller for thoughtfully engaging with my view that the law of non-international armed conflict (NIAC) applies to first strikes between State armed forces and organized armed groups. Kevin’s post raises a number of important issues, and I am grateful for the opportunity to discuss them. At the outset, I should note that my original post mostly discussed common article 3 of the Geneva Conventions and the customary law of NIAC. However, since Kevin primarily discusses Additional Protocol II, so will I.

Kevin writes that

the Tadic test, which requires both organization on the part of the armed group and adequately intense hostilities . . . is based squarely on Art. 1(2) of Additional Protocol II, … which provides that the “Protocol shall not apply to situations of internal disturbances and tensions, such as riots, isolated and sporadic acts of violence and other acts of a similar nature, as not being armed conflicts.”

Kevin goes on to write that

[t]he text of Art. 1(2) of AP II clearly contemplates actual hostilities, not a single act by government forces. What could be a more “sporadic act of violence” than a single act that does not meet with a response from the targeted group and may never be repeated by the government?

In my view, a military operation by State armed forces that meets with no armed response and may never be repeated is not a “sporadic act of violence” within the meaning of APII 1(2). On this point, I follow the ICRC Commentrary to APII, which negatively defines “isolated and sporadic acts of violence, as opposed to military operations carried out by armed forces or armed groups.” APII 1(2) describes disturbances and tensions created by disorganized or unarmed groups, criminal gangs, and individuals. APII 1(2) does not describe “acts of violence against the adversary in offence or defence” (that is, attacks as defined by API).

Unfortunately, the ICRC Commentary goes astray when it suggests that APII only applies to “sustained and concerted” military operations. According to the Commentary,

[a]t the beginning of a conflict military operations rarely have such a character; thus it is likely that only common Article 3 will apply to the first stage of hostilities.

Indeed, common article 3 applies to the first stage of hostilities. So does APII. The ICRC’s contrary view is hard to square with the text of APII 1(1), which refers to

organized armed groups which, under responsible command, exercise such control over a part of its territory as to enable them to carry out sustained and concerted military operations and to implement this Protocol.

APII 1(1) requires that a group has the ability to carry out sustained and concerted military operations, not that the group has already carried out such operations. In other words, this clause bears on the organization criterion, not on the intensity criterion.

The only term in APII 1(1) that bears on the intensity criterion is “armed conflict.” On the prevailing view, an “armed conflict” between States begins with the first use of military force by one against the other. In my view, an “armed conflict” between a State and an organized armed group begins with the first military operation by one against the other. Nothing in the text of APII 1(1) or APII 1(2) precludes this symmetrical approach.

As Kevin notes, I believe that we should interpret the term “armed conflict” in light of the object and purpose of IHL, namely to protect human beings against dangers arising from military operations. In general, we interpret all the terms of a treaty in light of the object and purpose of the treaty as a whole. If protecting human beings from military operations is the primary purpose of APII as a whole, then we should interpret every term of APII in light of that purpose. The term “armed conflict” should be no exception.

In contrast, Kevin argues that while IHL’s protective purpose should inform our interpretation of IHL’s substantive rules, it should not inform our interpretation of IHL’s field of application. I find this argument hard to accept. After all, there is no easier way to subvert the object and purpose of a legal rule than to deny its application. Indeed, according to the ICRC, “Common Article 3 . . . gave rise to a great variety of interpretations and in practice its applicability was often denied.” APII 1(1) was drafted in order “[t]o improve the protection of the victims on non-international armed conflicts” by ensuring that “the authorities concerned could no longer deny the existence of a conflict” and, with it, all the protections of IHL. Whether it succeeded in that aim is another matter.

Now, Kevin also suggests that we don’t need to look to the object and purpose of APII to interpret the term “armed conflict” because, in the language of the Vienna Convention, subsequent practice in the application of the treaty establishes the agreement of the parties that it does not apply to first strikes. It’s an empirical question, and Kevin may be right. We would have to look at State practice and see what we find. If Kevin is right, then I would indeed make the normative argument that States should change their view. As Kevin knows, I am quite fond of normative arguments.

