Bringing the military back in: The politicisation of the military and its effect on democratic consolidation

Abstract

Empirical research has shown how the role of institutions, civic culture, ethnic fractionalisation, the political economy, and demographic factors affect the stability of democracies, but none has explored the role of the military as an institution that can drastically influence the process of democratic consolidation. This led many scholars to notice how the study of comparative democratisation has been state-centric, but it has largely neglected the role of the military in determining post-transitional outcomes. Thus, this study takes on this challenge by examining the role of the military in the democratisation process more than 30 years after the third wave. The results suggest that highly politicised military structures among transitional states have a deleterious effect on the quality of democracy. More specifically, if the military has institutionalised its role in politics as being interventionist, the likelihood of democratic consolidation becomes difficult. The article highlights the importance of restricting reserved domains granted to the military in the transition process in order to strengthen prospects for democratic stability.

Keywords

Civilian–military relations coups democratic consolidation democratisation military in politics military intervention reserved domains

Introduction

The literature on how transitional states can consolidate their democracies is quite a fissiparous enterprise that lacks conceptual clarity. In fact, Guillermo O’Donnell (1996) has labelled the exercise of defining the process of consolidation an exercise in futility based on illusionary tendencies. However, this is disputed by Leonardo Morlino (1998, 2003), who argues that there are clearly defined attributes that typify consolidation: mainly, if state institutions engage in consensus-seeking and legitimation, emphasising that civil society groups are instrumental in the process of building durable democratic values. Other scholars have alluded to how the military influences prospects for democratic consolidation. As an example, Andreas Schedler (1998) posits that many ex-authoritarian polities that had military regimes are in danger of democratic collapse because of a threshold problem. This occurs when civilian governments are continually observed and monitored by the military. If civilian governments have a performance-legitimacy problem, the military will develop a tendency to reassert itself either through the threat of intervention or through an outright coup. Such tendencies generate political uncertainty and may erode the democratic gains a transitional state achieves. In other cases, the military may intervene and reverse the course of democratic reforms by supporting civilian leaders’ assumption of power in autogolpes (as happened in Peru under Fujimori) or by completely rigging election results and openly supporting civilian leaders that the military has an ideological connection with (as in the cases of Paraguay and Madagascar). In certain cases, endless coup plots against the state are foiled by loyalist segments of the military, which makes civilian governments chronically unstable, compelling the state to temporarily or permanently curtail basic freedoms and rights of citizens to neutralise the factional military threat. In other countries, the historical institutionalisation of the military as a vanguard institution of reform and change gives it the justification to impose a revolving-door policy of intermittent intervention in politics (Diamond, 1999).

The military’s extrication from political involvement is a necessary component of what is referred to as the “liberal tradition” of civil–military relations. However, in order for this to be a reality, there must be a tacit agreement among the state’s political elites that the military is an organ of the state that has continued legitimacy and relevance. Further, the military’s social welfare needs must be responded to in a timely manner in order to prevent it from assuming a politicised role in the post-transitional process (Nye, 1996). William Thompson (1973, 1980) formulated a corporate grievance model, which suggested that military institutions develop self-interested motives in ending civilian rule. When national governments do not sufficiently fund the military or when its funding is threatened when the state tries to adhere to austerity measures imposed by domestic, external, or populist forces, the military may end democratic rule. We should therefore expect that when the financial interests of the military are threatened, the survivability of democracy is at stake. It is expected that in transitional states that expend a larger share of their gross domestic product (GDP) on the military, the military will assume a more restrained role in politics and thus provide the institutional space that is needed to consolidate democracy.

Research in comparative democratisation has demonstrated the causal relationship between the role of institutions, civic culture, ethnic heterogeneity, political economy, and the type of transition, in the stability of democracies, but very few have explored the role of the military as an institution that can drastically affect democratic outcomes (Bunce, 2000). As Aguero (1995, 1997) asserts, the key to any consolidated democracy occurs when the military is subordinate to the authority of elected civilian officials. Indeed, the military is considered to be one of the most important agents that can determine the outcome of democratic politics, but there is a systematic lack of empirical work to assess the military’s effect on democracy’s consolidation. This was addressed early on by Alfred Stepan, who believes that democratisation research has been state-centric, but has led to a complete “neglect of the military as a central topic of empirical research” (Stepan, 1988: 9). This led Zoltan Barany to comment that “The omission is striking since the military is the most consequential actor in post-authoritarian transitions and the success or failure of these processes to a large extent hinges on its political behavior” (Barany, 1997: 1). Although these exhortations were made several years ago, there is no empirical study that addresses this issue. Therefore, this study takes on this challenge by examining the role of the military in the consolidation process more than 30 years after the third wave of democratisation. Since the military’s role in affecting democratic outcomes has been based entirely on anecdotal case studies and macro-historical assessments (Aguero, 1998; Bruneau and Matei, 2013; Bruneau and Trinkunas, 2006; Croissant and Kuehn, 2009; Diamond and Plattner, 1996; Karl, 1990; Kuehn and Völkel, 2011; Mares, 1998; Pion-Berlin, 2001, 2011; Rice, 1992; Serra, 2010), and considering how the military is deemed to be a primary institutional factor that determined the potential for democratic consolidation in Turkey (Heper and Guney, 2000), Bangladesh and Pakistan (Wilkinson, 2000), El Salvador (Miller, 2006), Taiwan (Keuhn, 2008), and the Philippines and Indonesia (Lee, 2009), the purpose of the article is to ask three fundamental questions and provide empirical support. The major impetus of the article is to assess if such historical cases can be further generalised into an empirically verifiable theory that has resonance and significance since the third wave. The three pertinent research questions sought in this study are:

Does a politicised military harm a nation-state’s prospects in consolidating its democracy and the promotion of democratic accountability?

Does the occurrence of coups, coup attempts, and failed coups prior to democratic transition weaken and reduce the likelihood of democratic consolidation?

Do the size of the military and the amount of money it receives from the national government affect democratic outcomes?

The article proceeds as follows. First, I assess past literature on how the military can affect the prospects of democratic consolidation. Then, I will discuss why institutional legacies regarding military intervention matter in influencing democratic outcomes. The third section will provide a discussion of the data, the operationalisation of variables, and the methodology used to test the hypotheses. I then present the empirical findings. The last section concludes and offers insights based on the empirical findings.

The politicisation of the military and democratic politics

There are many studies that assert that the military must restrain itself in the political arena in order for democracy to survive (Crouch, 1997; Hunter, 1994; Huntington, 1996; Linz and Stepan, 1996; Stepan, 1988). In fact, according to Aguero (1997, 1998), democratic states are still susceptible to collapse if the defence ministry is still headed by a military general and not a civilian leader post-transition. Democratic reversals in Latin America were facilitated by the failure of civilian governments to completely extricate the military from the executive apparatus. It is believed that civilian-led ministries of defence can impose checks and oversight on the military’s activities, thus lessening the possibility of factionalism and divisions within the military that can precipitate periodic coup plots and attempts, which can weaken democratic gains (Bruneau and Goetze, 2006). Furthermore, democratic consolidation is also hinged on whether or not transitional states can unilaterally impose civilian control on the military infrastructure (Cotino-Hueso, 2007; Hunter, 1992, 1997).

