If You’ll Be My Bodyguard: Presidential Guard Units and Leader Capture During Coups d’état

Abstract

Coup “handbooks” emphasize the capture the incumbent leader as a key objective for plotters to enact successful regime change. However, the literature has yet to empirically assess this relationship. We also lack a robust understanding of how leaders prevent their own capture during coups d’état. Using novel data on leader statuses during coups from 1950 to 2017, I find evidence that incumbent leader capture has a positive and significant relationship with the likelihood of coup success. The findings also suggest that leader capture will be less likely if the regime pre-emptively creates a counterweight presidential guard unit, responsible for providing proximate security for the core leadership. These data and findings provide new insights into the dynamics of in-progress coups, focusing on the operational strategies employed by both sides. It also demonstrates the diversity of independent effects that specific types of counterweight forces have on specific coup outcomes, encouraging further study in this area.

Keywords

coup d’état political violence civil–military relations

Introduction

The study of civil–military relations is a massively growing field, building upon foundational works of theory that explore the political and sometimes hostile dynamics between soldiers and civilians (Brooks, 1999; Feaver, 1996; Finer, 1988). Coups are forms of rebellion by which the military or other regime insiders attempt to illegally seize power from the incumbent government, sometimes with violent force but often without (De Bruin, 2019; Powell & Thyne, 2011). The publishing reaction by the scientific community may unfortunately be linked to the recent resurgence of coup around the world. From 2019 to 2023, military coups have overthrown both democratic and autocratic governments in Myanmar, Guinea, Chad, Sudan, Mali, and Burkina Faso. The mounting threat of coups has even prompted leaders in the Economic Commission for West African States to discuss forming a collective peacekeeping force tasked with helping members to resist these events (Onuah, 2022).

Historically, many works have studied the conditions under which coups are more likely to occur, aiming to facilitate better prediction by studying the social conditions that can prompt militarized seizures of power (Perlmutter, 1977). The more recent literature has brought an elevated empirical approach to this predictive study. To name a few, scholars have found that military coups are often the outcome of unbalanced power-sharing arrangements between regime elites (Frantz & Stein, 2017; Kim & Sudduth, 2021), contentious civil–military policies (Brooks & White, 2022; Matthews, 2022), and the personalization of security forces (Song, 2022). Despite the abundance of literature on predicting coups, few have focused on what makes these events more likely to end in regime change after they have been launched, during the coordination game known as the “execution” phase (Casper & Tyson, 2014; Little, 2017). What do rebel soldiers pursue during this phase to improve their odds of success? What can loyalist forces do to prepare for this situation?

What we do know of the execution phase revolves primarily around the identities of plotters. Specifically, that the military ranks of plotters bestow them with personal and professional experiences that they can leverage to help gain strategic advantages during a coup attempt (De Bruin, 2019). Middle-ranking officers (like majors or lieutenant colonels) have combat experience and direct control over troops, allowing them to deploy quickly and leverage command experience to their advantage (Singh, 2014). General-rank officers have command authority over thousands of soldiers and middle officers, exploiting the chain-of-command to order these forces (sometimes without their prior knowledge) to enact regime change. These generals also have insider knowledge of the political system, which may give them a strategic advantage in knowing the regime’s counter-coup planning systems (De Bruin, 2018).

Luttwak’s (1968) “practical handbook” on coups stands out among the literature in placing a large focus on the operational dynamics that take place after coups have begun. Beginning in the planning phase of a conspiracy, Luttwak argues that plotters should prepare their forces for an execution phase strategy that will optimize their chances of forcing the regime to surrender and hand over power. He argues that securing this surrender depends on whether the rebels can convince loyalists that the coup is unstoppable by securing critical strategic points around the capital. Each of these strategic points carries different benefits for the plotters if captured. For example, occupying the telecommunications stations (radio or television) allows plotters to both cut off an important line of communications for the regime to reach supporters, while also giving rebels a platform from which to transmit their own propaganda. Other works have argued that this propaganda helps the plotters to “make a fact” of the coup’s eventual outcome, using symbolic and authoritative language and imagery to convince third parties not to enter on the side of the regime (Bleck & Michelitch, 2017; Crabtree & Kern, 2018; Singh, 2014).

Along with capturing telecommunications stations, Luttwak (1968) also suggests that plotters will want to capture key capital infrastructure and symbolic buildings: bridges or intersections, the capital airport, the parliament building. Cutting off bridge entry points and major traffic intersections allow plotters to control the movement of their own troops around the operational zone. It also gives them a defensible chokepoint for keeping loyalist troops away from strategically important locations. Securing the airport helps rebels to both ground loyalist air support, while potentially delaying loyalist reinforcements from being transported into the conflict zone. Storming the parliament building not only gives rebels visual control of a place of important political symbolism (another propaganda advantage), but any politicians barricaded inside can also be detained as hostages (Krebschull, 1994; Roessler, 2011). Legislators can be ransomed or used as shields to force loyalist forces to surrender. Even if the attempt goes poorly for the plotters, these high value prisoners may be negotiation pieces to help secure a safe surrender, making them valuable under any circumstance.

Before they even attempt to seize these physical locations, Luttwak (1968, pp. 122–23) argues that coup plotters will first want to ensure that their troops have a plan to capture the most important objective of all: the regime’s leader. The president, king, prime minister, party general secretary, or chair of the ruling military junta. This individual is ultimately the figure that the rebel forces are aiming to oust with their regime change coup plot (Powell & Thyne, 2011). Thus, executing an effective plan to capture the regime leader can bestow many advantages upon the plotters, easing their way to a successful seizure of power. Luttwak (1968) even argues rather plainly that the initial strike of a coup should be from a special squad tasked with seizing the regime’s leader, suggesting that the element of surprise is best spent in decisively securing this critical objective. With the leader in plotter hands, the coup is not yet over, as loyalists can still resist. However, leveraging the detained leader can help to end that resistance, as plotters may force this figure to publicly broadcast their own resignation and order to their forces to stop resisting. Seeing the inability of the regime to even protect their top figure, loyalists may accept this stand-down order, thus securing a successful outcome for the coup leaders.

This relationship between leader capture and coup success is perhaps an intuitive one. Despite this theoretical presumption, the literature has yet to verify whether this is an empirically demonstrable relationship at the cross-national level. This is perhaps due to a lack of data on the internal dynamics of coups during the execution phase, with the quantitative focus instead being placed on studying coups onsets and aftermaths. Furthermore, without this data we are unable to study the connection question of how regime leaders effectively defend themselves from capture after a coup has started. What forces can be prepared and mustered to deny this critical advantage to the plotters, perhaps making regime change less likely? These shortcomings have left us with little understanding of the dynamic factors that make coups more or less likely to succeed after they have been initiated, when prevention and plotting are put to the test.