However, I suspect that States haven’t thought this question through. If they did, I suspect that they would insist that IHL requires humane treatment of a single soldier captured by an armed group, and that IHL constrains the initial advance of a group like Daesh. Or so I argued in my initial post.

Finally, Kevin argues that my purposive approach entails that IHL should apply to military operations by State armed forces against unorganized armed groups. As Kevin puts the point, “[a]ll of Adil’s arguments against the intensity requirement apply equally to the organization requirement.”

Of course, APII 1(1) expressly states that APII applies only to armed conflicts between State armed forces and organized armed groups that possess responsible command, territorial control, and the ability to carry out sustained and concerted military operations as well as to implement APII. While the terms of APII’s organization requirement should be interpreted in light of APII’s object and purpose, they cannot be ignored.

For their part, common article 3 and the customary law of NIAC apply to all organized armed groups, irrespective of territorial control. Importantly, it is a structural principle of IHL that IHL applies equally to opposing parties and their armed forces. The legal obligations that IHL imposes on armed groups presuppose that such groups are sufficiently organized that they are capable of conforming to those obligations. Where such organization is absent, individuals must seek legal protection in human rights law, the law of genocide and crimes against humanity, and domestic criminal law. IHL is designed to protect human beings from one source of danger. Unfortunately, there are others.

First Strikes & NIAC: Thoughts on the Haque/Horowitz Debate

by Kevin Jon Heller

I have been following with great interest the debate at Just Security between Adil Haque and Jonathan Horowitz over whether the existence of a non-international conflict (NIAC) exists the moment a state launches a “first strike” at an organized armed group or whether hostilities of a certain intensity between the two are required. Adil takes the former position (see here, here, and here); Jonathan takes the latter one (see here and here).

Though Adil’s posts exhibit his typical brilliance, my sympathies lie with Jonathan. To begin with, as a matter of the lex lata, I don’t think the argument is even close: the Tadic test, which requires both organization on the part of the armed group and adequately intense hostilities, has overwhelming support from states. After all, the test is based squarely on Art. 1(2) of Additional Protocol II, ratified by 168 states, which provides that the “Protocol shall not apply to situations of internal disturbances and tensions, such as riots, isolated and sporadic acts of violence and other acts of a similar nature, as not being armed conflicts.” And, of course, as the ICRC notes in its new commentary on the First Geneva Convention, the AP II standard is used by a number of more recent conventions that apply to all NIACs — Common Article 3 or AP II — such as the Rome Statute (1998), the Second Protocol to the Hague Convention for the Protection of Cultural Property (1999), and the Convention on Certain Conventional Weapons (2001).

Adil, it is worth noting, has a different interpretation of AP II, one that does not require intensity:

In my view, if an organized armed group has the capacity to sustain military operations then any military operation by or against that group should be constrained by the law of armed conflict. The organization and capacity of the group is sufficient to distinguish military operations by or against the group from “internal disturbances and tensions, such as riots, isolated and sporadic acts of violence and other acts of a similar nature.”

This is a difficult position to defend. The text of Art. 1(2) of AP II clearly contemplates actual hostilities, not a single act by government forces. What could be a more “sporadic act of violence” than a single act that does not meet with a response from the targeted group and may never be repeated by the government? More importantly, despite some stray practice cited by Michael J. Adams and Ryan Goodman in this post, states have simply never interpreted the AP II standard to require only organization.

Even more problematic, though, is Adil’s argument that the “object and purpose” of IHL counsels against conditioning the application of IHL on adequately intense hostilities:

In my view, we should interpret both the substantive rules of IHL and the conditions for the application of IHL in light of the object and purpose of IHL. The primary object and purpose of IHL is to protect human beings against dangers arising from military operations. Accordingly, IHL should apply to all such military operations. To postpone the application of IHL until a first strike triggers an armed response, or until military operations reach a high level of intensity, would be inconsistent with the object and purpose of IHL.