Barany (1997) developed a theory on how the military infrastructure in Eastern Europe affected the quality of its democratic transitions. On the one hand, states whose military infrastructure was professionalised pre-transition (i.e. no military official occupied a seat in the politburo/government) were more likely to consolidate their democracy post-transition. This is what occurred in Poland and Hungary. On the other hand, polities where the military is politicised pre-transition (i.e. the military assumed powerful positions in government and its leadership structure was constantly reshuffled) were more likely to have turbulent transitions. This is what occurred in Romania.

Unlike its counterparts in Eastern Europe, Latin America went through prolonged periods of military dictatorships, thus institutionalising the role of the military as an active and regular actor in politics. Stepan (1988) posits that consolidation cannot occur in the Latin American region if the military fails to extricate itself from wholly civilian affairs. In fact, in post-transitional politics, the military may insist on certain “prerogatives” or “reserved domains” (Linz and Stepan, 1996; Valenzuela, 1992). These prerogatives may include: limits on the presidency in controlling the affairs of the military; a constitutional directive that preserves the autonomy of the military as being in charge of domestic security; military participation in the executive’s cabinet and the ownership of public enterprises; laws prohibiting the military from being prosecuted in civilian courts (which includes immunities as regards to testifying in legislative oversight hearings); the subordination of the police infrastructure to the military as an institution; giving the military utmost authority in determining decisions regarding the promotion and termination of officers; the continued existence of military courts as distinct from civilian courts; and blanket amnesty given to ex-military officials who committed human rights violations in the past (Stepan, 1988: 94–97). However, some countries in the Latin America region have cracked down or completely eliminated these prerogatives as they relate to immunity, which facilitated rapid democratic consolidation. Hagopian (2005: 325) cites the example of Argentina, which actively prosecuted ex-military generals under the Alfonsin government during the famous junta trials. Bolivia also followed suit when its civilian government actively pursued the prosecution of the repressive military-dictator Luis Garcia-Meza (Mayorga, 1997).

African transitions to democracy exhibited a divergent status compared with Latin America because reserved domains are still largely in effect. Since most of the military regimes and civilian-led military regimes in Africa inculcated neo-patrimonialism, the military became a key player in managed “pacted” or negotiated transitions. As such, ex-military generals like Babangida of Nigeria, Rawlings of Ghana, and Museveni of Uganda imposed conditions on how the future democratic institutions of their respective countries should work. As such, these leaders left their dictatorial imprints on the new democratic order by specifying special rules about the military’s prerogative in the formation of civil society groups, regulating the role and behaviour of political parties, and setting timetables for the holding of founding elections (Bratton and Van de Walle, 1997). Further, Africa’s stalled democratic progress is partly a result of the military’s reserved domains, which remain untouched, and the military’s “near monopoly of coercion affording military rulers significantly enhanced control over transition dynamics and outcomes” (Bratton and Van de Walle, 1997: 171).

Institutional legacies and the survivability of democracies

Institutional legacies pertaining to the military do matter. If a country has had a very active military in terms of overthrowing civilian rule pre-transition or post-transition, then the prospects for strengthening democratic institutions are dim. This is in light of the theoretical argument that if a coup happened in the past to overthrow a civilian regime, it may be difficult for the military to return to the barracks and remain uninvolved in the political process post-transition (Londregan and Poole, 1990; O’Kane, 1983). Thus, a coup event in a country’s pre-transitional past may induce the military to position itself as a politicised organisation, which can obstruct democratic development post-transition (Nye, 1996). Accordingly, when the military has institutionalised its role in politics as being interventionist, that is, it has engaged in coups, counter-coups, and coup plots in the recent past, the likelihood for democratic consolidation becomes difficult (Diamond and Plattner, 1996). Further, it is also expected that militaries in transitional states that experienced a coup in their pre-transitional histories are those that are more likely to demand “reserved domains” during the post-transitional process and to ask for a major role in drafting a new civilian constitution that is favourable to the military’s self-interests (Huntington, 1996). This path dependency is clearly a manifestation of how militaries that have been activist in the past are hesitant to let go of their political influence in the post-transitional process and would like the new constitutional order to legitimise their right to influence and dictate political outcomes with the goal of retaining their prerogative in ending civilian rule if need be. Thus, many transitional states that experienced coups in their pre-transition histories had constitutional provisions that laid the groundwork for military intervention if civilian leaders should lose the consent of the governed (Siaroff, 2009).

Based on the aforementioned theoretical discussion, the following hypotheses are formulated:

H1: Transitional states that have a politicised military infrastructure are less likely to have accountable governments.

H2: Transitional states with a politicised military (in terms of the presence of reserved domains) are less likely to consolidate their democracy.

H3: Transitional states that have an interventionist military in their pre-transitional past are less likely to consolidate their democracy.

H4: Transitional states that expend a larger share of their GDP on the military are more likely to have better democratic quality and consolidate their democracy.

Measuring the politicisation of the military

Measuring the politicisation of the military is a difficult empirical exercise. There are two existing measures that seek to capture Linz and Stepan’s (1996) assessment of how the military may ask for reserved domains post-transition. One measure is provided by Alan Siaroff (2009), who operationalised the degree of military influence in politics based on Samuel Fitch’s (1998) categorisation.¹ Siaroff operationalises his measure by examining the extent to which the military is subordinated to civilian-democratic control. As such, transitional democracies can be under the complete dominion of the military or the military may be completely subject to the authority of civilian governments. The Siaroff index of civil–military relations examines the presence or absence of reserved domains that the military has for each transitional state. The list of reserved domains are listed in Table 1, which includes the extent to which the military exerts control over security policy, whether the military has a constitutional prerogative to intervene in times of a national crisis, and whether or not the military controls its own internal processes, including decisions on promotion and assignment/reassignment of positions. Transitional states that have the highest value for military influence are known to be in the category of “military rule”, where state policy is controlled by military generals and where cabinet members of the executive are recruited from the military. States that have the least degree of military influence are known to be in the category of “civilian supremacy”, where the civilian government has effective control over security policy, the military is held accountable for any human rights violations in the past and are prosecuted post-transition, and civilian governments largely influence the internal processes of military-related spending, including crafting the budget on defence-related appropriations and influencing the promotions of generals. Thus, a transitional state can receive a civilian–military relations score of 0 if it is under complete military rule or 10 if civilian supremacy over the military is apparent. This measure provides scholars of comparative democratisation with a static measure of the military’s role in transitional politics because the role of the military does not drastically shift, change, or go at the extreme ends of the continuum scale over time (Siaroff, 2009).

