Where in the Chain of Command Do Security Force Abuses Originate?

Recently, two pieces that touch on security force abuses in the Sahel caught my eye.

One is Héni Nsaibia’s excellent piece for ACLED, “State Atrocities in the Sahel: The Impetus for Counterinsurgency Results Is Fueling Government Attacks on Civilians.” An excerpt:

Ahead of the end of the rainy season in August 2019, ISGS [the Islamic State in the Greater Sahara] and JNIM [Jama’at Nusrat al-Islam wa-l-Muslimin, the Group for Supporting Islam and Muslims] – in tandem – launched an offensive in the tri-state border area, also known as the Liptako-Gourma. It was a campaign in which military outposts were overrun like dominoes, forcing government troops to tactically withdraw from the border areas and leave previously contested territory under militant control. These developments underscored the lack of cooperation and coordination between the constituents of the struggling regional G5-Sahel force, for years promoted as an effective coalition to address the jihadi threat (Ouest-France, 2018). Amid mounting popular discontent in Mali over the presence of foreign forces as Malian soldiers were killed in scores (RFI, 2020), French president Emmanuel Macron summoned the leaders from the G5-Sahel countries to clarify their positions on France’s role in the Sahel. During the summit convened on 13 January 2020 in the French town of Pau, a roadmap was outlined to counter the jihadi onslaught (The Conversation, 2020). France decided to deploy 600 supplemental troops to its Barkhane mission (Le Monde, 2020), ensued by the official launch of ‘Takuba’, a task force gathering special forces from several European countries aimed at shoring up Barkhane and Malian forces in the fight against jihadi groups (Ouest-France, 2020). The primary focus of the counter-offensive was to be ISGS, now the ‘Greater Sahara’ faction of the Islamic State West Africa Province (ISWAP). Although the group carried out some of the deadliest attacks targeting security forces to date, this announcement largely neglected the comparable threat posed by its Al Qaeda counterpart, JNIM (Liberation, 2019; DW, 2020).

In the wake of the Pau meeting, state violence targeting civilians increased in all three countries as local and foreign forces stepped up their operations. If the Pau Summit did not encourage civilian targeting, it evidently appears to be a direct consequence.

The other piece that struck me was Judd Devermont and Leanne Erdberg Steadman’s “Defending the U.S. Military Presence in Africa for Reasons Beyond Counterterrorism,” published at Lawfare. Obviously, from the title, their article is only partly about security force abuses, but here is what they said on that score:

In places like Burkina Faso, Mali, Nigeria and Somalia, U.S. forces and partner militaries can support longer term efforts to prevent violent extremism from taking hold in the first place. The United States can help partners set up joint operations centers where, in real time, the U.S. military can showcase how intelligence-driven operations reduce accidents, lower civilian casualties, and foster information sharing (including with gendarmes and police) that saves lives. And it can do more to promoteopen dialogues with communities on threats and prevention strategies.


If the United States wants to reduce the threat of terrorism in Africa, partnerships need to mean much more than just limited counterterrorism objectives. In too many African contexts, terrorists easily step in to establish themselves as a viable alternative to the government when those in power are corrupt, venal, and direct security forces to kill and abuse civilians for political gain. [Emphasis mine – Alex.] In those cases, the U.S. military’s most effective intervention is assisting partner nations to improve their behavior and rebuild trust. Because when U.S. forces support efforts to repair the broken bonds of governance and positively impact the behavior of African security forces, it reinforces the idea that the state’s job is to protect people, instead of going after them. And that might end up being the most potent type of counterterrorism the United States can ever help with.

These posts got me thinking: Where exactly do security force abuses originate within the chain of command? I am definitely open to the argument that they are top-down. Nsaibia builds a strong, if circumstantial, case for the effect of the Pau summit. But I find Devermont and Steadman’s characterization too strong – I don’t think security force abuses are necessarily directed from the top, and I don’t think that when civilian leaders order or tolerate abuses, it is necessarily because they see political advantage in them. Sometimes leaders may be cognizant that an abuse-laden strategy of collective punishment is a poor approach – they may even realize that is counterproductive – but then acquiesce to it or order it anyways, perceiving themselves to have no alternative. None of this is to exculpate the Sahel’s leaders – but I am not sure that individual-level venality, or even the venality of different countries’ political classes as a whole, has that much explanatory power.

Unfortunately, I couldn’t find any academic literature on these questions in my initial searches on Google Scholar (that doesn’t mean it’s not out there!).* If any readers know of relevant work, please point it out.

Here are a few points that occurred to me. As a provisional framework, it made sense to me to think about five levels where the decision to commit abuses might originate: the individual soldier, the unit, the theater/campaign, the top military leadership, and the civilian leadership. Building on Nsaibia’s piece, one could add a sixth level – the foreign patron government – if one wants to be cynical about who takes orders from whom, or at least who responds to pressure from whom.

1: There Is a Difference Between Top-Down Systems of Abuse and Systems of Impunity; But Impunity for Major Abuses Runs, By Definition, All the Way Up the Chain.

