Defeat in Interstate War and the Probability of Political Liberalization

Abstract

The study examines whether defeat in war increases the probability that states will be involved in an international crisis (as a diversionary policy), enter a process of political liberalization, or alternatively, curtail political rights. More generally, it examines the impact of the leaders’ weakness on their tendency to adopt these different strategies in order to overcome internal unrest. We look at defeat in war as an indicator of the leader’s weakness since we can assume a strong correlation exists between the two. The results showed a significant positive connection to political rights liberalization, indicating that defeat in war increases the probability of liberalization in political rights but does not significantly increase the probability of de-liberalization and diversionary policies. Therefore, the study strengthens the claim that a leader’s weakness tends to push him or her toward initiating political reforms.

Keywords

defeat war liberalization de-liberalization diversion

This article examines whether political leaders employ liberalization, repression, or diversion following defeat in war as a strategy to mitigate its political outcomes. Wars have massive national repercussions—the costs include many casualties, both physical and emotional, huge sums of money, and often the mobilization of society to such an extent that practically every aspect of public life is affected (Jervis, 1976). Thus, not surprisingly, victory and defeat in war tend to have serious political, economic, and social repercussions. Nonetheless, while war is a frequent topic in international relations, most studies analyze the causes of wars rather than the ramifications of their outcomes. Furthermore, only a limited number of studies discuss the effects of war on regime type, transition to democracy, economic progress, and radicalization (e.g., Enia & James, 2015; Reiter, 2001; Stein & Russett, 1980; Zabad, 2012).

National leaders are often held directly responsible for defeat in war, significantly affecting their support base. Leaders must also contend with the numerous challenges generated by the conflict, such as economic downturn and social tensions. Thus, defeat in war often elicits domestic unrest and threatens leadership ability to remain in power (Bueno de Mesquita & Siverson, 1995; Colaresi, 2004). For example, the humiliating defeat of the Arab states in the Six-Day War (1967) caused political unrest and weakened the legitimacy of the Egyptian regime, even leading to the immediate resignation of President Abdel Nasser (Ibrahim, 1993).

Likewise, Leopoldo Galtieri, the president of Argentina, encountered widespread domestic discontent following the country’s defeat in the Falklands War. Taken overall, defeat in war paints the leadership as worthless and incompetent, fanning the public’s desire to punish the guilty and establish an effective deterrent for future leaders (Downs & Rocke, 1994).

Nevertheless, leaders may employ certain strategies to survive the political instability which frequently arises following defeat in war (Davies, 2016; Oakes, 2006). Incumbent leaders seeking to guarantee their political survival may impose a series of repressive measures to weaken or eliminate their opponents. One such example are Saddam Hussein’s oppressive actions against the Kurds and the Shiits in Iraq following his army’s defeat in the First Gulf War (1991).

Alternatively, leaders may employ the opposite strategy, initiating political reforms such as allowing greater representation, loosening media control, or increasing decision-making transparency in an effort to offset internal pressure. For example, Muamar Gaddafi imposed such measures in Libya after his country’s defeat in the Chadian–Libyan War (1987), as did Argentinian president Leopoldo Galtieri following the Falklands War defeat (1982). Leaders can also ignite an external crisis or find a pretext to join an existing one in order to divert public attention from their failure to an external actor (scapegoat tactics). For example, Egypt began the War of Attrition against Israel (1968–1969) almost immediately after its defeat in the Six-Day War (1967).

This article examines whether and how defeat in war influences the strategies employed by political leaders following defeat in war to mitigate political pressure. The first part of this article reviews current academic literature regarding the implications of defeat in war for the political leadership. The second section presents a theoretical model explaining how military failure presents the leadership with a series of major challenges that could threaten its very survival. These challenges may drive the leadership to implement political reform, tighten its internal control by curtailing civil liberties and restricting the media, or become involved in an external crisis in order to relieve internal pressure. The third section reports the findings, and the final section summarizes the study and discusses its implications.

Military Defeat and Leadership Survival

Scholars highlight a close connection between the outcome of a war and political change (Skocpol, 1979). Indeed, many claim that defeat creates fertile ground for radicalization (Harkavy, 2000; Huntington, 2006; Zabad, 2012). A large group of these studies explores the link between defeat in war and the fate of the political leadership. These studies find that military defeat negatively affects incumbents’ political survival (Bueno de Mesquita, Siverson, & Woller, 1992; Bueno de Mesquita & Siverson, 1995; Colaresi, 2004; Goemans, 2000).

