How the U.S. Supported Rebels in Afghanistan, Iraq, and Syria
In her book, Erica Gaston outlines the perils and practices of these partnerships.
Over the last two decades, the United States has supported a range of militias, rebels, and other armed groups in Afghanistan, Iraq, and Syria. Critics have argued that such partnerships have many perils, from enabling human rights abuses to seeding future threats. Policymakers, however, have sought to mitigate the risks of partnering with irregular armed groups. Militia group leaders in these war-torn countries were subjected to background checks, instructed about international law and human rights, and their funding was cut when they crossed red lines. To what extent have such mechanisms curbed the dangers of proxy warfare, and what unforeseen consequences has this approach unleashed?
In Illusions of Control: Dilemmas Managing U.S. Proxy Forces in Afghanistan, Iraq, and Syria, Erica Gaston, an adjunct professor at Columbia SIPA and head of the Conflict Prevention and Sustaining Peace Programme at United Nations University Centre for Policy Research, draws on a decade of field research and hundreds of interviews with stakeholders. She unpacks the problems of attempting to control proxy forces, and demonstrates that, although the tools U.S. policy makers used to constrain partners’ behavior increased in number and sophistication, they never fully addressed the range of political, security, and legal concerns surrounding these forces.
Gaston discusses the book with Columbia News, as well as what she reads for relaxation, surprising books that line her shelves, and her ideal dinner companions.
What was the impetus behind this book?
In 2008-2009, I was working as a human rights advocate in Afghanistan, focused on issues of civilian harm. At the time, U.S. Special Forces were experimenting with mobilizing tribal militias, as a counterinsurgency force against the Taliban (then an insurgent force). It was immediately controversial. If there’s one country that might make you think twice about mobilizing militias, it’s Afghanistan—the long history of war crimes and abuses perpetrated by irregular militias there, their contribution to the collapse of the state and cycles of violence in the 1990s, not to mention the ultimate example of proxy blowback: When Afghan militias that the U.S. had supported in the 1980s later transformed into Al Qaeda under Osama bin Laden.
But for each of these potential risks and downsides, Special Forces had a mechanism or response: This time, there would be vetting and background checks, training in human rights, and a range of accountability checks, from embedding these forces in tribal structures, to having Special Forces mentor and monitor them, and placing them under Afghan institutions. It was almost like Special Forces was saying, “Yes, these are risky forces, but this time will be different,” because Special Forces had figured out how to mitigate or address the risks.
It was an intriguing idea, but one that didn’t pan out. Most of the risk mitigation measures that were proposed lapsed as soon as this militia force—later known as the Afghan Local Police—was authorized. Many held little promise of doing much even if they had been implemented.
But this wasn’t just an isolated incident. I then worked on related issues in Iraq and Syria, and saw some of the same back-and-forth in those countries—critics raising risks and concerns about backing militias, rebels, or other local forces, and proponents largely defraying these critiques by proposing a range of technical or tactical-level controls, which State Department bureaucrats and U.S. military commanders would then do their best to apply in complex and messy conflict environments.
As an academic, I was interested in how these experiences inform our assumptions and theories about proxy warfare and proxy control, and also what they say about the sort of bargaining, and the sort of players, that must be factored into decision-making models when it comes to 21st-century foreign policy. With my international lawyer and foreign policy practitioner hats, I wanted to know more about the effects of these checks and controls: Should we think of them as one approach to due diligence obligations under international humanitarian law? Finally, given that they have come to be a fairly prominent part of the American approach to security assistance, do they work? Is it possible to regulate irregular forces, in environments least disposed to control? And if it is not—as many of the examples suggested—what are the larger costs for policy-making?

What are some examples from the book of dilemmas the U.S. has faced trying to control proxy forces?
Many of the militias, rebels, and tribal or community forces the U.S. has supported come with past histories of grave human rights abuses, war crimes, and atrocities. Many may be part of extended criminal networks or linked to warlords, and thus represent more of a source of instability and conflict than a cure for it. Some had ongoing links with terrorist or insurgent groups. If the U.S. provided weapons, equipment, or other support to these groups, they might pass that support on to terrorist organizations, generating new security concerns and also legal liability under strict American laws about not providing “material support” to terrorists. They might even go rogue themselves, attacking U.S. trainers partnering with them, or later emerging as their own security threat—the risk of seeding another Al Qaeda, as happened with support to the mujahideen in Afghanistan.
U.S. policymakers and commanders were not unaware of these risks, so they tried to establish a range of checks and “control mechanisms,” as I describe them in my book. These included increasingly standardized vetting and background checks, setting red lines or standards, providing training in human rights, asking people to pledge not to violate international humanitarian law, monitoring and reporting processes, and tracking weapons. The CIA, for example, briefly forced Syrian rebels to collect spent missile casings before they would be provided with new ones, among other checks.
Overall, these various checks and controls fared about as well as academic theories on principal-agent dynamics and proxy control expect them to: They were extremely costly, difficult to implement, and, ultimately, extremely limited in their ability to mitigate risks. But the larger issue I highlight in my book was not just that the controls did not work, but that they generated a larger moral hazard: Because these control mechanisms offered the promise of addressing the risks or drawbacks of working with proxy forces—whether that was furthering war crimes, inadvertently supporting the Islamic State, or (in the case of U.S. support to Kurdish forces in northeast Syria), damaging the relationship with a key NATO ally (Turkey)—the mechanisms created a sort of false confidence, which allowed these initiatives to be authorized and to go forward.