Table 1.

List of reserved domains that the military can impose on the state (post-transition).

The military retains control over security policy.

The military has control over policy areas like defence budgets.

The military has a constitutional and historical prerogative to end civilian rule if it loses political legitimacy.

The head of government is a member of the military.

The defence minister is from the military.

Some members of the cabinet are from the military.

The military has the ability to appoint/override civilian positions in government.

The constitution gives the military policy responsibilities in governance and a consultative role.

The military has dominion over national resources, controls a specific industry, and has its own resources.

The military is not accountable for human rights abuses in the past.

The military controls all of its internal processes like budget procurement and promotions.

Source: Derived from Siaroff (2009: 90).

One weakness of the Siaroff measure is that it provides a numerical measure of the military’s role in domestic politics immediately after each country’s transition to democracy. Although it is true that the institutional powers the military may practise are fairly static over time because it is mostly a function of historical institutionalisation, it is very important to assess how the drafting of new democratic constitutions post-transition may affect and drastically alter the military’s dominion over domestic politics. For instance, Argentina’s post-authoritarian constitution restricts by law the intervention of the military in resolving internal civilian conflict (Hagopian, 2005). Thus, to test the long-term impact of the military and its effect on democratic politics, a more valid and reliable measure is the Military in Politics measure provided by the Political Risk Services Group (PRS hereafter).² The military involvement measure approximates the degree to which the military is involved in civilian affairs because of a perceived internal/external threat and its perceived capacity to overthrow civilian governments. The PRS compiles the measure of military involvement in politics in time-series format for transitional states from 1984, with the most recent data released in 2008, such that states with lower levels of military activism receive a score of 1, while states with a very politicised military are given a score of 6.³

The third measure looks at the historical politicisation of the military. This variable is coded as a dummy variable wherein countries that experienced a coup event (attempted, failed, or successful) in their pre-transitional past receive a score of “1”; all others receive a “0”. The measure is provided by the Integrated Network for Societal Conflict Research. This measure examines the historical involvement of the military in overthrowing civilian governments since a country gained independence or full sovereignty. It is appropriate to utilise this measure because scholars have long argued that military interventionism in overthrowing civilian rule is a path-dependent phenomenon (Londregan and Poole, 1990; O’Kane, 1983).

Data, measures, and methods

Cases

The empirical analysis utilises cases from those countries that are labelled as transitional states. Such states are those that: (1) transitioned from a one-party system to a multiparty system; (2) transitioned from authoritarian rule to democratic rule; and (3) transitioned from a civil war to a new democratic state (Karatnycky and Ackerman, 2005). The 44 cases utilised in this article are listed in Appendix 1 together with each country’s transition year, its scores on the Siaroff index of military involvement, the score of the politicisation of the military index provided by the PRS group (mean score, 1984–2008), and whether or not each state experienced a coup event in its pre-transitional history (coded as a dummy: 0 = absence of an event; 1 = presence of an event). The 44 states were selected because they have data points for all measures that tap into the military’s role in politics (the PRS measure, the Siaroff Index, and the existence of a coup in a state’s pre-transition history). To ensure that the results are not driven by potential outliers— more specifically, transitional states that did not go through a democratic transition— the regression modelling also includes a reduced sample of 39 states. The reduced sample excludes the following countries, which are categorised by the Freedom House Organisation as “not free” in 2008: Azerbaijan, Belarus, Kazakhstan, Russia, and Zimbabwe.

Dependent variable

The purpose of this article is to examine if having a politicised military in transitional states has an enduring effect on democratic outcomes; more specifically, if having a politicised military has a negative effect on prospects for democratic consolidation. It can be argued that democratic consolidation, defined as the likelihood that a state will not revert back to authoritarianism, is contingent on whether or not there is political accountability in terms of the level of checks and balances within a given political system, and whether or not the state promotes democratic values that discourage political marginalisation and encourage political openness (Shin, 1994).

Thus, the dependent variables utilised in this study measure the extent to which the state has strengthened its trajectory toward democratic consolidation. To do this, I obtain data from three organisations that look at: (1) the degree to which a state is accountable to mass publics; (2) the degree to which the state promotes the politics of inclusion, equality, pluralism, and fairness in governance; and (3) the degree to which a transitional state is likely to become an authoritarian state. Thus, all of these variables are employed in this study because of their ability to gauge whether or not a democratic system is more likely to collapse or endure.

For the measure of democratic accountability, I utilise the PRS Group’s measure. The data on political accountability examine the nature of democratic politics in each transitional state from 1984 to 2008. This allows the analysis to examine the impact of the military variables on variations in levels of democratic accountability for a 24-year period. According to the codebook, this measure assesses how responsive the state is to the demands and voices of the people by assessing its governance style, whether or not the country protects political and civil liberties, and whether it has veto players in the government. States with high levels of political accountability receive a score of 6, while states with democratic deficits in terms of a lack of responsiveness and accountability receive a score of 0.⁴

The second measure to proxy the extent of democratic consolidation is the Economist Intelligence Unit’s (EIU’s) Score of Democracy. This is not available on a yearly basis; therefore, I use the scores released in various years and pool them in the regression analyses. The Economist, a reputable organisation that provides an assessment of substantive democratisation, creates a democracy score for each state based on 60 indicators grouped in five different categories: electoral process and pluralism; civil liberties; functioning of government; political participation; and political culture. Each country receives a score ranging from 1 to 10, where countries achieving scores closer to 10 are considered full democracies that are less likely to revert back to authoritarian rule, while countries closer to 1 are considered to be more susceptible to democratic collapse.

The third measure I use is the Bertelsmann Transformation Index (BTI), a widely used index of democratisation,⁵ which is also not released on a yearly basis. Therefore, I use the scores released in various years and pool them in the regression analysis. The indexed score ranges from 1 (low) to 10 (high), where higher scores (those approaching 9 and beyond) indicate consolidated democracies that are highly unlikely to revert back to authoritarian rule (BTI, various years).

Control variables

Several control variables are utilised in the study that may have a long-standing effect on the process of democratisation. Data sources for the control variables are listed in Appendix 2. They include institutional variables that may have beneficial or deleterious effects on democratic quality, such as: the role of parliamentarianism in democratic accountability (Gerring et al., 2009);⁶ the deleterious or mixed effects of federalism on democratic quality (Lane and Ersson, 2005);⁷ and how a socialist past may prove to be a hindrance to democratic development because of the absence of civic associationalism (Bernhard and Karakoc, 2007).⁸ The role of socio-economic development, as captured by the natural log of gross national product (GNP), is also controlled for with a lag of the previous year’s data.⁹ The absence of a homogeneous society is also known to affect the prospects of consolidation and this is accounted for by using ethnic fractionalisation as a control (Horowitz, 1993).¹⁰ States that have large Muslim populations are also deemed by scholars as having a democratic deficit attributed to the prevalence of gender inequality (Fish, 2002).¹¹ Lastly, I also control for two related variables that tap into the role of the military in the democratisation process, such as the military participation ratio, which is the size of the army per 1000 population, and military expenditures as a percentage of GDP.¹² Theoretically, we should expect that a nation-state with a large military apparatus will see its military branch become more politicised over time (Nye, 1996; Siaroff, 2009).