Goemans (2000) found that a limited military defeat (low battle deaths per capita) is more likely to reduce the chances of a leader’s survival in a democracy than in an authoritarian regime. By contrast, a serious defeat is found to be more likely to have an equal effect of all regime types. However, other researchers argue that under certain circumstances, authoritarian leaders can be equally susceptible to the consequences of defeat (Chiozza & Goemans, 2004; Debs & Goemans, 2010). Therefore, our research builds on the above studies which suggest that the outcome of a war severely affects the political leadership, seriously jeopardizing a leader’s chances of political survival. However, we claim that leaders do not passively await their anticipated fate. Rather, they can act to prevent their fall from power by employing one (or more) of three main strategies: democratization, repression, and diversion.

Previous studies suggest that war may lead to democratization processes in the defeated country due to direct intervention by the victorious party (Gates & Strand, 2008; Grimm, 2008; Grimm And & Merkel, 2008; Jarstad & Sisk, 2008). For example, democracies often promote democratization in the vanquished states, as it occurred following the two World Wars and in the case of United States intervention in Afghanistan (2001) and Iraq (2003). Another study claims that leaders of a country in a state of war sometimes initiate political reforms themselves. As elites confront an external threat, they may engage in political democratization and wealth distribution in order to elicit patriotism and assist the war effort (Ticchi & Vindigni, 2006). The social modernization that took place in France and Prussia after the 1870 war and expansion of the social and political reforms in several European countries during and after the two World Wars are cases in point (Ticchi & Vindigni, 2006). Mansfield and Snyder (2010) however question the impact of participation in war on democratization.

Several studies discuss democratization processes following a civil war (Cederman, Hug, & Krebs, 2010; Fortna & Huang, 2012; Gurses & Mason, 2008; Hippler, 2008; Joshi, 2010; Wantchekon & Neeman, 2002). Several researchers suggest that, under specific conditions, civil war may lead to democratization, for example, when the military conflict between political factions ends in a draw (Gurses & Mason, 2008; Joshi, 2010). However, these findings are not conclusive (Fortna & Huang, 2012). None of the studies cited above examined the correlation between defeat in war and a leader’s decision to undertake political reforms.

A number of studies examined the effect of participation in war on political repression. Scholars found evidence to suggest that the winning party in a civil war often conducts a purge against the opposition (Baalen & Höglund, 2017; Boyle, 2009; Herreros, 2011). One example is the brutal repression carried out by General Franco’s forces at the end of the Spanish Civil War (1936–1939; Richards, 1998). Several scholars examined the effect of international wars on political repression (Grandi, 2013; Poe & Tate, 1994). Poe and Tate (1994), for example, found that participation in wars correlates positively with domestic repression.

Strikingly, the linkage between consequences of war and diversion strategy has not been examined in the literature. Most studies of diversion focus on social or economic problems as motivation for adopting a diversion strategy (Blomdahl, 2017; Mueller, 1970, 1973; Savun & Gineste, 2019). Our study makes a novel contribution to the literature by examining proactive steps adopted by leaders following defeat in international war by focusing on three strategies—liberalization, repression, and diversion—as alternative courses of action.

Survival Strategies after Military Defeat

In this section, we argue that defeat in war significantly influences the leader’s domestic or foreign policy. Furthermore, we claim that it may influence the leader’s choice of strategy. Defeat in war, more than any other outcome, dramatically affects both state and society, materially and psychologically. Such an outcome in war is usually accompanied by feelings of national humiliation and strong public desire for revenge (Harkavy, 2000; Löwenheim & Heimann, 2008) as well as economic decline, reparations payments, and loss of sizable territorial holdings to the victor.