This article aims to address these theoretical shortcomings and create an empirical record of the effect of leader capture during coup attempts. I draw upon new data on regime leader statuses during coup attempts worldwide from 1950 to 2017 to assess this relationship. Data on leader statuses were collected using a wide range of source materials, including (but not limited to) firsthand media accounts and secondary narratives from scholarly observers. Using these regime leader status insights, I also extend the focus beyond the success outcome relationship to also explore how these targeted leaders avoid capture through prior institutional design. I argue that the establishment of presidential guard units, acting as counterweights to the coercive force of rebel soldiers, are effective coup-proofing tools for ensuring that regime leaders do not fall into the hands of the plotters. Presidential guard units protect the regime leader through their capabilities and traits of being armed, organizationally independent from the military, and proximately located to the leader’s residence to allow for rapid defensive deployment. By being on site and ready to hold attackers at bay, presidential guards can deprive plotters of the key benefits that come from capturing the regime leader, thus reducing their overall success likelihood. Taken in tandem, these findings provide a more nuanced story of how coups progress from start to finish after being launched, incorporating a new factor of analysis that describes how coups are fought before they are settled.

Incumbent Capture as a Strategic Objective

Plotters during coups will seek to capture various “objectives” around the capital at the onset of a regime change attempt, executing complex strategies to ensure that these areas are held by their forces rather than remaining under the control of the loyalist side. Rebel plotters will undertake this risky effort because these critical locations provide important strategic benefits to the occupier that may improve their odds of successfully seizing political power. One such objective involves capturing the airport, blocking off runways to halt potential loyalist reinforcements from entering the capital zone by air, along with grounding the air force if they are not on the side of the conspiracy (Krebschull, 1994). Plotters will also attempt to seize radio and television stations, aiming to sever the regime’s communication lines and using these facilities broadcast their own propaganda that shapes the narrative around the events going on (Lee, 1976). Securing each of these strategic objectives gives the plotters certain advantages and deprives the loyalist forces of those tools, lowering their ability to effectively resist regime change.

Each of these objectives discussed has so far been physical facilities, but individuals can also serve important purposes for both plotters and loyalists during the contestation of a coup game. One of the most critical individual objectives that both sides will want to control (through defense or detention) is the very person whose removal is the key objective underlying the coup attempt: the regime leader. This person is the effective head of the entire political system of the country, being a president, prime minister, monarch, general secretary, or chair of a military junta (Goemans et al., 2009). So vital is this person to the overall outcome of a coup that Luttwak (1968, pp. 178–179) suggests that plotters should expend their first strike advantage during a coup on a lightning attack to capture the regime leader before shifting to seizing other objectives like the airport and radio stations.

Plotters should ensure that a specialized capture squad immediately detains and keeps control of the regime leader at the coup’s onset, as this allows the plotters to extract powerful propaganda advantages early on that can more quickly decide the event’s outcome. Similar to the previously theorized importance of telecommunications stations, credibly demonstrating that the leader is under detention can make “the victory of the coup attempt seem inevitable and resistance futile, thus creating self-fulfilling expectations around the coup’s success” (Singh, 2014, p. 15). Making a fact of victory dampens the desire of loyalists to fight on, while also disrupting the regime’s chain of command. This disruption lowers the ability of senior figures to credibly issue orders and to radio uninvolved military units to for reinforcements. Even if uninvolved units can be reached, their commanders may determine that siding with the regime is a futile effort, electing to join the plotters or just stay in their barracks.

If rebels capture the incumbent, it could help them “head off potential resistance” by using the physical person of the incumbent for a propaganda advantage (De Bruin, 2022, p. 5). Successfully securing the leader means the plotters can disseminate photographs and video of the capture, using the visual narrative to demoralize loyalist forces and signal that the coup’s success is a foregone conclusion. Even if obviously made under duress, publicizing forced abdications by captured regime leaders may help plotters to smooth the way for their transition. The image of a leader’s resigning on live television while flanked by soldiers can be a powerful visual to the population of a country, signaling that the success of the coup has been settled and demoralizing any remaining loyalists who may still desire to resist.

During the 2020 Malian coup d’état, this strategy was used against President Ibrahim Boubacar Keita (Cole, 2020). The unprotected president was swiftly captured by rebel soldiers and forced at gunpoint to read a resignation statement on national television. His disheveled state and distressed words sent a powerful signal to those watching that victory for the National Committee for the Salvation of the People (the officers leading the coup) was inevitable (Reuters Staff, 2020). Keita expressed the hopelessness to his supporters over the airwaves: “If today, certain elements of our armed forces want this to end through their intervention, do I really have a choice?” (Diallo, 2020). Officers in the National Guard, an internal troop unit ostensibly charged with guarding political buildings, elected to only protect international embassies and otherwise remained passive to the political situation (Andreasen, 2023, p. 44). Keita indeed had little choice, resigning at gunpoint. In the days that followed, the CNSP solidified their hostile takeover of political power and achieved a successful regime change.

By arresting the incumbent, plotters also ensure that the leader will also not be able to serve as a coordinator for loyalist forces, lacking the capability to rally their supporters to resist in the streets. Detaining the incumbent may also can sway neutral military units toward the side of the rebels, by creating cascading defections (Luttwak, 1968, pp. 122–123; Singh, 2014). If neutral units know the incumbent is in captivity, they may be less likely to come to their aide, not wanting to risk mobilizing against their fellow soldiers on behalf of a government they perceive to be on its way out. For example, after the arrest of President Reid Cabral of the Dominican Republic by a 1,000-man group of Army coup plotters on April 25, 1965, most military forces in the country remained passive (Crandall, 2006, pp. 56–57). They waited as bystanders, not wanting to risk intervening for the detained Cabral when it appeared that regime change was inevitable (Lowenthal, 1973, p. 357). Even the National Police, organized as a counterweight to the armed forces, ended up joining the rebel forces or simply stripped their uniforms and disappeared into the crowds (Lowenthal, 1970; Palmer, 1989, pp. 19–20). Cabral, at the same time, was unable to call for outside military assistance, lacking the capacity under detention to make an effective appeal for their intervention.

Based on these assessments, we should expect that capture of the regime leader during a coup has an important effect on the event’s outcome. If captured, the plotters receive benefits that make successful regime change more probable. If the regime leader remains outside their hands, the plotters lose these potential advantages, suffering more obstacles to their seizure of political power. This logic motivates the first hypothesis:

Hypothesis 1 (H1): Capturing the regime leader during a coup d’état makes success for the rebel forces more likely.

Considering the dire effects that leader capture can have on a coup’s outcome, what can the regime do prior to an attempt that will improve their side’s chances of political (and sometimes personal) survival? Are there any coup-proofing tools that designed to ensure this situation is less likely, even if the military has decided to launch a strike against the state? The following section explores this question, centering on one particular type of institution designed to ensure the safety and security of regime leaders during times of unrest. These powerful defensive tools are presidential guard units, which are trained and tasked with being the leader’s proximate last line of defense, the final bodies standing them and a rebel capture squad during a military coup.