Method and hypotheses

I utilise panel data ordinary least squares (OLS) regression analysis to assess if there is a relationship between the politicised nature of the military and the consolidation of transitional democracies.¹³ However, because I utilise time-invariant data in the modelling, we are presented with problems concerning serial auto-correlation and heteroskedasticity. A Philipps-Perron test of panel unit root confirms groupwise heteroskedasticity but not in all panels. To mitigate this deficiency, I use a model that uses panel-corrected standard errors, as advised by Beck and Katz (1995, 1996). The models are also run with the dependent variable lagged at t – 1 on the right-hand side of the panel equation to further offset the potential problem of bidirectionality (how the present score of democracy is correlated with past scores, and vice versa). Variance Inflation Factor (VIF) scores are all within standard levels below the numerical value of 2 (Fox, 1997). Summary statistics are provided in Appendix 2.

The data is panelled according to country-year format and has the equation summarised below:

y_{i t} = β_{0} + X_{i t} β + Ζ_{i} γ + α_{i} + μ_{i t},

where: y_it is the dependent variable observed for country i at time t; X_it is the time-variant regressor; Z_i is the time-invariant regressor; αi is the unobserved country effect; and u_it is the error term. The term αi represents historical and institutional factors (country-level data).

Results

Panel regression results presented in Table 2 confirm the hypothesis that a politicised military infrastructure affects democratic politics. For instance, the military index of politicisation (provided by PRS) is negatively correlated with democratic accountability at (p < .05). Thus, states that have politically active military institutions are more likely to face lower levels of political accountability. We can also infer from Table 2 that the Siaroff index of military activism is positively correlated with higher levels of democratic accountability (p < .01), suggesting that states that have restrictions on “reserved domains” are more likely to be responsive to political rights and civil liberties, and have checks and balances in their political system. The role of institutionalised military intervention, reflected by the occurrence of coup events in the historical past (prior to transition), also has a negative effect on democratic accountability (achieving statistical significance at p < .10), largely confirming how the historical legacy of institutional structures may affect the nature of transitional politics (Pop-Eleches, 2007). Interestingly, none of the military-related variables concerning the size of the national army and spending on the military as an institution are correlated with democratic accountability.

Table 2.

The effect of a politicised military on democratic accountability; pooled cross-national regression results with PCSE (1984–2008).

	Model A	Model B	Model C	Reduced Sample	Reduced Sample	Reduced Sample
Military index of politicisation (PRS)	−0.457** (0.189)	—	—	−0.259** (0.122)	—	—
Siaroff index of military’s activism	—	0.387*** (0.133)	—	—	0.247** (0.109)	—
Coup dummy prior to transition	—	—	−0.096* (0.050)	—	—	−0.224* (0.131)
Ethnic fractionalisation	0.039 (0.896)	−0.271 (0.904)	−0.042 (0.896)	0.089 (0.903)	−0.159 (0.897)	0.052 (1.012)
GNP per capita_(t–1)	0.240 (0.209)	0.368* (0.184)	0.506** (0.221)	0.343 (0.243)	0.349* (0.204)	0.483** (0.215)
Parliamentarianism	0.631*** (0.208)	0.689** (0.208)	0.547** (0.238)	0.386* (0.203)	0.485* (0.256)	0.436* (0.224)
Federalism	0.195 (0.495)	−0.285 (0.608)	−0.302 (0.642)	0.051 (0.554)	−0.149 (0.556)	−0.149 (0.534)
Socialist past	−1.120** (0.510)	−1.553** (0.608)	−0.934* (0.642)	−0.413 (0.436)	−0.759 (0.519)	−0.383 (0.484)
% Muslim	−0.010** (.0000)	−0.010** (0.000)	−0.011** (0.000)	−0.003 (0.005)	−0.005 (0.004)	−0.004 (0.005)
Military participation ratio	−0.118 (0.089)	−0.128 (0.093)	−0.149 (0.110)	−0.073 (0.097)	−0.066 (0.095)	−0.070 (0.099)
Military expenditures as a % of GDP_(t–1)	0.033 (0.196)	0.033 (0.174)	0.102 (0.205)	−0.011 (0.180)	−0.055 (0.166)	−0.094 (0.167)
Constant	3.766 (1.664)	−1.092 (1.863)	1.239 (1.625)	2.798* (1.632)	0.254 (1.621)	−0.094 (0.167)
Number of countries	44	44	44	39	39	39
Number of observations	1047	1047	1047	927	927	927
R Square	0.450	0.450	0.366	0.393	0.434	0.382

Notes: Unstandardised coefficients are reported. Panel-corrected standard errors in parentheses. Based on a two-tailed test. Significance: *** p < .01; ** p < .05; * p < .10.

We now move to examine the effect of a politicised military on a more substantive and maximalist measure of democracy, the EIU measure. Table 3 conveys how all the relevant measures—the PRS measure (significant at p < .05), the Siaroff index (significant at p < .05), and the coup index prior to transition (significant at p < .01)—have coefficient signs in the expected direction and are statistically significant across the board. Out of all the three variables capturing the politicisation of the military, the variable that has the most salient effect on promoting the politics of inclusion, equality, pluralism, and fairness in governance is the dummy variable of whether or not a state had a coup in its pre-transition history. Hence, states whose militaries were engaged in coups prior to transition are less likely to strengthen the nature of democratic politics, an indication that activist militaries in the pre-transition period are more likely to remain restive and destabilising in the post-transition period, largely affecting and eroding democratic gains. Hence, this comports with Stepan’s (1988) argument that historical legacies of military adventurism matter in shaping how far states move forward in their transitional trajectory toward consolidation.

Table 3.

The effect of a politicised military on the level of democracy (EIU measure pooled for 2003, 2006, 2008, and 2010 ratings).