Scholars often claim that defeat in war gives rise to serious internal political difficulties and can even lead to revolutions (Gates & Strand, 2008; Mitchell, Gates, and Hegre, 1999; Sills, Merton, & Laqueur, 1968; Skocpol, 1979). Defeat weakens leaders both within their own societies and in their dealings with other international actors. Various groups may exploit the situation, forcing leaders to relinquish some or all of their political privileges. Indeed, numerous historical examples highlight the connection between defeat in war and political and social upheaval, among them France’s defeat in the war against Prussia (1870) which sent shock waves throughout French society, led to a regime change and shaped French revanchism for almost half a century; Germany’s defeat in World War I, which the country viewed as an unacceptable national humiliation, led to economic and political chaos, and ultimately played a significant role in the rise of Nazism; the defeat of Japan and Germany in World War II led to fundamental changes in all aspects of both countries’ societies.

How does military defeat affect the political environment? Which actors are mobilized by defeat? Defeat in war can undermine a leader’s political power in three ways. First, defeat deals a blow to the leader’s prestige, seriously weakening his reputation and legitimacy (Bueno de Mesquita, 2005). It evokes feelings of national humiliation because citizens identify powerfully with the state. Furthermore, war exacts a high social toll in fatalities, quality of life, and material destruction, and this is far harder to swallow in case of a defeat. It is psychologically difficult to admit or accept that the nation as a whole is accountable for the defeat, leading to a search for specific persons or groups to blame, motivated by psychological incentives. The political and military leaderships constitute the most obvious targets for this blame due to their paramount role in decision-making during the war. The leader’s loss of legitimacy is not only the product of emotional factors but is also linked to public perception of the leader as incompetent. For example, the protest movement in Israel after the Yom Kippur/October War in 1973 specifically targeted the leadership—Prime Minister Golda Meir, Defense Minister Moshe Dayan, and Chief of Staff David Elazar. This applies yet more forcefully when personal decisions made by the leader or senior leadership led to failures on the battlefield. The public tends to punish leaders who fail to supply public goods (Fearon, 1997; Schultz, 2001; Tomz, 2007; Weeks, 2008), and defeat is a conspicuous case of such a failure. Compared to other political failures, such ineffective management of an international crisis or an unsuccessful economic policy, defeat in war is far more equivocal and thus attracts particular public attention (Bueno de Mesquita & Siverson, 1995). Presumably, the exceptionally high material and psychological costs involved intensify the public’s desire to punish the leader and remove him from power. Culpability increases considerably when the leader intentionally grooms public expectations of a glorious victory, a course of action that magnifies his failure (Fearon, 1994; Schultz, 1998). Furthermore, punishing leaders by removing them from office deters future leaders from making rash public policy decisions, limits damage to the state’s international reputation (Croco, 2011; Downs & Rocke, 1994; McGillivray & Smith, 2008), and minimizes any further damage that could be caused by the current leader’s views (Deegan, 1993; Esposito & Piscatori, 1991; Ibrahim, 1993).

The leader’s perceived responsibility is not necessarily linked to the way the war was actually conducted but at times to the fact of its very occurrence. When a war is perceived as nonessential, namely a war of choice, the public will tend to attribute more blame to the leader for its outcomes, especially if the leader was intimately involved in initiating the conflict (Croco, 2011). Examples are the extremely weak support for George W. Bush during the war in Iraq (2003) or the defeat of the former Spanish Prime Minister, José María Aznar, in the 2004 elections. The latter followed the terror attack in Madrid, which was perpetrated in reaction to Spain sending troops to support the international forces in Iraq. Another focus of public criticism concerns the manner of the conflict’s conclusion. A leader may be regarded as a defeatist who surrendered prematurely or criticized for exhibiting cowardice and submissiveness in negotiating the peace terms. Such perceptions may in turn cause the public or ruling elite to question the leader’s qualities and competence in making rational, intelligent decisions (Richards et. al., 1993). Consequently, the ruling elite may demand that it be accorded a greater role in decision-making or at least call for greater transparency and openness to criticism. If the leader complies, this will lead to certain decentralization of power. However, refusal may cause a schism within the ruling elite, at times leading to a total breakdown of the regime (Bunce, 2000; Carothers, 2002; Huntington, 1993; Palma, 1990). On occasion, defeat generates even harsher reactions: the desire to punish the leader by removing him from office or at least clipping his wings significantly (Dassel & Reinhardt, 1999; Gelpi, 1997; Oakes, 2006).

