Why Militaries Support Presidential Coups

In July 2021, Tunisia’s duly elected president, the constitutional scholar Kais Saied, froze the country’s parliament, dismissed its prime minister, and then rewrote the constitution to greatly expand presidential powers. Saied’s self-coup effectively ended Tunisia’s post–Arab Spring experiment with democracy, a decade after the Jasmine Revolution began.

Saied’s presidential putsch is just one of many to have taken place across the world in the last ten years. When self-coups happen in democracies — as in the case of Tunisia — they more often than not mark a turn toward autocratic rule.

While self-coups are a form of democratic breakdown distinct from military coups — incumbent executives attempting to seize more power for themselves versus military putschists aiming to overthrow a sitting government — focusing on this distinction can obscure the fact that militaries sometimes play important roles in self-coups. France’s Napoleon III in 1851, Peru’s Alberto Fujimori in 1992, and Tunisia’s Kais Saied in 2021, among others, deployed the military in their autogolpes to help shut down parliaments, repress protests, or prosecute opponents in military courts.

Moreover, in other cases, self-coup attempts failed when militaries refused to play these roles. In Guatemala in 1993, Indonesia in 2001, and, most recently, Peru and Brazil in 2022, presidents were unable to broaden their powers or extend their stay in office after militaries refused to back their takeovers. Some political scientists contend that “what separates successful self-coups from those that fail is whether the military backs the undertaking.”

Why would a military support a self-coup? Which types of militaries are more likely to do so? As I show in my book Soldiers of Democracy? Military Legacies and the Arab Spring (2023), the type of military that stages a coup of its own is distinct from the type that facilitates a self-coup: It is the powerful and politicized militaries that most often mount their own coups, while relatively weaker and apolitical militaries tend to facilitate self-coups.

Two mechanisms underlie this trend: the military’s self-interest and its degree of professionalism. Generally, it is easier for a president to entice into his self-coup a military that has historically been neglected than one already enjoying the lion’s share of power and wealth. For Tunisia’s long-marginalized armed forces, for example, gaining even one ministerial position represented a major win. Ahead of Saied’s self-coup, he appointed a military doctor as minister of health, empowered the military to take the lead in the covid-19 response, and promoted the top officers to military ranks they had rarely seen. These institutional and personal incentives were one reason why the military obeyed his orders to shut down the parliament.

Weaker, more apolitical militaries that have been kept far from power are also more likely to have internalized norms of civilian control. So if the commander-in-chief orders such forces to shut down parliament, their sense of professionalism may impel them to follow those orders and to see refusing as too political. As one Tunisian Army General I spoke to mused, “To say ‘no, this is unconstitutional’ would have been intervention into the political arena. Is that acceptable? Would that have been better for the country?”

By contrast, militaries that have historically been more politicized have had fewer qualms saying no to a president, as in Guatemala, Indonesia, and Brazil. Although deference to the civilian president can help prevent military coups, too much deference can be dangerous when that civilian attempts to draw the military into politics.

Countering Self-Coups

The international community can do more to discourage militaries from supporting self-coups. While some Western donors and regional organizations have laws on the books to suspend aid in the event of military coups, most do not in the event of self-coups. The United States, for its part, did recently expand its coup clause beyond “military coups” to also include civilian coups “in which the military plays a decisive role.” This clause should be expanded further to include executive aggrandizement and not just ousting the elected leader.

Moreover, the United States needs to consistently and quickly enforce this clause in response to all types of coups. The prospect of losing much-needed foreign aid will help militaries to realize that facilitating self-coups might not actually advance their interests. After Saied’s self-coup, the United States did cut military aid to Tunisia, but only gradually and only in half — never enough to change the military’s or the president’s calculus. In Brazil, by contrast, it was reportedly U.S. pressure that decisively convinced the generals not to back Jair Bolsonaro’s attempts to cling to power in 2022.

Second, the international community should signal to militaries that facilitating a self-coup is indeed unprofessional behavior, and provide guidance on what to do when a president gives them quasi-legal but undemocratic orders. In Tunisia, the United States has done neither, with the State Department even certifying to Congress that the Tunisian military remains an apolitical and professional force despite shutting down the parliament. The United States should include in its International Military Education and Training programs guidance on how militaries should react when civilians attempt to drag them into politics.

Critical to that effort will be the development of judicial channels for assessing whether orders are illegal or undemocratic. Military officers cannot be expected to judge the legality or constitutionality of all orders themselves. As an open letter by thirteen former U.S. secretaries of defense and chairmen of the joint chiefs emphasized: “Civilian control is [also] exercised within the judicial branch through judicial review of policies, orders, and actions involving the military.” Establishing the equivalent of the U.S. Defense Department’s Office of General Counsel within defense ministries as well as channels for both militaries and parliaments to refer orders to competent judicial authorities can help to remove politics from a military’s decision not to support self-coups.

The international community must recognize that a self-coup presents a fundamentally different challenge than a conventional coup: The danger is a military that is overly obedient to a civilian, not one that flouts civilian control. Countering self-coups, then, will demand new methods. When an executive attempts to drag them into politics, militaries must be equipped to say no.