	Model A	Model B	Model C	Reduced sample	Reduced sample	Reduced sample
Military index of politicisation (PRS)	−0.410** (0.153)	—	—	−0.180** (0.094)	—	—
Siaroff index of military’s activism	—	0.293** (0.124)	—	—	0.098** (0.0425)	—
Coup dummy prior to transition	—	—	−0.151*** (0.049)	—	—	−0.282** (0.127)
Ethnic Fractionalisation	0.284 (0.770)	−0.480 (0.795)	−0.675 (0.770)	−0.936* (0.529)	−1.059* (0.613)	−0.936* (0.466)
GNP per capita_(t–1)	0.503** (0.188)	0.643*** (0.167)	1.870*** (0.666)	0.580** (0.232)	0.627*** (0.223)	0.670*** 0.166
Parliamentarianism	0.626*** (0.179)	0.663*** (0.219)	1.312 (0.862)	0.376** (0.180)	0.404* (0.203)	0.4581** (0.183)
Federalism	−0.013 (0.463)	−0.399 (0.432)	0.553 (2.056)	−0.145 (0.543)	−0.244 (0.537)	−0.363 (0.442)
Socialist past	−1.378** (0.543)	−1.641** (0.628)	−8.856*** (2.904)	−0.598* (0.341)	−0.642 (0.384)	−0.735** (0.341)
% Muslim	−0.005** (0.000)	−0.007** (0.000)	−0.008** (0.000)	−0.006 (0.005)	−0.007 (0.006)	−0.006 (0.005)
Military participation ratio	−0.097 (0.110)	−0.112 (0.113)	−0.855** (0.415)	−0.081 (0.079)	−0.084 (0.078)	−0.067 (0.077)
Military expenditures as a % of GDP_(t–1)	−0.104 (0.190)	−0.153 (0.191)	−0.081 (0.685)	−0.084 (0.117)	−0.117 (0.110)	−0.159 (0.106)
Constant	4.293 (1.445)	−0.138 (1.162)	−7.569 (5.864)	3.488** (1.660)	1.972 (1.204)	2.539** (1.058)
Number of countries	44	44	44	39	39	39
Number of observations	176	176	176	156	156	156
R Square	0.562	0.560	0.553	0.714	0.665	0.595

Notes: Unstandardised coefficients are reported. Panel-corrected standard errors are in parentheses. Based on a two-tailed test. Significance: *** p < .01; ** p < .05; * p < .10.

Finally, what effect does an activist military have on the prospects for democratic consolidation? The results presented in Table 4 show that the PRS measure is negatively correlated with democratic consolidation (significant at p < .01) and, based on the predicted estimates shown in Figure 1, we see that the BTI measure of consolidation is at its highest when the PRS measure is 1 (states with apolitical military institutions), while the BTI score is at its lowest when the PRS measure is 6 (states with politicised military institutions), even when holding all control variables constant at their means and modes. These results are mimicked when we look at the effect of the Siaroff index of military activism on democratic consolidation. Table 4 shows a positive relationship between the Siaroff index and democratic consolidation (significant at p < .01), and the predicted estimates illustrated in Figure 2 show that as the restrictions on reserved domains increase, so do the prospects for democratic consolidation. To substantiate further, the BTI transformation index reaches its maximum value when the Siaroff index is 10 (connoting lower levels of reserved domains), while the BTI transformation index reaches its minimum level when the Siaroff index is 2 (connoting higher levels of reserved domains). This shows that states which have militaries that control security, defence, and social policies, particularly those whose post-transitional constitutions gave the military responsibilities and positions in government, are less likely to consolidate or strengthen their democratic institutions. What about the effect of the military’s historical prerogative in ending civilian rule in the past? Does this activism persist even in the post-transition period, producing problems as it relates to democratic consolidation? The results in Table 4 confirm the deleterious effect of prior coup events on the prospects of democratic consolidation, although this time the prior coup variable attains only marginal significance (at p < .10), results which attained a higher level of statistical significance when the EIU measure of democracy was used.

Table 4.

The effect of a politicised military on democratic consolidation (BTI measure pooled for 2003, 2006, 2008, and 2010 ratings).

	Model A	Model B	Model C	Reduced sample	Reduced sample	Reduced sample
Military index of politicisation (PRS)	−0.448*** (0.155)	—	—	−0.243* (0.134)	—	—
Siaroff index of military’s activism	—	0.366*** (0.119)	—	—	0.199** (0.090)	—
Coup dummy prior to transition	—	—	−0.107* (0.054)	—	—	−0.224* (0.127)
Ethnic fractionalisation	0.264 (0.596)	−1.062 (0.679)	−1.378 (0.780)	−0.451 (0.445)	−0.663 (0.547)	−0.483 (0.519)
GNP per capita_(t–1)	0.597*** (0.171)	0.728*** (0.192)	0.850*** (0.159)	0.812*** (0.175)	0.837*** (0.204)	0.942*** (0.177)
Parliamentarianism	0.585*** (0.176)	0.638*** (0.199)	0.500** (0.198)	0.232 (0.181)	0.306 (0.189)	0.284 (0.189)
Federalism	0.165 (0.455)	−0.294 (0.418)	−0.367 (0.506)	−0.192 (0.455)	−0.362 (0.439)	−0.389 (0.439)
Socialist past	−0.624 (0.592)	−1.018 (0.619)	−0.479 (0.571)	0.036 (0.557)	−0.199 (0.569)	0.047 (0.534)
% Muslim	−0.001** (0.000)	−0.003** (0.000)	−0.004** (0.000)	0.000 (0.004)	−0.000 (0.004)	0.000 (0.004)
Military participation ratio	−0.081 (0.112)	−0.093 (0.106)	−0.110 (0.126)	−0.060 (0.081)	−0.057 (0.073)	−0.056 (0.073)
Military expenditures as a % of GDP_(t–1)	0.077 (0.195)	0.028 (0.182)	−0.062 (0.224)	0.075 (0.142)	0.033 (0.139)	−0.003 (0.132)
Constant	3.688 (1.319)	−1.021 (1.472)	1.325 (0.945)	2.101* (1.140)	−0.172 (1.113)	0.696 (1.024)
Number of countries	44	44	44	39	39	39
Number of observations	176	176	176	156	156	156
R Square	0.576	0.603	0.513	0.707	0.717	0.704

Notes: Unstandardised coefficients are reported. Panel-corrected standard errors are in parentheses. Based on a two-tailed test. Significance: *** p < .01; ** p < .05; * p < .10.

Figure 1.

Predicted estimate of military politicisation on the BTI index.

Figure 2.

Predicted estimate of the Siaroff index on the BTI index.

It should also be noted here that the reduced models that drop the number of countries from 44 to 39 states also yield results that corroborate the main models in all the regression analyses. Thus, even if we restrict the sampling to states that strictly transitioned to democracy, the pernicious effect of a politicised military on democratic accountability and democratic consolidation remains statistically significant.

Discussion and conclusion

The empirical findings of this study convey that transitional states where reserved domains granted to the military are high, specifically states where the military apparatus has a large role in shaping national policies, are more likely to have lower levels of democratic accountability and are more likely to reverse democratic gains. Furthermore, politicised military institutions and the institutional legacy of overthrowing civilian regimes in the pre-transitional past have an enduring effect in eroding prospects for democratic consolidation.