Presidential Guard Units and Incumbent Defense

Prior research has found that many leaders (particularly dictators) rely on state security forces as tools of survival, being particularly effective at resisting coups in progress (De Bruin, 2018; Janowitz, 1977). In the event of a coup, security forces like militias and militarized police can be deployed to defend vital points, guarding strategically important areas in the combat zone (Böhmelt & Clayton, 2017; Escribà-Folch et al., 2020). Using coercive capabilities that counterbalance the military, these security forces ensure that rebel plotters do not go unchallenged. Luttwak (1968, p. 68) emphasizes how even if security forces are outnumbered, “even one single formation loyal to the regime could intervene and defeat the coup attempt.” This makes a limited presence of state security forces an effective coup-proofing tool for regime leaders.

For state security forces to most effectively threaten or use coercion against coup forces, they must have been specifically designed as a “counterweight” force (Horowitz, 1985). The first qualification for being a counterweight is whether the security force is stationed in or very near the political capital of the state (De Bruin, 2020b). This proximity allows them to rapidly deploy to strategically important and politically symbolic locations if a coup launched, preventing plotters from seizing them for various conflict advantages (Talmadge, 2015). The second feature is that these forces are organized outside the direct command of the military hierarchy. Instead, counterweight security forces are directly commanded by a ruling party, non-military state apparatus, or even the regime leader personally (Greitens, 2016; Song, 2022). This choice helps to ensure that the counterweight force’s command authority is not linked to potential plotters within the military, who could credibly command them against the regime if they possessed this authority. Instead, they are recruited and organized to be “parallel” to the military, designed to view them as antagonists that could potentially put an end to their preferential status if the patron leader is ousted (Quinlivan, 1999).

Of the different types of counterweight state security forces (militias, secret police, etc.), only one is designed to specialize in last-line protection in close proximity to the core VIPs of the regime: presidential guard units. De Bruin (2020c, pp.5-6) qualifies presidential guards as “small, elite units tasked with the physical protection of the executive and other members of the regime.” These variants of counterweight security forces have long histories of effectively safeguarding leaders in both democracies and dictatorships, often acting as the immediate armed defenders of core leadership during military coups and other violent events (Barany, 2011). They are able to effectively accomplish their missions through specific design features, which make them the best positioned among other security force types to prevent regime leader capture, even if challenged with coercive force by military coup plotters.

Presidential guards typically station within or adjacent to the executive residence, positioning them physically close to the incumbent. This locational advantage allows them to rapidly deploy during times of emergency to quickly secure the regime leader’s person. They can also create a perimeter around the VIP by arming from on-site armories and stationing pre-arranged defensive positions (Roessler, 2011). Barany (2013) notes that presidential guard units often enjoy equipment and professional advantages over the regular armed forces, which helps to make them more effective coup-fighters and creates an incentive for them to credibly protect their regime leader.¹

An example of the protective powers of presidential guard units can be observed in the failed 1960 coup in South Vietnam against President Ngo Dinh Diem. Early on November 11, the presidential residence (Independence Palace) was attacked by Col. Nguyen Chanh Thi and LTC Pham Van Lieu, who mustered a combined-branch force to breach the residence and capture Diem (Michaels, 2014; Moyar, 2006). At the same time that Independence Palace was being surrounded, plotters executed other “textbook” coup strategies: seizing broadcast stations, occupying the army headquarters, and closing the airport (Associated Press, 1960; Nevard, 1960a). But despite their advances elsewhere in Saigon, the rebel squads of Cols. Nguyen and Pham faced a slowdown, rebuffed from Independence Palace by gunfire from Diem’s elite presidential guard (Nevard, 1960b). The guard had been alerted to the rebel forces converging on the capital, swiftly acting to secure Diem, barricade the palace compound, and take up defensive firing positions (Miller, 2013).

As the presidential guard successfully maintained the palace perimeter, Diem’s insiders made pleas for help to the Fifth and Seventh Infantry Divisions outside Saigon. Loyalist officers in these divisions responded positively and deployed to the rescue (Nevard, 1960a; UPI, 1960). The lightning strike advantage of the plotters faded as they were prevented from breaching the palace, leading to the surrender of rebel squads as they were surrounded by the loyal Fifth and Seventh Divisions. Routed, the rebel leaders were apprehended and jailed, concluding the coup in a failure for the plotters, securing Diem’s continued political survival (Masur, 2009; Nevard, 1960b).

Contrast the execution of this coup with the 1975 effort against President Mujib of Bangladesh, who did not possess a counterweight presidential guard unit at the time of his downfall. Mujib had been fearful of the military for some time, but rather than organizing a dedicated presidential guard unit, he instead decided to counterbalance the armed forces with a national militia: the Jatiyo Rakki Bahini (Khan, 1981; Maniruzzaman, 1976). JRB forces were armed, independent, and had stations inside the capital and across the country (Ahsan, 2014; De Bruin, 2020b). The Mujib-aligned 30,000-person militia was offered preferential pay, rapid promotions, and large institutional funding that came at the expense of the regular army’s budget (Makeig, 1988, p. 212). However, the rapid expansion of the powerful JRB still left Mujib personally vulnerable, as he never organized a presidential guard unit alongside the prioritized militia.

Mujib was essentially a sitting duck in the Dhanmondi 32 presidential residence as two Army battalions led by junior officers blitzed Dacca during the early morning of 15 August (Khondker, 1986). Outgunned and taken by surprise, two attendants outside Dhanmondi 32 were shot by attackers and the gates were breached by rebel soldiers. Mujib attempted to telephone for assistance, but by that point he was already under fire inside from the compound (Ahmed & Manik, 2018). The capture squad sent to find Mujib faced no organized resistance and almost immediately accomplished their objective (Chowdhury, 1995). Elsewhere, rebel soldiers isolated and besieged the JRB’s stations, keeping them from deploying to other strategic locations (Chin et al., 2021, p. 93). Mujib was summarily executed inside his residence by the capture squad and his weakened forces around the capital surrendered. The coup concluded in a success for the plotters and the regime of the deceased Mujib was swept away.

The outcomes of these two coup attempts were very different; Mujib was executed within hours and Diem survived to die another day. For Diem, his survival in 1960 can be to a large part attributed to the stalwart defense of his residence by the presidential guard, who quickly armed and deployed at the first signs of a rebellion to secure their regime leader. Holding off rebel troops with trained gunfire, the presidential guard of South Vietnam bought Diem enough time for loyalist divisions to arrive in the capital, ensuring his rescue and survival. Mujib was not so fortunate. He had created the JRB to counterbalance the military but had neglected to ensure he had a security force to specialize in guarding his immediate location during a coup. This left him vulnerable as rebel troops flooded into his compound and immediately removed him from the political content, striking a quick blow against his embattled forces.

