What do Appropriate Due Diligence Measures for Nonstate and Substate Partners Look Like? Part II: A Sample of US Practice in Iraq, Syria, and Afghanistan

[Erica Gaston is an international lawyer and non-resident fellow with the Global Public Policy institute (GPPi). She has been engaged in research on IHL and conflict-related human rights issues in Afghanistan, Iraq, Yemen, and other countries since 2007.]

As introduced in Part I of this two-part blog, Western states are now more than ever finding themselves in partnerships (direct or tacit) with nonstate or substate armed groups. This has raised the question of what obligations they have under International Humanitarian Law (IHL), or other legal theories, to ensure that supported groups do not commit war crimes or human rights abuses. Part I discussed some of the legal debates and standards surrounding this issue, in particular whether states have affirmative duties to prevent partner forces from engaging in IHL violations as part of their due diligence obligations under Common Article 1. This second part will instead focus on existing and emerging practice, seeking examples of how states have tried to prevent or address violations by nonstate or substate partners in practice.

The examples will be drawn from a recently published ODI-GPPi joint report, “Regulating Irregular Actors: Can Due Diligence Checks Mitigate the Risks of Working with Nonstate and Substate Forces?” The study explores due diligence and risk mitigation mechanisms in seven US partnerships with nonstate or substate armed groups in Afghanistan, Iraq, and Syria (from 2009 to the present), including provision of training, regular monitoring, establishing rules and standards, threatening or applying sanctions, and other mechanisms.

There are a few important caveats to make up-front. Despite their legal-esque tenor, most of these measures were adopted by the US on a policy basis. The below case studies will also not get into an analysis of the degree of state responsibility or obligation that might be inferred based on the degree of US support to, or control or direction over, the groups in question. In addition, it is worth noting that the ODI-GPPi study itself is not purely focused on legal risks; it also considers how such mechanisms attempted to address other policy commitments, for example, to mitigate security risks or diplomatic consequences surrounding these forces. However, a substantial focus of many of these mechanisms was to try to prevent or deter partner forces from committing atrocities or gross violations of human rights, and thus they provide some concrete examples of what measures or mechanisms might be considered as potential due diligence practices.

Syria

There were a number of different iterations of US support to nonstate armed groups in Syria from 2012 to the present, both lethal and non-lethal in the form of assistance, covert and overt in status, and with groups fighting under either anti-Assad or anti-ISIL banners. All involved mechanisms or procedures aimed at mitigating or preventing abuse or misconduct by supported Syrian groups – to a surprising degree given the conflict context, the limited US force presence in Syria, and the absence in many cases of US domestic legal requirements to take such steps.

The earliest US assistance to Syrian armed groups was the State Department non-lethal assistance (e.g., trucks, communications equipment, medical and humanitarian supplies) to select groups associated with the Free Syrian Army (FSA), which was underway by early 2013. From mid-2013, the CIA followed suit, covertly providing arms, weapons and training to some of the same FSA groups, with the assistance distributed via joint intelligence hubs in Jordan and Turkey (alongside support by Saudi Arabia, Qatar, the UAE, the UK, France, and others). In October 2014, Congress authorized a $500 million program to overtly “Train and Equip” Syrian forces to fight against ISIL. The initial vision of creating a “New Syrian Force” from scratch was quickly abandoned, but DoD has continued to oversee different iterations of the anti-ISIL Train & Equip programming for a range of Syrian forces since 2015, including Arab groups in northern Syria, some around al-Tanf, and those associated with the Syrian Democratic Forces in northeast Syria.

Across all of the programs, there were fairly detailed vetting and screening mechanisms before funding was provided, with the conduct and human rights record of supported groups among the key criteria. The overt State Department and Defense Department assistance programs all adopted vetting criteria based on the Leahy Law, which blocks funding to individuals (and their surrounding units) for whom there is “credible information” of Gross Violations of Human Rights (GVHR) (at p. 2). The decision to do so was notable given the substantial challenges of finding and verifying any information on fighters in Syria, and that this arduous process was not legally required. The Leahy Law – an instrument traditionally aimed at US support to other states – was deemed inapplicable (at p. 25) to the nonstate opposition groups in Syria. The Leahy Law is also not applicable for covert assistance; nonetheless,  past abuses and misconduct also appeared to factor into the CIA vetting process, according to FSA accounts and public reporting. As one US official commented (anonymously): “We routinely vet the groups we work with and support, as you might expect, and their human rights record figures prominently in that.”