The analyst cannot always be sure where abuses originate within the chain of command; the data are often limited and hard to confirm, and the workings of any given military chain of command and military culture can be opaque to outsiders. It is worth it, though, to make an analytical distinction between “systems of abuse” and “systems of impunity,” because they might look roughly the same from a distance but have very different internal mechanisms and very different policy ramifications (more on this below). Just because abuse is widespread does not necessarily mean there is a top-down, centrally-directed system of abuse; not everyone in the chain of command is automatically implicated in the abuse itself. Yet impunity for abuses does implicate everyone, by definition, because even if the top military leadership or the civilian authorities are not ordering abuses, their acquiescence is needed to create a system of impunity. Consider three possibilities:

  1. Civilian authorities create a system of abuse and a system of impunity;
  2. Abuses originate down the chain, so there is not necessarily a system of abuse, but civilian authorities provide a system of impunity in response or by default;
  3. Abuses originate down the chain and civilian authorities punish abusers.

2: Widespread Individual-Level Abuses Point to a System of Impunity

This is an extension of the previous point. Even if, for the sake or argument, all abuses in a given country were generated at the individual level (for example, the soldier pressuring a woman for sex in an IDP camp), I think that if such abuses are widespread it indicates that a system of impunity, by design or by default, is in place – otherwise most individuals would not continue to abuse. In other words, the “bad apples” argument cannot be sustained if a lot of individuals are committing abuses.

3: Unit-Level Abuses Can Be Self-Generated or Directed from Above

When a particular unit becomes infamous for committing abuses, that in and of itself does not necessarily tell us whether there is a wider system of abuse in place. On the one hand, the unit could be a “rotten egg” whose abuses are known and deplored, but which is also provided impunity by superiors (military or civilian) in the chain of command. Perhaps the unit is particularly feared, or the costs of disciplining it are perceived as too great, and so superiors do not act. On the other hand, an abuse-prone unit may in and of itself constitute a top-down system of abuse if superiors deliberately leverage that unit for a particular purpose, “letting it off the leash” while keeping other units restrained – or attempting to play “good cop, bad cop.”

4: Theater-Level Abuses Would Be Difficult to Sustain Without Buy-In from the Top Military Brass

As discussed above, if one finds a single unit regularly committing abuses, it’s plausible under certain circumstances that the unit itself is generating those abuses without orders from above. But a pattern of abuses throughout an entire theater of operations – say, northeastern Nigeria – is very unlikely to persist without approval and probably direction from the top military brass. That is, in a context where multiple units are committing abuses within the same theater, and where the pattern persists despite rotations in theater commanders and other key personnel, then a real system of abuse very likely exists all the way up and down the chain of command.

5: Theater-Level or Military-Wide Abuses Can Be Generated by Commanders or by Pressure from Civilian Leaders; If the Former, Civilian Leaders Are Likely Cowed

If an entire theater, or an entire military, is characterized by abuses, this automatically implicates the civilian leadership in a system of impunity, but it does not mean that civilian leaders are directing the abuses. Civilian leaders could, conceivably, be deploring the abuses but feel powerless to stop them – perhaps out of fear of a coup, or through more subtle forms of intimidation that militaries can bring to bear against civilian leaders (strategic leaks of government deliberations, for example).

6: Civilian Leaders Can Encourage/Order Abuses Enthusiastically or Reluctantly

Now we are discussing situations where the system of abuse is truly systemic, and runs not only up and down the military chain of command but includes the civilian leaders as well. Even in such situations, however, there are still multiple variants. Consider a scenario like that suggested by Devermont and Steadman above: Civilian leaders enthusiastically order abuses in the pursuit of perceived political gain, punishing political opponents and intimidating the population writ large. But other scenarios are possible: civilian leaders, particularly in states with few resources and where the central state’s knowledge of restive peripheries is thin, might feel that they have no real option other than to “crack a few heads.” Civilian leaders may be influenced by select interlocutors from the conflict zones, who promote harsh approaches, demonize particular communities, and sell the central authorities on a strategy of deliberate abuse. And then there is the scenario that Nsaibia suggests – pressure from above, in other words from a major foreign power.

7: The Net Effects of Unit-Generated Abuses Combined with Systems of Impunity, Versus Top-Down Systems of Abuse, May Appear Similar – But the Policy Responses May Have to Be Different

This takes us back to the discussion in points 1 and 3 – it is not always easy to tell how deliberately or systematically the civilian leaders are promoting abuses. The net effects of different scenarios may look the same: in other words, widespread abuses and suffering. If widespread abuses are encouraged up and down the chain, then the policy response by international actors may need to involve substantial pressure on the actors at the top of the chain to make them cut off the flow of orders promoting abuses. But if one concludes that civilian leaders are only reluctantly allowing impunity for abuses, that civilians are hostages to the military leadership on this issue, or that the top military brass’ role in the abuses is ambivalent because of fear of a particular unit or of backlash from the rank-and-file generally, then the policy focus may need to shift elsewhere – to creating conditions that would give space to civilian leaders to discipline those abuse-prone units, etc.

*There is, obviously, a wide literature on civil-military relations. My superficial sense is that much of the literature on civil-military relations in Africa concerns coups. Then there is the massive NGO literature on incidents of abuse, and I am implicitly drawing on some of that here, but I have not seen an NGO report present a systematic typology of where and how abuses occur; impressionistically, many reports seem to describe abuses at what I am calling the unit and theater levels.

2 thoughts on “Where in the Chain of Command Do Security Force Abuses Originate?

  1. Pingback: Burkina Faso: The Apparent Summary Executions at Tanwalbougou and Wider Questions of Accountability | Sahel Blog

  2. Pingback: Niger: A Devastating Report on Summary Executions of Civilians in Tillaberi, and a Hard Denial from the Ministry of Defense | Sahel Blog

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s