Based on these two cases and the overall development of the theoretical arguments above, I expect that there should be a relationship between presidential guard units and capture of the regime leader during a coup attempt. To be effective, these presidential guard units need to be organized as counterweight forces: armed, independent of the armed forces, and centrally located near the leader. Without this specialized counterweight unit in place, capture squads sent by coup plotters storm in unchallenged during a first strike to apprehend the regime leader. This logic leads to my second hypothesis on the relationship between presidential guard units and regime leader capture during a coup attempt.

Hypothesis 2 (H2): Capturing the regime leader during a coup d’état is less likely if a counterweight presidential guard unit exists.

Data and Method

I test the two hypotheses outlined above by using an event-level dataset of all coup attempts worldwide since 1950, identified by Powell and Thyne (2011). They define coups attempts as “illegal and overt attempts by the military or other elites within the state apparatus to unseat the sitting executive” (Powell & Thyne 2011, p. 252). An event sample using this definition provides clear advantages for evaluating the hypotheses of this article. First, it omits coup plots and conspiracies that never come to fruition, only focusing on actualized attempts. Second, their definition clearly states that these events are launched with the aim of unseating the sitting executive, so this excludes events like autogolpes (or self-coups) and mutinies, where ousting the regime leader is not initially the desired outcome of plotters (Dwyer & Tansey, 2020). It also ensures that the sample of events includes attempts aimed at the removal of the incumbent executive.

I also edited the coup event sample in two additional ways for greater conceptual accuracy. A small number of coup attempts in the sample were found to be autogolpes perpetuated by the regime leader to unconstitutionally assume additional political powers. These events are conceptually at odds with the research questions, as capturing the incumbent not a primary objective for plotters in autogolpe. Second, one coup attempt in the sample (North Yemen, August 31, 1968) was incorrectly assigned to another country in the original source data. These adjustments resulted in a final data sample of 387 coup attempts.

Coup success is the dependent variable for assessing the first hypothesis. I apply Powell and Thyne’s (2011, p. 252) definition of success: “the perpetrators seize and hold power for at least seven days.” It should be noted that leader capture is not an absolute condition of what makes a coup successful in this definition. Rather, it measures the ability of the plotters to credibly take control of the state for a set temporal period. This focuses on a post hoc assessment of the event, rather than the process steps that led to its conclusion. Put short, this definition ensures that regime leader capture is not a criterion for measuring success. Success is dichotomously, with a value of 1 indicating rebel success and 0 indicating failure. Of the 387 coup attempts, 193 ended in rebel success and 194 in failure, creating an overall success rate of 49.87%.

Leader capture measures whether plotters successfully apprehended the regime leader during the coup attempt. This serves as the dependent variable testing the first hypothesis and the key independent variable for the second. Determining the capture status of leaders during coups worldwide since 1950 necessitated the collection of original data. I began by using political leader databases and coup narratives to identify which incumbent leader was being targeted for removal by the coup attempt (Chin et al., 2022; Goemans et al., 2009; Marshall & Marshall, 2019). Once the targeted regime leader was identified, I then shifted to assess what physically happened to that person during each coup attempt. Evidence was compiled on leader statuses from a range of source materials, including (but not limited to): radio reports, newspaper articles, secondary narratives, personal biographies, and scholarly manuscripts. If the source materials agree that the targeted regime leader at some point in the coup attempt was detained by rebel forces, I coded these as successful capture events. Figure 1 depicts the sum of leader statuses (captured/not captured) across the two different types of coup outcome (failure/success), demonstrating some initial descriptive evidence in support of the first hypothesis. Note in Figure 1 that regime leader capture is not synonymous with coup success (or vice versa), but that these outcomes do appear to be strongly linked.

Figure 1.

Incumbent Regime Leader Capture by Coup Outcome (Failure, Success).

Caveats for this variable’s coding outcomes must be addressed to ensure clear sample specifications. First, I do not include in the capture category any coup attempts where the regime leader was physically “surrounded” within a defended location but never detained. Although held to an isolated location, they were never actually captured by the plotters, still able to act independently and protected. Second, I do include in the capture group instances where the leader was detained, but subsequently escaped. Rebels in these situations still had control of the regime leader for some period that allowed the capture advantages to be exploited (such as showing the leader on television under detention). Thus, they still enjoyed the propaganda and command interruption benefits of successful capture, even if the leader subsequently escaped after a time. Finally, I code in the capture group leaders who were killed during the course of their detention, as they were effectively rendered unable to offer any further resistance as if they were continuously detained.

Presidential Guard is a dichotomous variable indicating whether this type of counterweight security force was present at the time of the coup attempt. Data for presidential guard units are collected from De Bruin, defined as security forces “tasked with the physical protection of the executive and other members of the regime” (2020c, p. 6). This definition is conceptually helpful because it clearly indicates that these forces were designed to act as proximate personal security for the regime leader, including during coup attempts. Also important to note are these forces being intentionally designed to counterbalance the violent monopoly of the military, serving as an coercive force for regime protection (Quinlivan, 1999). Their command structure being outside the direct oversight of the military aims to ensure their loyalty to the regime during crises like coups, where military generals may otherwise attempt to co-opt their support through their direct command authorities. It is also critical that they are physically stationed within the capital city, allowing rapid deployment to the leader’s side whenever a rebellion is first detected (De Bruin, 2020a).

As the primary dependent variables are all dichotomous, I selected to test the hypotheses with logistic regressions using robust country-clustered standard errors. For the models exploring the effects of presidential guard units on regime leader capture, I excluded some cases based on several factors that made certain events incompatible with the overall research questions pursued in this work. First, sample events were excluded if the presidential guard units were themselves involved in planning the coup attempt. Clearly, these forces cannot in earnest protect the regime leader during a coup they are perpetrating, so studying their use as a coup-proofing tool in those instances would be illogical.² Second, I excluded any coup attempts where the regime leader was abroad when the event was launched, as it would not have been possible for them to be captured (or not) at zero hour by domestic-side coup rebels while outside the country.³

In the main statistical models, I control for variables that may have important effects on the two outcomes of theoretical interest. Coup cliques benefit from the identities of their lead plotters, who possess traits and experiences that can be brought to bear during the coup’s execution. General staff officers have been found to be more effective plotters, often succeeding in seizing power without deploying coercive violence (De Bruin, 2018, 2019). They do so by drawing on insider knowledge of the regime’s weaknesses and leveraging the soft power authority granted by their high offices. Coups become well-staffed and carefully strategized, directed by key military figures who can effectively coordinate across large numbers of military units. These theoretical rank advantages should also allow coup outcome models to proxy for the quality of the plotters’ plans, signaling whether they had access to advantageous resources or quality personnel that should make their effort more likely to succeed from launch. I control for these planning and execution conditions with a dichotomous variable for whether the coup was led by a general officer, using plotter rank data from De Bruin (2019).