As regular readers know, whenever I see arguments based on the supposed “object and purpose” of a treaty, I reach for my pen. All too often, such arguments simply use object and purpose to justify interpreting a treaty in a manner that specifically contradicts the intention of the states that drafted and concluded it. And unfortunately I think that is what Adil does here. He defends applying IHL to first strikes by claiming that the “object and purpose” of IHL is “to protect human beings against dangers arising from military operations” (emphasis mine). But that is misleading: the object and purpose of IHL is to protect human beings against dangers arising from military operations in armed conflict. If there is no armed conflict, IHL has nothing to say about the danger of military operations — because IHL doesn’t apply. And as discussed above, states have always insisted that a first-strike military operation is not enough to create an armed conflict — IHL applies only once there are adequately intense hostilities between government forces and the organized armed group.

Adil is free, of course, to normatively argue that IHL should apply to first strikes in NIAC because doing so would better protect human beings. I would disagree, but the claim is coherent and deserving of discussion. What he can’t do is base that claim on the object and purpose of IHL, because that would be to use an object and purpose that only applies within armed conflict to justify changing the definition of armed conflict itself. The definition of when IHL applies cannot be determined by reference to what the goals of IHL are once it applies. That definition has to be sought outside of the IHL system — and again, it is clear that states do not want IHL to apply to first-strike military operations against organized armed groups.

There is, however, an even deeper problem with Adil’s argument that the need to protect human beings from military operations counsels a definition of NIAC that does not require adequately intense hostilities: if that is true, there is also no reason why the application of IHL should require armed groups to be organized. All of Adil’s arguments against the intensity requirement apply equally to the organization requirement. If we need to protect human beings from the dangers of first-strike military operations by states against organized armed groups, surely we also need to protect them from the dangers of first-strike military operations by states against unorganized armed groups. After all, Adil’s central argument is that the inherent danger of military operations means that IHL should apply to a first-strike regardless of whether that military operation leads to any kind of hostilities.

I see no convincing response to this criticism. It is tempting to argue that the organization requirement is important because a first-strike military operation against an organized armed group is much more likely to lead to actual hostilities than a first-strike military operation against an unorganized armed group. But Adil rejects the idea that hostilities are relevant to the application of IHL. He believes IHL should apply even if a first-strike military operation meets with no response whatsoever.

Another potential response would be to argue that first-strike military operations against organized armed groups pose greater dangers for innocent civilians than first-strike military operations against unorganized armed groups. But that would be a difference of degree, not of kind — and thus far from a convincing basis for applying IHL to the former and not the latter. I’m also not sure whether the claim is even empirically sound. It is at least equally plausible to assume that states are more willing to use military force against unorganized individuals whom they can assume will not fight back (or will not fight back effectively) than against an organized armed group with the capacity to respond to a first strike with military operations of its own.

Adil’s desire to protect human beings from the dangers of military operations is laudable, but his claim that IHL should apply to first strikes against organized armed groups cannot be sustained. Not only have states insisted that IHL applies only to hostilities that reach a certain level of intensity, the idea that protecting individuals from the danger of military operations requires eliminating the intensity requirement is underinclusive. Those dangers exist for all military operations, even those against unorganized armed groups. So the only consistent — if still objectionable — position is that IHL applies to any military operation launched by a state, regardless of its object. I’m curious whether Adil would be willing to take that position.

Clarifying the Alternative: International Criminal Justice Options for African Union Member States

by Oliver Windridge

[Oliver Windridge is a British lawyer specializing in international human rights and international criminal law. Oliver is founder of The ACtHPR Monitor, an independent blog and website dedicated to the African Court on Human and Peoples’ Rights, on twitter @acthpr_monitor. In June 2014 he was one of five non-African lawyers to be appointed to the Court’s inaugural list of Legal Aid Counsel (pro bono). The views expressed herein are those of the author and do not necessarily reflect the views of the United Nations or any other organization affiliated to the author.]

Following moves from Gambia, Burundi and South Africa in the past weeks to withdraw from the ICC, much thought is now being given, and keyboards worn down, by the international community as it considers what this news will mean for these countries individually, Africa more generally and of course the ICC.