Once past the initial vetting stage, all of the programs adopted fairly complex monitoring and reporting mechanisms – from State Department officers in DC and Turkey regularly making calls to grantees and community members, to third-party monitoring, to CIA demands for after-action reporting and in-person debriefs at intelligence hubs in Jordan and Turkey. Often the questioning at intelligence hubs focused on conflict and battlefield developments – no doubt motivated by their intelligence value – but FSA commanders also said they were frequently questioned about allegations of misconduct or war crimes by their affiliates or other groups.

Where FSA groups crossed a “redline” – to include allegations of war crimes or human rights abuses – they were not infrequently cut. The most infamous example was the blocking of the Al-Zenki group from both State Department and CIA support after repeated reports of rights abuses, including the beheading of a youth who was detained.  Depending on the program in question, there were also other interim repercussions, such as limitations on the type of support, temporary blocks or re-training. One FSA commander interviewed said that one group given anti-aircraft TOW missiles through the CIA program misfired and struck a civilian target; in response, their permission to use or be given TOW missiles was revoked and they were required to go through training again and to re-affirm their commitment to a sort of protocol focused on avoiding civilian harm.

Attention to training and IHL-derivative protocols was even more visible with the overt DoD assistance. Recruits for the first iteration of Train & Equip programming received a full “Law of Land Warfare” training (at pp51-52) and had to commit to a pledge (p. 20), which reads as a layman version of core IHL obligations. Some key excerpts include:

“I swear by my God, as a member of the … New Syrian Forces, to defend the Syrian people from Da’ish… I announce my unconditional commitment to the rule of law and following rules of the law of armed conflict.

I will only fight combatants. I will not harm civilians who are not directly participating in the fight… I will treat humanely all who surrender or are captured. I understand all detainees must receive medical treatments, food, water, shelter, basic hygiene… I will not kill or torture detained personnel.  …

I will facilitate relief efforts and humanitarian work. Accordingly, I will protect hospitals, ambulances, and humanitarian and relief workers…

I will treat all civilians humanely. I understand that attacks against civilians are strictly forbidden…

I will prevent violations. … If in command, I will investigate all credible complaints regarding violations…..”

Although the Pledge was not carried forward when the Train & Equip program was revamped in 2015, the subsequent Train & Equip program programs still involved some form of IHL-related training where direct support was provided to units, requirements to vet recruits against evidence of past GVHR, and monitoring and inquiries into whether supported forces recruited or used child soldiers (at pp. 37, 44), among other legal obligations.

This is not to suggest perfect, or even robust, implementation of all of these due diligence measures. In addition, even when substantial efforts were made, there was limited evidence of impact. ‘There is no such thing as perfect vetting in that environment,’ was a sort of truism repeated by many of those involved in Syria support programs. Many also questioned whether a remotely provided primer on IHL, or remote or indirect monitoring could add any value in a context like Syria. One review of Dutch non-lethal assistance, which deployed similar risk mitigation measures to the US programs, found that particular challenges to monitoring for human rights abuses, and that overall many of the monitoring and requirements for Syrian assistance were out of step with the conflict reality in Syria.

Afghanistan

In Afghanistan, US forces and agents worked with a range of militias, tribal groups, and other nonstate or substate forces from 2002 to the present. Some, like the Counter-terrorism Pursuit Teams or “campaign forces” accompanying CIA and SOF on counter-terrorism missions have never had any clear legal tether, and appeared to have few  human rights or conduct-based restraints (at least to judge by the substantial abuses linked to them). However, when US Special Forces decided to mobilize local community and tribal forces in 2009 as part of their counter-insurgency strategy of winning community support, some substantial due diligence measures were imposed (at least on paper).

The initial pitch by proponents of local defense forces was that community-based forces would be less likely to engage in misconduct because they would be drawn from and embedded in communities, and restrained by informal “kinship and family” ties. These community restraints – i.e. community selection and oversight – were incorporated in the model, but in the debate over expanding local defense forces, critics successfully argued for greater formal oversight measures. As a result,  the different community-based forces that US SOF mobilized in 2009 were formally brought under the Afghan Ministry of Interior (MoI) in August 2010, and baptized the Afghan Local Police. With this, ALP came under the command of Afghan National Police in their local areas, and were subject to regular MoI disciplinary procedures and regulations. By mid-2012, there was a special monitoring and investigation unit (at p. 22-23) for the ALP, which regularly investigated allegations of misconduct, and referred members for prosecution or disciplinary action.