I include a dichotomous variable of democracy to account for democratic regimes being less likely to coup-proof (Pilster & Böhmelt, 2012), classified as regimes with a 6+value on the autocracy-democracy value on the Polity5 index (Marshall et al., 2015). Other studies have argued that less provisioned militaries exhibit more coup-generating grievances, but also that their weakened capacities could make them worse at executing regime changes (Collier & Hoeffler, 2007; Nordlinger, 1977). Furthermore, the actual size of the military can create coordination issues by that make it more difficult for coup conspirators to safely disseminate strategies across the vast ranks (Powell, 2012). Spending per soldier (military expenditures divided by military personnel) and military personnel control for these influences, with values from the National Material Capabilities dataset (Singer, 1987; Singer et al., 1972).

GDP per capita (in 1994 US$) and change in GDP account for economic fragility that can motivate some coup behaviors (Gleditsch, 2002). Recent revolution is a binary indicator capturing whether there was a revolution in the state in the previous 3 years with data from Colgan (2012). Prior coup is the count of years since a coup was observed in that state, along with associated polynomials to account for possible time dependencies (Carter & Signorino, 2010). Finally, I control for the impact of the Cold War (1946–1991), as coup plotters during that era of superpower competition often benefited from some advantageous foreign support, while those afterward faced greater international scrutiny (Thyne, 2010; Yukawa et al., 2022). Summary statistics for all variables are reported in Supplemental Table A1 of the Online Appendix, which is available through the journal website. The dependent variables (and event-restricted indicators like plotter rank) were lagged one calendar year.

Results

Table 1 presents the results of a series of six logistic regressions testing the two hypothesized relationships outlined earlier. As a reminder, leader capture and coup success (the key dependent variables for both hypotheses) are both modeled dichotomously. Models 1 to 3 focus on the coup outcome, examining relationships between the included variables and the likelihood of an attempt ending in successful regime change. Models 4 to 6 change the outcome to regime leader capture during a coup attempt, prior to the conclusion of the event. These models are used to assess the relationship between the presence of counterweight presidential guard units and leader capture, with the expected relationship being negative as suggested in the second hypothesis. Results for each model are depicted with estimated coefficients and robust standard errors clustered by country in the parentheses.

Table 1.

Determinants of Coup Success and Leader Capture.

	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6
Variables	Success	Success	Success	Capture	Capture	Capture
Leader capture	2.781***	3.213***	3.114***
	(0.326)	(0.394)	(0.514)
Presidential guard				−1.096**	−1.354***	−1.316***
				(0.483)	(0.442)	(0.461)
General-led coup		1.174***	1.646***		1.090***	1.090***
		(0.351)	(0.531)		(0.371)	(0.371)
Democracy		−0.852*	−1.224*		0.294	0.265
		(0.489)	(0.631)		(0.490)	(0.510)
Spending per soldier (ln)		−0.261**	−0.423*		−0.245	−0.241
		(0.124)	(0.253)		(0.208)	(0.210)
Military personnel (ln)		−0.170	−0.243		0.011	0.016
		(0.112)	(0.151)		(0.120)	(0.127)
GDP per capita (ln)		−0.085	0.015		−0.470**	−0.466**
		(0.249)	(0.268)		(0.236)	(0.235)
Change in GDP (%)		0.921	2.196		−0.761	−0.802
		(1.741)	(2.124)		(1.932)	(1.939)
Recent revolution		−1.187**	−2.098***		−0.193	−0.198
		(0.518)	(0.754)		(0.623)	(0.625)
Since last coup		0.228*	0.267		0.002	0.006
		(0.137)	(0.179)		(0.147)	(0.151)
Cold war		−0.577	−0.647*		0.920	0.916
		(0.361)	(0.381)		(0.592)	(0.586)
Counterweights (ln)			−0.284			−0.080
			(0.412)			(0.305)
Constant	−1.087***	1.598	2.576	0.220	4.338*	4.309*
	(0.169)	(1.746)	(2.699)	(0.165)	(2.275)	(2.277)
Pseudo R ²	.265	.367	.398	.017	.107	.107
Pseudo log likelihood	−197.147	−146.546	−88.942	−122.985	−107.49	−107.465
Observations	387	334	213	181	174	174

Note. Robust standard errors in parentheses. Polynomials of since last coup included, but not reported.

p < .1. **p < .05. ***p < .01.

Model 1 tests for the more limited relationship between leader capture and the success of a coup d’état, without additional control variables. Model 2 expands to include nearly the full battery of control variables, only omitting counterweights, due to the reduction in model observations that its inclusion will create from source sample limitations. Model 3 includes counterweights alongside all previous variables in testing for coup success. Model 4 shifts the dependent variable to regime leader capture, looking initially at the limited effect of presidential guard units on the likelihood of a regime leader being captured, without additional controls. Model 5 incorporates the battery of control variables, again minus counterweights, while Model 6 incorporates all variables to examine their full effects.

Models 1 to 3 demonstrate support for the first hypothesis, depicting that the relationships between regime leader capture and coup success are consistently positive and statistically significant. These findings are robust to inclusion of the battery of control variables, including assessing whether the coup was led by a general and if it occurred presence of other counterweight forces. Figure 2 depicts the substantive effects of varying regime leader capture outcomes on coup success with 95% confidance intervals included, drawing on the results from Model 3. These outcomes show just how substantively important capture of the regime leader can be in deciding the outcome of a coup attempt. All other factors held constant at their medians, plotters who fail to capture the regime leader will only have around a one in four chance (25.2%) of achieving overall outcome success. Contrast this with coup attempts where the rebel forces apprehend the regime leader, which achieve around 84% likelihood of success in adverse regime change.

Figure 2.

Predicted Probabilities for Coup Success Given Value Outcomes of Leader Capture. Vertical Bars Depict 95% Confidence Intervals.

The results from this initial set of models also confirm prior theories from the extant literature explaining coup outcomes. Similar to De Bruin’s (2018) findings, these models demonstrate that general-led plots enjoy a strong success advantage compared with those led by actors outside the military or by the military lower ranks. The aftereffect of a recent revolution also has a robustly negative and significant effect on coup success across Model 2 and Model 3. Recent revolutions appear to make coups less successful, suggesting that the reorganization of the state and military during these periods creates conditions where plotters are more easily repelled by the mobilized forces of the revolution or institutionally weakened through that process. This lends support to the notion that there are important institutional changes that take place in post-revolutionary societies, which help to protect them from military uprisings (Lachapelle et al., 2020; Levitsky & Way, 2022).