I want to slightly side-step some of these issues though and address the seemingly confused narrative circulating on the African alternative to the ICC. This seems especially important given the South African Minister for Justice Michael Masutha’s first statement on South Africa’s withdrawal from the ICC which reportedly included the line “South Africa will work closely with the African Union and with other countries in Africa to strengthen continental bodies, such as the African Court on Human and People’s Rights, created to deal with such crimes and to prosecute the perpetrators, whilst at the same time continuing to participate and honour its commitments under international human rights instruments.”

Let’s be clear, the African Court on Human and Peoples’ Rights does not and will not, barring an extraordinary turn of events, ever have the jurisdiction to try cases of genocide, crimes against humanity and war crimes. The African Court is a human rights court, similar in many ways to its European and Inter-American cousins. It has jurisdiction to consider alleged violations of the African Charter on Human and Peoples’ Rights and other international human rights instruments to which the member state in question has signed up to, but nothing in the African Court’s statute, Protocol or Rules gives the African Court the power to prosecute individuals for crimes such as genocide, crimes against humanity and war crimes. To point to the African Court as an alternative to the ICC is like Germany or France withdrawing from the ICC and then pointing to the European Court of Human Rights to handle things from now on.

There is another possibility for AU member states though on the horizon and one that not only has the same jurisdiction as the ICC but actually one much greater; the African Court of Justice and Human Rights (ACJHR). At present however this court exists on paper only in the form of the Protocol on the Statue of the African Court of Justice and Human Rights and the subsequent 2014 Draft Protocol on Amendments to the Protocol on the Statute of the African Court of Justice and Human Rights. Read through this second Draft Protocol and there emerges details of a mega-court that would subsume the current African Court on Human and Peoples’ Right which would then exist as a “Human and Peoples’ Rights Section”, alongside a “General Affairs” section and an “International Criminal Law” section. The International Criminal Law section would have three chambers, a Pre-Trial Chamber, a Trial Chamber and an Appellate Chamber and have jurisdiction to try individuals on charges of genocide, crimes against humanity and war crimes. Sound familiar?

There are however a number of differences between the ICC and the International Criminal Law Section of the ACJHR. To start with, in addition to genocide, crimes against humanity and war crimes, the ACJHR would have jurisdiction over a further 11 crimes including corruption, mercenarism and the trafficking in persons and drugs, creating an ambitious roster of 14 crimes under its jurisdiction (Article 28A, Draft Protocol). Additionally, the International Criminal Law Section of the ACJHR would include immunity from prosecution for serving AU heads of state or government and other “senior state officials” whilst in office- a vague definition if there ever was one (Article 46A bis, Draft Protocol). But above concerns over immunity and jurisdiction exists the real kicker; to take the ACJHR off the page and into reality requires at least 15 AU member states to ratify the amended Protocol (Article 11, Draft Protocol). Currently however, not a single AU member state has ratified the Draft Protocol and its international criminal law section- not one. This complete lack of ratification goes some way to demonstrate the clear gap that exists between those advocating for withdrawal from the ICC ostensibly to allow the AU to handle matters, and those who have actually signed up to the AU’s alternative.

So as Gambians, Burundians, South Africans and the wider world starts to come to grips with what withdrawal from the ICC could mean, let’s be clear on the regional alternative – the African Court of Justice and Human Rights and where it currently exists- on paper only.

Torture and the U.S. Military

by Deborah Pearlstein

Cross-posted at Balkinization

There should by now be little doubt that various members of the incoming administration, including the President himself, would be willing to torture terrorist suspects should opportunity arise. On the campaign trail, Donald Trump expressed a desire to return to “waterboarding” terrorism suspects and “worse.” Mike Pence declined to rule out torture when asked about it expressly this past weekend. Nominee for CIA Director Mike Pompeo opposed President Obama’s decision in 2009 to close C.I.A. black-site prisons and also to require interrogations to comply with the rules of the Army Field Manual. Army Lt. Gen. Michael Flynn, the choice for national security adviser, is perhaps more equivocal. As a firsthand witness to the counterproductive effects of abusive interrogation, he has said that “I would not want to return to ‘enhanced techniques,’ because I helped rewrite the manual for interrogations.” On the other hand, “if the nation was in grave danger from a terrorist attack involving weapons of mass destruction, and we had certain individuals in our custody with information that might avoid it, then I would probably OK enhanced interrogation techniques within certain limits.”