Although institutionalization was intended as a partial check, the US also adopted additional bilateral or external due diligence measures. US forces kept a close hold on the program even after it was formalized under the MOI, with US SOF directly selecting, mobilizing, and working with ALP units in the field for years after the program’s inception. This allowed a greater degree of US influence on vetting and selection. Initially SOF engagement and community consultations were supposed to informally screen out abusive actors and militias. At least by 2014, the US Leahy Law was also formally applied to the ALP (the same commander- or unit-based screening as for the rest of Afghan forces).

Training, mentorship, and standards of good conduct were also a big part of the ALP model from the earliest pilot programs. There were special rules and restrictions intended to restrain ALP from behaviour or deployment patterns that had led to abuse in the past – for example, preventing them from engaging in detention operations, or operating more than 1 kilometer from their home village. ALP were provided 21 days, and later a full four weeks (at p. 59) of training, including on IHL and human rights. SOF also would embed in the area for a number of weeks, which was intended to create a degree of informal oversight, mentorship and reinforcement of these standards. After a Human Rights Watch report offered documentation of rampant abuses, the Special Forces command in charge issued a new rule requiring that its forces report allegations of abuse by ALP and that where “probable cause” existed, US forces should cease contact with the alleged perpetrators.

While all of this suggests a fairly robust due diligence practice vis-à-vis the ALP, it is important to note that many of these checks were only minimally applied in practice.  Program requirements to get community approval and consultation for local forces were quickly abandoned when the decision was made to rapidly expand the program from 2010 to 2012. Restrictions on operating outside of the home community or on abiding by the code of conduct were routinely ignored, and forces not held to account. There is no evidence that the Special Forces cut-off rule was ever applied. Even after the Leahy law was applied, it is not clear how many (if any) abusive ALP commanders were blocked from funding; in general DoD was taken to task for over-using  exceptional carve-outs within the Leahy Law in Afghanistan to exempt large numbers of Afghan forces from funding blocks.

Iraq

When ISIL took control of substantial parts of northern Iraq in mid-2014, the US responded by supporting a range of state, substate, and nonstate Iraqi forces. In addition to supporting Iraqi forces and the Kurdish regional forces, the US tried to encourage a Sunni tribal uprising against ISIL, hoping to recreate the success of the 2006 Sons of Iraq or sahwa (“Awakening”) movement that had helped defeat Al Qaeda in Iraq. US funding for the Tribal Mobilization Force (alternately known as the Hashd ash-Shairi or “Tribal PMF”) began in early 2015 in Anbar province, and then expanded to Ninewa province. US enthusiasm for the program and support began to wane (at 22) with the end of active hostilities against ISIL.

In terms of checks and safeguards, the TMF structure in many ways resembled that of the ALP. They were officially an Iraqi force, with all weapons and equipment channeled through the Iraqi government. Once the law legitimizing the much larger Popular Mobilization Force was passed, they were technically legalized along with it and brought under its structures (although the force was kept somewhat distinct at least until 2019, given US hostility to the larger PMF leadership). However, even though “Iraqi-owned” the US kept close hold on the program and imposed its own demands and requirements. The result in terms of due diligence practice was a mixed model of some Iraqi institutional constraints and some measures imposed directly by the US.

On the Iraqi side, the TMF were subject to Iraqi background checks (and also Kurdish background checks for forces in Ninewa) and regular payroll and accountability procedures. They came under the direct command of  Iraqi forces – a three-star general based in Baghdad overall, and the nearest Iraqi Security Forces or Federal Police command at a local level.  They were subject to all the regular rules and disciplinary procedures of other Iraqi forces. In some cases, TMF who broke the law (mostly infighting among themselves, theft) were prosecuted and punished via regular Iraqi law enforcement processes.

On the US side, the TMF were subject to Leahy Law human rights vetting, and follow-on monitoring. US personnel at US embassies, consulates, and bases regularly tracked which forces were mobilized under the TMF banner, where they were deployed, whether they received training or intended support, and any allegations of misconduct or human rights abuses.  Some groups were cut from the program where there were credible allegations that they tortured or mistreated suspected ISIL fighters whom they captured, recruited underage minors into their force, or predated upon the population. US and other Coalition forces provided training with some material on laws of war and human rights obligations. However, training was not mandatory and many TMF were deployed without it.