Several sources of threat emerge when a leader’s legitimacy is questioned. The first source is the general public. Defeat is often followed by public criticism in the form of demonstrations, protest marches, strikes, and armed attacks on government institutions (Carey, 2010) calling for leaders to be held accountable. Where a free press exists, public dissatisfaction spreads throughout the media, fanning public unrest and disseminating it to large swathes of society. For example, following the Yom Kippur/October War in 1973 and the Vietnam War in 1975, the popularity of the political leaderships in the respective countries (Prime Minister Golda Meir in Israel and President Gerald Ford in the United States) declined. President Ford was defeated in the 1976 elections.

In authoritarian states, dissatisfaction produces unrest that can eventually ignite revolt. A threat to the leader may also emerge from the military, economic, and political elites close to him and whose support is crucial to his survival. Faced with the country’s humiliating failure, these elites may demand that the leader shares decision-making with a wider group and also submits to greater oversight of his actions. Already in political distress, the leader is forced to bow to such pressures because, now more than ever, his survival depends on the continued support of these circles. In sum, defeat in war may result in a real challenge to the leader’s rule, irrespective of a regime type.

Among the typical characteristics of a defeat which may influence a leader’s choice of strategy is the erosion or even total collapse of his deterrence capacity, which is especially relevant in nondemocratic regimes. Authoritarian leadership is commonly based on coercion and force: possessing coercive means and a reputation for using them deters other actors, such as opposing parties, the general public, or even other members of the ruling elite, from seeking a greater share of political power. Defeat in war often destroys the domestic security forces on which the leader depends. Furthermore, the victor may impose postwar terms that considerably limit the scope and liberty of the police and army. Defeat can also discredit the repressive mechanisms that the leader wields by ridiculing them in the public eye. The loss of deterrence can lead to a resurgence of preexisting opposing forces, which sense an opportune moment to challenge the leader openly: Indeed, this occurred following the First Gulf War when Kurds and Shiites openly rebelled against Saddam Hussein’s regime.

A second typical consequence of defeat is the increased vulnerability of the vanquished state to pressure from external forces, such as the victory in the war or international organizations. These may pressure the leader to carry out political reforms, either for ideological reasons or to diminish the leader’s powers if he is perceived as a threat to these external forces (Grimm And & Merkel, 2008).¹ In some cases, the victor may demand political reform as a condition of peace. For example, under the Armistice Agreement signed in 1918, the Entente Powers demanded that Germany institutes a series of political reforms (including the removal of the Kaiser). Considering that democracies generally win more wars, the external pressure for political change tends to be toward political liberalization (Mitchell et al., 1999; Reiter & Stam, 1998).² Likewise, the colossal economic cost of a defeat increases the leader’s reliance on outside aid in order to rebuild the country’s economy and reequip its army. These external forces in turn acquire even more leverage than they previously possessed and may make economic, military, or political aid contingent on a series of political reforms (Levitsky & Way, 2005). The special conditions generated by a defeat suggest that, at least theoretically, liberalization may be the optimal strategy for leaders. However, as mentioned above, scholars disagree on this issue. Therefore, in the framework of this research, three hypotheses were tested and are discussed in the empirical section:

Defeat in war pushes the leader toward a strategy of political liberalization.

Defeat in war pushes the leader toward a strategy of de-liberalization of political rights.

Defeat in war pushes the leader toward a strategy of diversionary policy.

Research Design

Data

In order to test the above research hypotheses, the study measured four variables: war outcome, change in liberalization policies, de-liberalization, and diversion. We used Polity IV to measure liberalization and de-liberalization policies. This measurement was chosen for the following reasons: It is the most common measure of democracy, it spans the entire period, 1800–2010, the number of wars in recent decades has fallen, and the use of Polity IV serves to increase the number of observations.

We employed the International Crisis Behavior Project (ICB2), which covers international crises in the period 1918–2006 (Bricher & Wilkenfeld, 2007) to measure the adoption of a diversionary policy. We used involvement in crisis rather than participation in a war or interstate militarized dispute because it is usually unreasonable for a leader to become involved in a war in order to relieve the pressure resulting from a previous defeat in war. However, involvement in an international crisis can divert the tension from the leader’s constituency without necessarily assuming the form of violent activity.