Even with all best intentions, Congress and the courts are unlikely to play much role at the outset in reining in this particular kind of presidential ambition. There are clear statutory prohibitions against the use of torture as it is; and the courts are empowered to act only once an actual case or controversy is before them. It was in no small measure in the face of the same dilemma during the first George W. Bush administration that so many legal scholars turned to focus on the role of internal, intra-branch checks on executive power – the Justice Department Office of Legal Counsel, agency Inspectors General, and others. It also became apparent that the uniformed military could be included among potentially available checks on executive power.

After the attacks of 9/11, military lawyers and others in the Pentagon played a critical role resisting efforts by the Bush Administration to evade laws barring torture and cruelty to detainees in U.S. custody. Not only was such treatment illegal, they argued, authorizing techniques the troops had long been trained were prohibited was disastrous policy: it sowed confusion in the field, compromised operational effectiveness, endangered our troops, and undermined the mission they had been sent to carry out. Well beyond the Pentagon, it was a young Army specialist who helped blow the whistle on the torture that permeated the U.S.-run Abu Ghraib prison in Iraq, and a Major General whose investigation made clear to Congress how inadequate resources, training, and accountability helped allow the abuse to endure and spread. Elsewhere, military lawyers urged Congress to investigate whether war crimes trials at Guantanamo could ever actually succeed in delivering justice. And it was an Army Reserve lieutenant colonel whose declaration about his experiences at Guantanamo extraordinarily persuaded the Supreme Court to change its mind and agree to decide whether the detainees there had a constitutional right to have their cases heard. Entirely apart from the military’s duty to disobey manifestly unlawful orders, both active duty military, and retired leaders, played a pivotal role in preventing America’s torture crisis from becoming worse than it was.

Yet as laudable as this service was, and especially as the incoming administration peoples itself with retired generals galore, the idea that the military might limit the President in the pursuit of his goals should seem at some level exactly backwards. The President is, after all, the Commander in Chief of the military, a symbol of our country’s bedrock principle of civilian control. That principle was born in part from a (Revolutionary War-era) fear of military oppression in ordinary life, a fear that seems unlikely today. But it was also driven by the worry that the military – whose political popularity is unsurpassed in contemporary American life – was capable of exercising outsized influence over democratic decision-making. The image of the “man on horseback” came to symbolize the concern that a particularly successful and charismatic commander could effectively lead the public down a path contrary to its own democratic interests, undermining the ability of elected leaders to accomplish the policy goals the People wanted them to fulfill.

While the military has of course changed dramatically since the Constitution was drafted, the enduring concern that the military might unduly influence politics led to a series of regulations beginning in the early twentieth century restricting active-duty military from engaging in political activities. Congress came to prohibit officers from holding civil elective office, and to impose criminal penalties for using “contemptuous words” against the President, members of Congress, or other elected officials. Today, active duty military personnel are prohibited from participating in partisan political fundraising, rallies, or conventions; using official authority or influence to interfere with an election; or soliciting votes “for or against a partisan political party, candidate, or cause.”

Such proscriptions are sensible. But these rules, coupled with powerful career incentives, have too often been understood to limit the honest expression of professional military dissent. There was in Washington’s time, and today remains, a critical difference between a military expression of partisan alliance and one of professional judgment. And there is certainly a difference between expressions of political disagreement, and an insistence on adherence to law. The era of Abu Ghraib taught us that there is a range of ways in which the military can, consistent with their own duty to uphold the nation’s Constitution and laws, help to steady the ship of state. Of course the military is no panacea. Plenty of troops supported Donald Trump, and not all would oppose a return to torture. But it is also clear that the military is capable of performing at least a part of the same service Americans should expect of all our political institutions: as a platform from which people of good will and a commitment to law can make their voices heard. Those concerned about a return to torture should reach out. For it is as least as likely as any of our institutional checks to help constrain whatever policy adventurism is to come.