Finally, it should be noted that the expected relationship in Model 3 between the number of counterweights and coup success does not achieve reportable statistical significance. This can perhaps be explained through the inclusion of the leader capture variable, which suggests that the importance of these security forces is not so much directly linked to coup success, but rather they have greater significance when deciding the dynamics of coup attempts. Put short, perhaps strategic dynamics like leader capture matter more for the outcome of a coup, while security force types and numbers matter more for how those objectives are exploited during coup execution (which is partially confirmed in Models 4–6). Although not possible to confirm at this time, the reduction in observations for the model because of the more limited counterweight data sample may also have an important effect on the control outcomes. If more data were available for counterweight forces globally, there may be significant changes to the relationships observed.

Models 4 to 6 present the results from testing the effects of selected variables on the likelihood of regime leader capture during a coup attempt. In line with the expectations of second hypothesis, there is a robustly negative and statistically significant relationships between the presence of counterweight presidential guard units and the likelihood of regime leader capture. These combined findings suggest that these particular variants of counterweight security forces are particularly effective in their designed purpose of VIP protection, even from armed attackers. Their presence makes coup plotters far less likely to successfully capture the leader than they are in situations where presidential guards are absent from the regime’s institutional forces.

Figure 3 presents the predicted probabilities of regime leader capture depending on presidential guard unit presence, demonstrating the substantive effect of these forces to further illustrate the statistical significance. These predicted probabilities indicate that coups launched in the absence of presidential guards have a 55.6% likelihood of capturing the regime leader. For coups where presidential guard units have been deployed beforehand, the likelihood of capture drops to 28.1%, a notable decrease. Clearly, we can see that the negative and significant relationship between presidential guard and the incumbent leader’s status during a military coup is substantively significant for all sides in these high-stakes coordination games. These findings lend additional strong support for the second hypothesis and demonstrate the critical magnitude of this relationship.

Figure 3.

Predicted Probabilities for Leader Capture Given Presence of Counterweight Presidential Guard Units. Vertical Bars Depict 95% Confidence Intervals.

I now continue to additional analysis of the other relationships depicted in Models 4 to 6. The findings concur with previous literature on the benefits that coup plots receive when being led by general rank officers (De Bruin, 2019; Singh, 2014). It appears that they are also incredibly effective as successfully coordinating the capture of regime leaders during these events, successfully executing this strategy far more often than their junior, middle rank, or non-military peers. However, even within general-led coups, substantive effects using Model 6 indicate that presidential guard units still reduce the likelihood of capture from 68.81% in their absence to 39.27% when PGUs are present, negating a large amount of the execution advantage these coups enjoy.

The overall count of counterweight forces (including militias, militarized police units, etc.) does have a significant relationship with regime leader capture when accounting for individual types of counterweight forces present at the time of the coup. This suggests that additive coup-proofing does not appear to be the most important factor for regime survival, at least when studying individual strategic outcomes occurring prior to the coup’s conclusion. Rather it seems that having specialized security forces to accomplish specific tasks could be more important than simply stacking with larger numbers of security force institutions. Again, it must be noted that the coverage of the counterweight data leads to a moderate drop in observations. This may influence the lack of significance reported, although the reduction in observations difference is not as numerically substantive as in Models 1 to 3. As new data are released on counterweights globally and across time, this would be an important relationship to reassess across the full universe of coup attempts.

I include in the Supplemental Online Appendix additional analyses testing the robustness of results with different model specifications. Supplemental Table A2 checks the key findings with inclusion of year fixed effects, while Supplemental Table A3 incorporates country fixed effects. Supplemental Table A4 substitutes the main coup observations with those identified by the Colpus dataset (Chin et al., 2021), aiming to ensure findings are not reliant on a particular sample set. Supplemental Table A5 incorporates the presidential guard unit variable into the coup success models. Supplemental Table A6 includes a variable capturing the military’s share of cabinet seats, proxying for the military’s influence over the prior government (White, 2021) to check the effects of military control over governance.

Supplemental Table A7 excludes the Cold War variable, while Supplemental Table A8 replaces the dichotomized democracy indicator with its original continuous index form. Supplemental Table A9 accounts for whether the relationship between presidential guards and leader capture is robust when accounting for plotters selecting into a coup attempt. Plotters may not select into a coup attempt if they believe their chances of capturing the regime leader are already low, perhaps due to the extant deployment of a counterweight presidential guard unit. If they do select into a coup, there may be other endogenous factors not accounted for in the model that motivate this choice, even in the face of robust presidential guard units. Thus, I test the selection effect of opting into a coup by using a two-stage model of coup attempts and leader capture. The key results are robust across nearly all these alternate specifications, as depicted in the Supplemental Online Appendix, which again is available through the journal website.⁴

Conclusion

The findings in this article should interest scholars of civil–military relations and political violence. Recent years have demonstrated the continued challenge that coups pose to the global community, which should encourage us to conduct greater study of these destabilizing events. This article aims to shift some of the focus of this field onto the strategic dynamics that take place after the coup has started, but before the outcome has been decided. It specifically focuses on the important role that the targeted incumbent leader plays in deciding the event by asking whether their capture provides a demonstrable benefit for plotters? In addition, how can these leaders prevent their capture from taking place through coup-proofing strategies? Using new data on leader statuses during coup attempts, I empirically demonstrate that prior theoretical intuitions are valid: capturing the regime leader makes coup success more likely.

However, it is also important to consider the role of coup-proofing institutions when considering how likely leader capture is during an attempted regime change. Specifically, has a counterweight presidential guard unit been deployed to ensure the regime’s leader is given immediate protection in the event of a coup or other security situation? Drawing again on the novel leader status data, results robustly support the notion that having a counterweight presidential guard unit in place makes the capture of an incumbent leader during a coup attempt significantly less likely, greatly improving the regime’s odds of survival.

There are additional avenues of research emerging from these findings that I would like to suggest. The first could focus on the objective-specific effects of other security force types. Are militias and militarized police better at securing other strategically-critical areas during coups, such as the airport or broadcast stations? Second, does the composition of presidential guard units have an impact on their ability to effectively resist coups? Stacking literature has emphasized that the demographic composition of coercive institutions can have effects on their battlefield effectiveness and political loyalty (Allen & Brooks, 2023; Harkness, 2016, 2018). Are stacked presidential guard units more committed to defending the regime leader at all costs, because their fates are more closely linked to their patron’s? In addition, are certain strategic objectives (such as the airport and broadcast stations) more important than others in determining the outcome of the coup? This would require the collection of additional coup-level event data within attempt episodes, which could open additonal avenues of investigation for scholars.