Scholars provide the normative prescription that reserved domains given to the military should not be provided by the state. In fact, such domains must be immediately dismissed at the onset of roundtable negotiations during “pacted” transitional processes (Diamond, 1999). Civilian governments should not grant the military a “golden parachute” to escape prosecution for human rights violations committed in the authoritarian past. For instance, there is consensus that one reason behind the relative success of Chilean, Brazilian, and Argentinean democratisation is attributed to amendments in their constitution that openly targeted the reserved domains granted to the military. These actions drastically lessened the military’s activism post-transition (Hagopian, 2005). The high-publicity trials of military generals involved in a coup or past human rights violations should be encouraged, to deter the military from pursuing a political role in the affairs of the state. Immunity deals granted to military generals who have denigrated the rule of law should be avoided so as to have a deterrent effect on the military’s potential activism (Siaroff, 2009).

In light of democracy-building initiatives of international organisations, Fitch (1998) recommends that new democracies should discourage military generals from participating in constituent assemblies/constitutional conventions. It is also recommended that generals or retired generals not be invited to take positions in the civil service or bureaucracy. Many also recommend that a civil–military equilibrium must be reached in which civilian governments retain control of the economy, especially in determining budgets that affect the military, their pay scales, and issues concerning the promotion and demotion of highly ranked officers. Lastly, the military’s role in influencing public policy and the consultative role it grants to the executive should be restrained or severed altogether as these reserved domains may threaten democratic values and institutions (Alagappa, 2001; Kohn, 1997; Trinkunas, 2001).

The results of this study also find that other military-related variables, like the size of the army and the percentage of GDP spent on the military, have no effect on democratic quality and consolidation. This corroborates the empirical findings of McKinlay and Cohan (1975) and Zuk and Thompson (1982), which allude to how the corporate model of greed and grievance coming from the military is not a salient factor in shaping its political agenda and motivations. Thus, the effect of the military on democratic outcomes is largely not a structural issue per se; rather, it is more a function of how the military’s activism in the pre-transitional process leads them to demand reserved domains at the cusp of constructing post-authoritarian constitutions. As these domains accrue over time, it provides the impetus for the military to further erode democratic gains, thus paving the way for the return of creeping authoritarianism.

Scholars of democratisation should not neglect the role of the military in influencing democratic outcomes. Further disaggregating and unpacking the political nature of military institutions in transitional democracies and studying their pernicious effect on the process of democratisation is still very much underdeveloped and empirically lacking. In light of the recent transitions in Egypt and Tunisia in the wake of the Arab Spring in 2011, it is important to monitor whether their military institutions insist on reserved domains or allow themselves to be subordinated to civilian rule because their militaries will very likely influence prospects for democratic consolidation. For instance, scholars have begun to offer normative prescriptions on how the unstable transitions in the Middle East and North Africa should learn lessons from the historical legacies of politicised armies in Spain (Martinez, 2012) and Latin America (Varas, 2011) and consider how some countries were able to avoid interventionist armies, which facilitated democratic consolidation.

Footnotes

Appendix 1

Table A1.

Transitional cases and military involvement scores.

Country	Year of transition	Siaroff civil–military score	PRS measure	Coups in their pre-transitional history
Albania	1990	10	2	1
Argentina	1982	8	2.5	1
Armenia	1989	8	3.5	0
Azerbaijan	1989	8	3.5	1
Bangladesh	1990	2	6	1
Belarus	1989	8	4	1
Bolivia	1983	6	4	1
Brazil	1985	8	3	1
Chile	1988	8	2	1
Croatia	1999	10	2	0
El Salvador	1992	6	4	1
Estonia	1989	8	2	0
Ethiopia	1991	8	6	1
Ghana	2000	6	4	1
Honduras	1981	6	4	1
Hungary	1989	10	1	0
Indonesia	1998	6	4.5	0
Kazakhstan	1989	8	2	0
Latvia	1989	10	2	0
Lithuania	1989	10	2	0
Madagascar	1990	6	6	1
Malawi	1992	6	3	1
Mali	1991	6	3.5	1
Mexico	2000	8	2.5	0
Moldova	1989	8	3	1
Mongolia	1990	8	2	0
Mozambique	1992	8	3	1
Nicaragua	1990	8	4	1
Nigeria	1998	6	5	1
Panama	1989	8	2	1
Paraguay	1989	6	5.5	1
Peru	2000	6	2	1
Philippines	1986	8	3.5	1
Poland	1989	10	1	1
Russia	1989	8	2.5	1
Senegal	2001	8	5	0
South Africa	1990	10	2	1
Tanzania	1989	8	3	1
Thailand	1992	2	5	1
Turkey	1981	6	5	1
Uganda	1985	6	5	1
Uruguay	1984	8	3	0
Zambia	1990	8	2	1
Zimbabwe	1976	4	5	1

Notes: The PRS measure is the average score from 1984 to 2008. Transition year data is derived from Karatnycky and Ackerman (2005).

Appendix 2

Table A2.

Summary statistics.

Variable	Mean	Standard deviation	Minimum	Maximum	Data source
PRS military score	3.45	1.21	1.14	5.58	PRS group (various years)
Past coups	0.728	0.450	0	1	Integrated Network for Societal Conflict Research (various years)
Siaroff index	7.69	1.89	2	10	Siaroff (2009)
GNP (Ln)	6.78	0.993	4.72	8.18	World Bank (various years) The world development indicators
Ethnic fractionalisation	0.48	0.246	0.045	0.930	Alesina et al. (2003)
Parliamentarianism	1.45	0.791	1	3	Gerring and Thacker (2004)
Federalism	0.090	0.290	0	1	NSD (various years) Database of political institutions
Socialist past	0.318	0.471	0	1	La Porta et al. (1999)
% Muslim	17.44	29.32	0	99.2	La Porta et al. (1999)
Military participation ratio	2.99	2.78	0	14.67	International Institute for Strategic Studies (various years) Military balance
% GDP spent on the military	1.882	1.325	0.201	6.5	SIPRI (various years)
Economist Intelligence Unit measure on democracy	6.15	1.41	2.53	8.17	EIU (various years)
Bertelsmann Transformation Index	6.95	1.58	3.39	9.56	BTI (various years)
Democratic accountability measure (PRS)	4.50	1.31	1.07	6	PRS Group (various years)

Acknowledgements

An earlier version of this paper was presented at the 69th annual conference of the Midwest Political Science Association, Chicago, Illinois, 1 April 2011. I am grateful to the participants on the panel on comparative democratisation at the annual meeting, most especially, Miguel Centellas—the chair and the discussant. I would also like to thank the editor and the anonymous reviewers of the International Political Science Review for their constructive comments and suggestions.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Notes

Author biography

Rollin F. Tusalem is an Associate Professor of Political Science at Arkansas State University, USA. His main area of interest is in comparative democratisation, political violence, and political institutions. His most recent publications have appeared in the following journals: Comparative Political Studies, Journal of East Asian Studies, Asian Politics and Policy, and Taiwan Journal of Democracy.