In some cases—as with the dilemma between forgoing support to Kurdish forces (the only local forces deemed capable of defeating the Islamic State in Iraq and the Levant [ISIL]) or enraging Turkey—the checks and controls created a false middle ground between ultimately irreconcilable foreign policy choices or trade-offs. As viewed from this larger lens, the real challenge presented by such control mechanisms might not be the degree to which they fail to constrain risks, but that they worked all too well in terms of enabling risky choices in political decision-making.
What is the analytical toolkit for dealing with transnational bargaining and foreign policy deliberations that you outline in the book?
Bureaucratic Policy Analysis views foreign policy outcomes as the product of bargaining between different actors or players in the foreign policy process, and/or as the result of organizational scripts, routines, and processes. This paradigm has often been applied to illuminate foreign policy decision-making, and go beyond thinking about policies as the result of rational strategies made by a sort of unitary state actor.
However, the limitation of this classic bargaining model has been that it focused only on the different actors within the domestic arena—the president, and his circle of advisors, sometimes Congress or interest groups, but nothing beyond that. In some of the classic literature on this by Graham Allison and others, non-U.S. government actors are presented as the “stimuli,” which can generate a crisis that the president or the U.S. bureaucracy will react to, but not as players in the game.
This is a huge limitation when it comes to trying to analyze foreign policy dynamics in the 21st century. While arguably foreign players—from leaders or representatives of bureaucracies, to non-governmental organizations, or even non-state armed groups—likely always had greater interaction with U.S. foreign policy decision-making than these classic models represented, this is truer in today’s more globalized, interconnected world. Whether in the first scenario I described of bargaining for different checks on the Afghan Local Police, to the range of foreign policy leaders, interest groups, and members of foreign intelligence agencies that shaped CIA support to Syrian rebel groups, to understanding how a foreign policy decision was reached and how it devolved in practice, we have to consider the range of transnational players and inputs that go into such a decision.
What do you read for relaxation?
Reading (and writing) make up quite a bit of my day-to-day work—on fascinating, but often quite heavy subjects. So for relaxation, I gravitate more to non-fiction. Non-fiction can also be serious, of course, and deal with weighty subjects, but even in these cases, it tends to do so in a more creative or narrative way than what you might find in academic or policy writing. Also, because I do read and spend so much time in front of a screen in my work life, I have learned to embrace the audiobook, because I can get out and about with it. There is nothing like having a good audiobook to keep you company on a daylong bike ride or trek.
What books might people be surprised to find on your shelves?
A quite random collection of everything from Harry Potter to the Hunger Games to a collection of Greek myths in a grab-bag of foreign languages. I find light reading to be a great way to keep up with foreign languages, so I will happily pick up the latest Tom Clancy spy novel or Nicholas Sparks romance novel in French or German if I come across it at the airport, or in a used bookstore. If a book seems like it will keep my attention, and expand my foreign language vocabulary in a way that is not too taxing (my knowledge of magical words in French and German is amazing, thanks to Harry Potter), I’m fairly undiscriminating!
Which three scholars/academics, dead or alive, would you invite to a dinner party, and why?
Philosopher Richard Rorty, historian and anthropologist Nancy Hatch Dupree, and Columbia professor, philosopher, and literary critic Edward Said—all for different reasons, but also because I think they’d have a fun, spirited debate together.
As an undergraduate at Stanford University, I was lucky enough to have a small seminar with Richard Rorty. Although he was already a legend in his own time, he was open to having a discussion with anyone, and knew how to bring out debate and constructive dialogue, even with the most reluctant and shy undergraduate. All of which tells me he’d be a fantastic dinner party guest. I also recently came across some of his essays. In many ways, Rorty anticipated the current state of polarization that we’re experiencing, not just in the U.S., but also in many other parts of the globe. I think he’d have a lot to say on current dynamics, but in a more profound way than is typical of dinner party conversation or public debate.
I met Nancy Hatch Dupree over my many years of working in Afghanistan. Known as the grandmother of Afghanistan, Dupree (who was American by birth) first went there as a diplomat’s wife in 1962. She fell in love with Afghan culture and history (and another American archaeologist), and her path turned toward exploring and documenting the country’s rich, if often tragic, history. She continued to live in Afghanistan throughout most of her life, until her death in 2017. I remain very attached to Afghanistan and concerned about its current situation, so I would relish hearing her perspective on the turn of events there since the Taliban takeover in 2021. Dupree also had a wicked sense of humor, which would add sparkle to any dinner party.
I did think to include one scholar in my academic discipline, and though Said was not explicitly an international relations scholar, his work was transformative for this field because it challenges us to question the way we think about the assumptions embedded within the study of foreign relations, and the way we understand global power structures and struggles. A dinner discourse with him would also be incredibly timely, in light of the recent global protest movements that have sprung up surrounding the situation in Gaza.