This article confirms old assumptions and introduces new findings about coup events that should be of interest to both policymakers and scholars. Knowing the tell-tale signals of likely coup success will be important for policymakers in deciding how to best respond as these events unfold, leveraging real-time information and applying the theoretical expectations of their effects. The findings also highlight the importance of counterweight presidential guard for halting coups. This information may be vital for struggling democratic leaders in coup-prone states, for whom these defenders might be the literal line between life and death. In short, coup-proofing tools are critical for regime survival, but so is having the right tool for the right job.

Supplemental Material

sj-docx-1-afs-10.1177_0095327X231169480 – Supplemental material for If You’ll Be My Bodyguard: Presidential Guard Units and Leader Capture During Coups d’état

Supplemental material, sj-docx-1-afs-10.1177_0095327X231169480 for If You’ll Be My Bodyguard: Presidential Guard Units and Leader Capture During Coups d’état by Austin S. Matthews in Armed Forces & Society

Footnotes

Acknowledgements

The author wishes to thank the respective chairs, discussants, and participants that engaged with earlier versions of this article at the 2017 Midwest Political Science Association and 2022 American Political Science Association annual conferences. A special acknowledgment is owed to Barbara Geddes, who offered extensive feedback and encouragement. Finally, I express my gratitude to Alia Palmer for excellent research assistance on this project.

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The author acknowledges East Carolina University, the Thomas Harriot College of Arts and Sciences, the ECU Office of Research, Economic Development, and Engagement and the Department of Political Science for start-up funding that contributed to the publication of this article.

Supplemental Material

Supplemental Material for this article is available online.

Notes

Author Biography

Austin S. Matthews is an assistant professor of Political Science at East Carolina University, also associated with the Security Studies program. His research interests include dictatorships, elite-level politics, regime change, and political violence.

References

Ahmed

Manik

J. A.

(2018, August 15). Bloodbath on Road 32. The Daily Star.

Ahsan

S. B.

(2014, January 13). Myth, reality and Rakkhi Bahini. The Daily Star.

Allen

Brooks

(2023). Unpacking “stacking”: Researching political identity and regime security in armed forces. Armed Forces & Society, 49(1), 207–227.

Andreasen

(2023, March). A special program bolsters security at embassies and consulates in challenging environments. DSS special agents explain SPEAR and how it works in Mali. The Foreign Service Journal, 41–44.

Associated Press. (1960, November 11). Battle at palace: Pro-west president’s fate unknown—Coup chief U.S.-Trained. New York Times, A1.

Barany

(2011). Comparing the Arab revolts: The role of the military. Journal of Democracy, 22(4), 24–35.

Barany

(2013). Armies and revolutions. Journal of Democracy, 24(2), 62–76.

Bleck

Michelitch

(2017). Capturing the airwaves, capturing the nation? A field experiment on state-run media effects in the wake of a coup. Journal of Politics, 79(3), 873–889.

Böhmelt

Clayton

(2017). Auxiliary force structure: Paramilitary forces and progovernment militias. Comparative Political Studies, 51(2), 197–237.

10.

Brooks

(1999). Political-military relations and the stability of Arab regimes. Routledge.

11.

Brooks

White

P. B.

(2022). Oust the leader, keep the regime? Autocratic civil-military relations and coup behavior in the Tunisian and Egyptian militaries during the 2011 Arab spring. Security Studies, 31(1), 118–151.

12.

Carter

D. B.

Signorino

C. S.

(2010). Back to the future: Modeling time dependence in binary data. Political Analysis, 18, 271–292.

13.

Casper

B. A.

Tyson

S. A.

(2014). Popular protest and elite coordination in a coup d’État. Journal of Politics, 76(2), 548–564.

14.

Chin

J. J.

Wright

Carter

D. B.

(2021). The Colpus dataset: Case narratives for candidate coup events in Asia, 1946-2020. https://www.johnjchin.com/colpus

15.

Chin

J. J.

Wright

Carter

D. B.

(2022). Historical dictionary of modern coups d’état. Rowman & Littlefield.

16.

Chowdhury

M. H.

(1995). Dynamics of political parties in Bangladesh. South Asian Survey, 2(1), 63–76.

17.

Cole

(2020). Five things to know about Mali’s coup. USIP Publications. https://www.usip.org/publications/2020/08/five-things-know-about-malis-coup

18.

Colgan

(2012). Measuring revolution. Conflict Management and Peace Science, 29(4), 444–467.

19.

Collier

Hoeffler

(2007). Military spending and the risks of coups d’état (pp. 1–32) [Department of Economics Papers, University of Oxford].

20.

Crabtree

Kern

H. L.

(2018). Using electromagnetic signal propagation models for radio and television broadcasts: An introduction. Political Analysis, 26(3), 348–355.

21.

Crandall

(2006). Gunboat diplomacy: U.S. Interventions in the Dominican Republic, Grenada, and Panama. Rowman & Littlefield.

22.

De Bruin

(2018). Preventing coups d’état: How counterbalancing works. Journal of Conflict Resolution, 62(7), 1433–1458.

23.

De Bruin

. (2019). Will there be blood? Explaining violence during coups d’état. Journal of Peace Research, 56(6), 797–811.

24.

De Bruin

. (2020a). How to prevent coups d’état. Cornell University Press.

25.

De Bruin

. (2020b). Mapping coercive institutions: The state security forces dataset, 1960-2010. Journal of Peace Research, 58(2), 315–325.

26.

De Bruin

. (2020c). Online appendix and codebook: The state security forces dataset, 1960-2010. https://www.ericadebruin.com/research

27.

De Bruin

. (2022). Power sharing and coups d’état in post-conflict settings: Evidence from Burundi and Guinea-Bissau. Armed Forces & Society. https://doi.org/10.1177/0095327X221108370

28.

Diallo

(2020, August 18). Mali president resigns after detention by military, deepening crisis. Reuters.

29.

Dwyer

Tansey

(2020). Explaining divergent trends in coups and Mutinies: The end of the cold war and the role of military agency. Security Studies, 29(5), 864–893.

30.

Escribà-Folch

Böhmelt

Pilster

(2020). Authoritarian regimes and civil-military relations: Explaining counterbalancing in autocracies. Conflict Management and Peace Science, 37(5), 559–579.

31.

Feaver

P. D.

(1996). The civil-military problematique: Huntington, Janowitz, and the question of civilian control. Armed Forces & Society, 23(2), 149–178.

32.

Finer

S. E.

(1988). The man on horseback: The role of the military in politics (2nd ed.). Westview Press.

33.

Frantz

Stein

E. A.

(2017). Countering coups: Leadership succession rules in dictatorships. Comparative Political Studies, 50(7), 935–962.

34.

Gleditsch

K. S.

(2002). Expanded trade and GDP data. Journal of Conflict Resolution, 46(5), 712–724.

35.