References

Aguero

Felipe

(1995) Soldiers, Civilians, and Democracy: Post Franco Spain in Comparative Perspective. Baltimore, MD: Johns Hopkins University Press.

Aguero

Felipe

(1997) Toward civilian supremacy in South America. In: Larry

Diamond

Plattner

Marc

Chu

Yun-han

. (eds) Consolidating the Third Wave Democracies: Themes and Perspectives. Baltimore, MD: Johns Hopkins University Press, pp. 177–206.

Aguero

Felipe

(1998) Legacies in transition: Institutionalization, the military, and democracy in South America. Mershon International Studies Review 42(1): 383–404.

Alagappa

Muthiah

(2001) Investigating and explaining change: An analytical framework. In: Alagappa

Muthiah

(ed.) Coercion and Governance: The Declining Political Role of the Military in Asia. Stanford, CA: Stanford University Press, pp. 29–56.

Alesina

Alberto

Devleeschauwer

Arnauld

Easterly

William

. (2003) Fractionalization. Journal of Economic Growth 8: 155–194.

Barany

Zoltan

(1997) Democratic consolidation and the military: The East European experience. Comparative Politics 30(1): 21–43.

Beck

Nathaniel

Katz

Jonathan

(1995) What to do (and not to do) with time-series cross-section data. American Political Science Review 89(3): 634–647.

Beck

Nathaniel

Katz

Jonathan

(1996) Nuisance versus substance: Specifying and estimating time-series-cross-section models. Political Analysis 6: 1–36.

Bernhard

Michael

Karakoc

Ekrem

(2007) Civil societies and the legacies of dictatorships. World Politics 59(4): 539–567.

10.

Bratton

Michael

Van de Walle

Nicholas

(1997) Democratic Experiments in Africa: Regime Transitions in Comparative Perspective. Cambridge: Cambridge University Press.

11.

Bruneau

Thomas

Goetze

Richard

(2006) Ministries of Defence and democratic control. In: Bruneau

Thomas

Goetze

Richard

(eds) Who Guards the Guardians and How? Democratic Civil–Military Relations. Austin, TX: University of Texas Press, pp. 71–98.

12.

Bruneau

Thomas

Matei

Cristiana

(eds) (2013) The Routledge Handbook of Civil–Military Relations. New York, NY: Routledge University Press.

13.

Bruneau

Thomas

Trinkunas

Harold

(2006) Democratization as a global phenomenon and its impact on civil–military relations. Democratization 13(5): 776–790.

14.

BTI (Bertelsmann Transformation Index) (various years) Transformational scores on democratic change. Available at: http://www.bertelsmann-transformation-index.de/en/bti/

15.

Bunce

Valerie

(2000) Comparative democratization: Big and bounded generalizations. Comparative Political Studies 33(6/7): 703–734.

16.

Cotino-Hueso

Lorenzo

(2007) Supremacía y Control Civiles de la Defensa y las FAS. Seguimiento y Propuestas Para América Latina. Working paper, Documentos CIDOB América Latina, #32, Barcelona, CIDOB. Available at: http://www.cidob.org/alt/militares-latinoamerica/cuestionarios/legismilitar.pdf

17.

Croissant

Aurel

Kuehn

David

(2009) Patterns of civilian control of the military in East Asia’s new democracies. Journal of East Asian Studies 9(2): 187–217.

18.

Crouch

Harold

(1997) Civil–military relations in South-East Asia: Regional challenges. In: Diamond

Larry

Plattner

Marc

Chu

Yun-han

Tien

Hung-mao

(eds) Consolidating the Third Wave Democracies. Baltimore, MD: Johns Hopkins University Press, pp. 207–235.

19.

Diamond

Larry

(1999) Developing Democracy: Toward Consolidation. Baltimore, MD: Johns Hopkins University Press.

20.

Diamond

Larry

Plattner

Marc F.

(eds) (1996) Civil–Military Relations and Democracy. Baltimore, MD: Johns Hopkins University Press.

21.

EIU (Economist Intelligence Unit) (various years) Measure of global democracy. Available at: http://www.eiu.com/public/#

22.

Fish

M. Steven

(2002) Islam and authoritarianism. World Politics 55(1): 4–37.

23.

Fitch

Samuel

(1998) The Armed Forces and Democracy in Latin America. Baltimore, MD: Johns Hopkins University Press.

24.

Fox

John

(1997) Applied Regression Analysis, Linear Models, and Related Models. Thousand Oaks, CA: University of California Press.

25.

Gerring

John

Thacker

Strom

(2004) Political institutions and corruption: The role of unitarism and parliamentarism. British Journal of Political Science 34(1): 295–300.

26.

Gerring

John

Thacker

Strom

Moreno

Carola

(2009) Are parliamentary systems better? Comparative Political Studies 42(3): 327–359.

27.

Hagopian

Frances

(2005) Government performance, political representation, and public perceptions of contemporary democracy in Latin America. In: Frances

Hagopian

Mainwaring

Scott P.

(eds) The Third Wave of Democratization in Latin America: Advances and Setbacks. Cambridge: Cambridge University Press, pp. 319–362.

28.

Heper

Metin

Guney

Aylin

(2000) The military and the consolidation of democracy: The recent Turkish experience. Armed Forces & Society 26(4): 635–657.

29.

Horowitz

Donald

(1993) Democracy in divided societies. Journal of Democracy 4(4): 18–38.

30.

Hunter

Wendy

(1992) Back to the barracks? The military in post-authoritarian Brazil. PhD dissertation, University of California, Berkeley, USA.

31.

Hunter

Wendy

(1994) Contradictions of civilian control: Argentina, Brazil, and Chile in the 1990s. Third World Quarterly 15(4): 633–653.

32.

Hunter

Wendy

(1997) Eroding Military Influence in Brazil. Chapel Hill, NC: University of North Carolina Press.

33.

Huntington

Samuel

(1996) Reforming civil–military relations. In: Diamond

Larry

Plattner

Marc

(eds) Civil–Military Relations and Democracy. Baltimore, MD: Johns Hopkins University Press.

34.

Integrated Network for Societal Conflict Research (various years) Data on coups 1946–2009. Available at: http://www.systemicpeace.org/inscr/inscr.htm

35.

International Institute for Strategic Studies (various years) Military balance. Data provided by the International Institute for Strategic Studies. Available at: http://www.iiss.org/en/publications/military-s-balance/how-s-to-s-access

36.

Karatnycky

Adrian

Ackerman

Peter

(2005) How freedom is won: From civic resistance to durable democracy. Available at: http://agnt.org/snv/resources/HowFreedomisWon.pdf

37.