Some of the same issues manifested with Iraqi tribal forces as plagued the Syria and Afghan assistance programs. Even if US forces and officials had greater access in Iraq than in Syria, they were still thinly staffed and largely confined to bases and consulates in Baghdad and Erbil. This made it difficult to observe or respond to issues with tribal forces operating in remote areas of Anbar and Ninewa provinces. While Iraqi institutional oversight provided some substitute for this, it also introduced greater political intervention and subversion of the program. Many of the larger forces overseeing the TMF – both the Iraqi security forces and later the Popular Mobilization Force – had their own problematic records of abuse and corruption. This wider atmosphere of impunity gave less deterrent effect to threats of US cuts if TMF engaged in abuses.

Conclusion: Due Diligence Checks as the Norm?

The starting question for this blog was whether we might advance the debate over due diligence measures for partner forces not by revisiting the ambiguous (and contested) legal standards, but by looking at approaches being tried in practice. This blog, and the larger report it is based on, has documented one approach to due diligence practice with nonstate or substate partners. The US has consistently tried to minimize the risk of human rights violations or war crimes by partner forces by 1) vetting for past abuses; 2) providing some form of training, codes of conduct, or other conduct-based ‘redlines’ or standards; 3) ensuring some degree of monitoring and tracking of supported forces’ conduct; 4) instituting sanctions or cut-offs where abuses manifest; and 5) (where the US is acting in alignment with the sovereign state in question) bringing nonstate partners under the control and oversight of (host) state institutions. Although still not a boiler-plate approach or fully formed ‘toolkit’, such safeguards appear to be a fairly normalized response when US policymakers are faced with the prospect of supporting risky and potentially abusive nonstate partners.  

So far, these checks and safeguards have tended to manifest only when the US is dealing with larger, more overt forces – for example, forces as large as the 30,000-strong ALP, the 20,000-strong TMF, or the 60,000-strong SDF and their affiliates.  Such forces are not only more visible but also are often supported as part of counter-insurgency effort, and there may be strategic dividends from ensuring good conduct. So far, US due diligence checks have been less apparent in its support to the sort of shadowy militias associated with US global counter-terrorism operations, however that may be changing. The Fiscal Year 2021 defense appropriation act for the first time introduced a human rights-related condition on the funding line that supports the “127 echo” fund, the fund that supports Special Forces ‘proxy’ or surrogate forces in counter-terrorism operations worldwide.

What would it mean if the sort of approach documented above was considered an appropriate response to Common Article 1 due diligence obligations? Although in each of these cases, there were substantial shortfalls in implementation, this level of checks and safeguards nonetheless represents a fairly proactive and robust form of due diligence. Diplomats of other European countries who were interviewed in the course of this research said that their countries did not undertake anything like US vetting and tracking of partner forces (state or nonstate), in part because they were not sure it was effective, but also because they did not have the capacity and resources to do so. A safer approach to avoiding the legal risks, they said, was to avoid supporting nonstate armed groups in the first place, at least in terms of lethal assistance.

While in some cases not providing support to a given force may be the best choice, avoidance is not a full solution. As noted in the first part of this blog, in some situations, Western states’ insistence on only working through state forces has created perverse incentives to whitewash problematic forces by granting them state title. Instead of addressing rights violations, this may only entrench unaccountable forces, and grant them an additional degree of impunity. In addition, while many militias and nonstate forces have a reputation for abuses, in many cases, state forces can be equally bad. The decision to provide support or not should thus not rest entirely on legal status.

Nonetheless, there are also cautions against embracing the sort of due diligence approach illustrated by past US practice in Syria, Iraq, and Afghanistan. In addition to being resource-intensive, many US officials and forces – including those who internally championed greater accountability measures – argued that these sorts of technical checks and safeguards had not shown strong results. For example, US military commanders tended to say that while they supported legal awareness and training, unless supported by broader operational support, mentorship, and reinforcement, and over a long period of time, these measures would have little effect. Others worried that such checks were not only negligible in their effects, but potentially counter-productive. One former State Department official worried that mechanisms like vetting procedures too often become a ‘check the box’ exercise: “It creates a sort of artificial confidence about the groups you’re working with, and stops the degree of policy consideration of whether working with these groups is good,” he observed.

Given the important role of nonstate or substate armed groups in 21^st century conflict, this issue is not going away and cannot be sidestepped by working only through state partner forces.  The debate about the scope of state responsibility for nonstate partners is still very much an open one. While examples from practice will not decide it entirely, what has worked or not should inform the legal discourse to a greater extent, in order to help develop due diligence policies that not only satisfy notions of state responsibility but also might result in better protection in practice.