Goemans

Gleditsch

Chiozza

(2009). Introducing Archigos: A dataset of political leaders. Journal of Peace Research, 46(2), 269–283.

36.

Greitens

S. C.

(2016). Dictators and their secret police: Coercive institutions and state violence. Cambridge University Press.

37.

Harkness

K. A.

(2016). The ethnic army and the state: Explaining coup traps and the difficulties of democratization in Africa. Journal of Conflict Resolution, 60(4), 587–616.

38.

Harkness

K. A.

(2018). When soldiers rebel: Ethnic armies and political instability in Africa. Cambridge University Press.

39.

Horowitz

D. L.

(1985). Ethnic groups in conflict. University of California Press.

40.

Janowitz

(1977). Military institutions and coercion in the developing nations. University of Chicago Press.

41.

Khan

Z. R.

(1981). Politicization of the Bangladesh military: A response to perceived shortcomings of civilian government. Asian Survey, 21(5), 551–564.

42.

Khondker

H. H.

(1986). Bangladesh: Anatomy of an unsuccessful military coup. Armed Forces & Society, 13(1), 125–143.

43.

Kim

N. K.

Sudduth

J. K.

(2021). Political institutions and coups in dictatorships. Comparative Political Studies, 54(9), 1597–1628.

44.

Krebschull

H. G.

(1994). Operation “just missed”: Lessons from failed coup attempts. Armed Forces & Society, 20(4), 565–579.

45.

Lachapelle

Levitsky

Way

L. A.

Casey

A. E.

(2020). Social revolution and authoritarian durability. World Politics, 72(4), 557–600.

46.

Lee

O. H.

(1976). Sukarno and the pseudo-coup of 1965: Ten years later. Journal of Southeast Asian Studies, 7(1), 119–135.

47.

Levitsky

Way

(2022). Revolution and dictatorship. Princeton University Press.

48.

Little

A. T.

(2017). Coordination, learning, and coups. Journal of Conflict Resolution, 61(1), 204–234.

49.

Lowenthal

A. F.

(1970). The United States and the Dominican Republic to 1965: Background to intervention. Caribbean Studies, 10(2), 30–55.

50.

Lowenthal

A. F.

(1973). The political role of the Dominican armed forces: A note on the 1963 overthrow of Juan Bosch and on the 1965 Dominican “revolution.”Journal of Interamerican Studies and World Affairs, 15(3), 355–361.

51.

Luttwak

E. N.

(1968). Coup d’État: A practical handbook. Harvard University Press.

52.

Makeig

D. C.

(1988). National security. In Heitzman

Worden

R. L.

(Eds.), Bangladesh: A country study (pp. 204–246). Federal Research Division, Library of Congress.

53.

Maniruzzaman

(1976). Bangladesh in 1975: The fall of the Mujib Regime and its aftermath. Asian Survey, 16(2), 119–129.

54.

Marshall

M. G.

Gurr

T. R.

Jaggers

(2015). Polity IV project, political regime characteristics and transitions, 1800-2014. Center for Systemic Peace.

55.

Marshall

M. G.

Marshall

D. R.

(2019). Coup d’État events, 1946-2018. Center for Systemic Peace.

56.

Masur

(2009). Exhibiting signs of resistance: South Vietnam’s struggle for legitimacy, 1954-1960. Diplomatic History, 33(2), 293–323.

57.

Matthews

(2022). Don’t turn around, Der Kommissar’s in town: Political officers and coups d’état in authoritarian regimes. Journal of Peace Research, 59(5), 663–678.

58.

Michaels

J. H.

(2014). Helpless or deliberate bystander: American policy towards South Vietnam’s military coups, 1954-1975. Small Wars & Insurgencies, 25(3), 560–583.

59.

Miller

(2013). Misalliance: Ngo Dinh Diem, the United States, and the fate of South Vietnam. Harvard University Press.

60.

Moyar

(2006). Triumph Forsaken: The Vietnam War, 1954-1965. Cambridge University Press.

61.

Nevard

(1960a, November 12). Attack by Rebels in South Vietnam appears crushed. New York Times, p. 1.

62.

Nevard

(1960b, November 13). Top Rebels jailed in South Vietnam. New York Times, p. 1.

63.

Nordlinger

E. A.

(1977). Soldiers in politics: Military coups and governments. Prentice Hall.

64.

Onuah

(2022, December 5). West African leaders plan peacekeeping force to counter “coup belt” reputation. Reuters.

65.

Palmer

Jr. (1989). Intervention in the Caribbean: The Dominican crisis of 1965. University Press of Kentucky.

66.

Perlmutter

(1977). The military and politics in modern times. Yale University Press.

67.

Pilster

Böhmelt

(2012). Do democracies engage less in coup-proofing? On the relationship between regime type and civil-military relations. Foreign Policy Analysis, 8(4), 355–372.

68.

Powell

J. M.

(2012). Determinants of the attempting and outcome of coups d’état. Journal of Conflict Resolution, 56(6), 1017–1040.

69.

Powell

J. M.

Thyne

C. L.

(2011). Global instances of coups from 1950-present. Journal of Peace Research, 48(2), 249–259.

70.

Quinlivan

J. T.

(1999). Coup-proofing: Its practice and consequences in the Middle East. International Security, 24(2), 131–165.

71.

Reuters Staff. (2020, August 20). Factbox: Who’s who in Mali’s military junta? Reuters.

72.

Roessler

(2011). The enemy within: Personal rule, coups, and civil war in Africa. World Politics, 63(2), 300–346.

73.

Singer

D. J.

(1987). Reconstructing the correlates of war dataset on material capabilities of states, 1816-1985. International Interactions, 14(2), 115–132.

74.

Singer

J. D.

Bremer

Stuckey

(1972). Capability distribution, uncertainty, and major power war, 1920-1965. In Russett

B. M.

(Ed.), Peace, war, and numbers. SAGE.

75.

Singh

(2014). Seizing power: The strategic logic of military coups. Johns Hopkins University Press.

76.

Song

(2022). Dictators, personalized security forces, and coups. International Interactions, 48(2), 204–232.

77.

Talmadge

(2015). The dictator’s army: Battlefield effectiveness in authoritarian regimes. Cornell University Press.

78.

Thyne

C. L.

(2010). Supporter of stability or agent of agitation? The effect of United States foreign policy on coups in Latin America, 1960-1999. Journal of Peace Research, 47(4), 449–461.

79.

UPI. (1960, November 12). Diem troops rout Saigon rebels. Los Angeles Times, p. 1.

80.

White

(2021). Generals in the cabinet: Military participants in government and international conflict initiation. International Studies Quarterly, 65(2), 551–561.

81.

Yukawa

Kushima

Fujita

(2022). Coup d’état and a democratic signal: The connection between protests and coups after the cold war. Journal of Peace Research, 69(6), 828–843.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.05 MB