Karl

Terry

(1990) Dilemmas of democratization in Latin America. Comparative Politics 23(1): 1–22.

38.

Kohn

Richard

(1997) How democracies control the military. Journal of Democracy 8(4): 142–166.

39.

Kuehn

David

(2008) Democratization and civilian control of the military in Taiwan. Democratization 15(5): 870–890.

40.

Kuehn

David

Völkel

Philip

(2011) Explaining civil–military relations in new democracies: Structure, agency and theory development. Asian Journal of Political Science 19(3): 231–249.

41.

Lane

Jan-erik

Ersson

Svant

(2005) The riddle of federalism: Does federalism impact on democracy? Democratization 12(2): 163–182.

42.

La Porta

Rafael

Lopez-de-Silanes

Florencio

(1999) The quality of government. Journal of Law and Economics and Organization 15(1): 222–279.

43.

Lee

Terrence

(2009) The armed forces and transitions from authoritarian rule: Explaining the role of the military in 1986 Philippines and 1998 Indonesia. Comparative Political Studies 42(5): 640–669.

44.

Linz

Juan

Stepan

Alfred

(1996) Problems of Democratic Transitions and Consolidation: Southern Europe, South America and Post-communist Europe. Baltimore, MD: Johns Hopkins University Press.

45.

Londregan

John

Poole

Keith T.

(1990) Poverty, the coup trap and the seizure of executive power. World Politics 42(1): 151–183.

46.

McKinlay

Robert

Cohan

Alvin

(1975) A comparative analysis of the political and economic performance of military and civilian regimes: A cross-national aggregate study. Comparative Politics 8(1): 1–30.

47.

Mares

David

(1998) Civil–Military Relations. Boulder, CO: Westview Press.

48.

Martinez

Rafael

(2012) The democratic consolidation of the military administration: Objectives to be met in light of the Spanish experience. Arab Reform Initiative SSR project. Available at: http://www.arab-reform.net/democratic-consolidation-military-administration-objectives-be-met-light-spanish-experience

49.

Mayorga

Rene

(1997) Democracy dignified and an end to impunity: Bolivia’s military dictatorship on trial. In: McAdams

A. James

(ed.) Transitional Justice and the Rule of Law in New Democracies. South Bend, IN: University of Notre Dame Press, pp. 61–92.

50.

Miller

Andrew

(2006) Military Disengagement and Democratic Consolidation in Post-military Regimes: The Case of El Salvador. New York, NY: Edwin Mellen Press.

51.

Møller

Jørgen

Skaaning

Svend-Erik

(2010) Beyond the radial delusion: Conceptualizing and measuring democracy and non-democracy. International Political Science Review 31(3): 261–283.

52.

Morlino

Leonardo

(1998) Democracy between Consolidation and Crisis: Parties, Groups, and Citizens in Southern Europe. New York, NY: Oxford University Press.

53.

Morlino

Leonardo

(2003) Democrazie e Democratizzazioni. Rome: Il Muno Press.

54.

NSD (Norwegian Social Science Data Services) (various years) Database on political institutions. Available at: http://www.nsd.uib.no/macrodataguide/set.html?id=11&sub=1

55.

Nye

Joseph

(1996) Epilogue: The liberal tradition. In: Diamond

Larry

Plattner

Marc F.

(eds) Civil–Military Relations and Democracy. Baltimore, MD: Johns Hopkins University Press.

56.

O’Donnell

Guillermo

(1996) Illusions about consolidation. Journal of Democracy 7(2): 34–51.

57.

O’Kane

Rosemary

(1983) Towards an examination of the general causes of coups d’états. European Journal of Political Research 11(1): 27–44.

58.

Pion-Berlin

David

(2001) Civil–Military Relations in Latin America. Chapel Hill, NC: University of North Carolina Press.

59.

Pion-Berlin

David

(2011) The study of civil–military relations in new democracies. Asian Journal of Political Science 19(3): 222–232.

60.

Pop-Eleches

Grigore

(2007) Historical legacies and post-communist regime change. The Journal of Politics 69(4): 908–926.

61.

PRS Group (Political Risk Services Group) (various years) Data on the politicization of the military. Available at: http://www.prsgroup.com/

62.

Rice

Condoleeza

(1992) Problems of post-communism: The military under democracy. Journal of Democracy 3(2): 27–42.

63.

Schedler

Andreas

(1998) What is democratic consolidation? Journal of Democracy 9(1): 91–107.

64.

Serra

Narcís

(2010) The Military Transition: Democratic Reform of the Armed Forces. New York, NY: Cambridge University Press.

65.

Shin

Doh

(1994) On the third wave of democratization: A synthesis and evaluation of recent theory and research. World Politics 47(1): 135–170.

66.

Siaroff

Alan

(2009) The military in politics. In: Siaroff

Alan

(ed.) Comparing Political Regimes. Toronto: University of Toronto Press.

67.

SIPRI (Stockholm International Peace Research Institute) (various years) SIPRI military expenditure database. Available at: http://www.sipri.org/databases

68.

Stepan

Alfred

(1988) Rethinking Military Politics: Brazil and the Southern Cone. Princeton, NJ: Princeton University Press.

69.

Thompson

William

(1973) The Grievances of Military Coup-Makers. Beverly Hills, CA: Sage Publications.

70.

Thompson

William

(1980) Corporate coup-maker grievances and types of regime targets. Comparative Political Studies 12(4): 485–496.

71.

Trinkunas

Harold

(2001) Crafting civilian control in Argentina and Venezuela. In: Pion-Berlin

David

(ed.) Civil–Military Relations in Latin America: New Analytical Perspectives. Chapel Hill, NC: University of North Carolina Press, pp. 163–187.

72.

Valenzeula

Julio

(1992) Democratic consolidation in post-transitional settings: Notion, process, and facilitating conditions. In: Scott

Mainwaring

O’Donnell

Guillermo

Valenzuela

J. Samuel

(eds) Issues in Democratic Consolidation: The New South American Democracies in Comparative Perspective. South Bend, IN: University of Notre Dame Press.

73.

Varas

Augusto

(2011) Democratic transitions and the Latin American military. Arab Reform Initiative SSR project. Available at: http://www.arab-reform.net/democratic-transitions-and-latin-american-military

74.

Wilkinson

Steven

(2000) Democratic consolidation and failure: Lessons from Bangladesh and Pakistan. Democratization 7(3): 203–226.

75.

World Bank (various years) The world development indicators. Available at: http://data.worldbank.org/data-catalog/world-development-indicators

76.

Zuk

Gary

Thompson

William

(1982) The post-coup military spending question: A pooled cross-sectional time series analysis. American Political Science Review 76(1): 60–74.

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

0.14 MB

0.08 MB