The Correlates of War Project (COW; Version 4.0) was used to measure war outcomes. Not only does this data set report on all international wars from 1816 to 2007, which entailed at least a thousand battle-related fatalities, it also identifies victors and losers. We focused on states that participated in wars that ended between 1918 and 2005, partly due to data availability and partly motivated by a desire to include the two World Wars and increase the study’s external validity. From the list of states that were involved in wars during this period, we excluded states occupied by foreign powers in the course of a war that led to the demise of the old polity, as well as states that underwent a complete collapse of central political authority. In addition, if the variable in the Polity IV data set received a coding of 0 because the power transfer was unregulated or involved a transition to/from unregulated competitiveness lasting more than 3 years, it was treated as a missing value. The final data set for the period 1918–2005 consisted of 154–164 observations (depending on the number of observations available for the specific measure of the dependent variable). After inserting the control variables into the regression models, the missing data were between 22% and 28%. However, the missing data were almost equally distributed before and after World War II and relate to both major powers and small states in various regions.

Variables

The unit of analysis in the study was state-year. For some variables (such as leader change), the unit used was “state-year after the war ended”; for others (such as “change in political rights” or “change in GDP [Gross Domestic Product] per capita”), we used the unit “state-year after the war ended minus the measure for state’s last year of war.” For the control variable “prewar liberties,” we used state-year before the war began.

Dependent variables

The goal of the analysis was to explain patterns of change in policy following a defeat. The policy changes examined liberalization/de-liberalization of political rights and diversion. It is important to stress that these three strategic options are not the only tools available to leaders after suffering a defeat in war. Leaders can also take other routes, such as concealing the fact of defeat or reducing discontent by supplying economic goods. For example, after its defeat by China (1954), Taiwan embarked on an economic program that was designed to “sweeten the bitter pill.” However, we assume that the three strategic options outlined above are the most commonly used and significant. Furthermore, these are also the most frequently discussed options in current literature (e.g., Enia & James, 2015; Miller, 1995).

As noted, the political rights data were drawn from Polity IV (Marshall, Jaggers, & Gurr, 2011), which measures democracy. The main variable of this database consists of several subvariables, three of which are used by scholars to measure political rights. The first is “competitiveness of executive recruitment” (the extent to which prevailing modes of advancement give subordinates equal opportunities to become superordinates; Marshall et al., 2011). The score for this variable ranged from 1 to 3, with 1 representing the lowest score and 3 representing the highest. The second subvariable was “openness of executive recruitment” (recruitment of the chief executive is “open” to the extent that the entire politically active population has an opportunity, in principle, to attain the position via a regularized process). The score for this variable ranged from 1 to 4, with 1 representing the lowest and 4 the highest degree of this variable. The third subvariable was “competitiveness of participation” (the extent to which alternative preferences for policy and leadership can be pursued in the political arena). The scores for this variable ranged from 1 to 5, with 1 representing the lowest level and 5 the highest level of this variable. Although in most cases, Polity does not avoid measuring regime type, in some cases it avoids scoring cases of political rights during wartime when the data were coded as a transfer period (of up to 3 consecutive years), the measure for a year in question was calculated as the average of the last year before the transition period and the first year after the transition period.

To achieve the goal of this study, we calculated the score for each of the above variables in the year after the war, minus the score for the last year of the war. For the multinomial regression, if the result was positive (indicating liberalization of political rights), the code 2 was assigned. If the results revealed deliberalization, the code 0 was assigned. Other cases were coded 1. In the binary logistic models, de-liberalization and liberalization were coded 0 and 2, respectively, in each test, and other cases were coded 1.

The diversionary policy was measured by the ICB2 database (Bricher & Wilkenfeld, 2007). This variable measured the involvement of states in international crises. The measure of the data set was taken from the year after the war concluded and was coded 3 if the state was involved in an international crisis in the year following the end of the war and 1 if not.

Independent Variable

The independent variable was “war outcome,” which was measured using the COW (Version 4.0; Sarkees & Wayman, 2010). War outcome was coded as follows in the data set: Winner, Loser, Compromise/Tied, Stalemate, War transformed into another type of war, War ongoing as of 12/31/2007, Conflict continues at below war level, and Changed sides. The present study excluded the last four war outcome codes because we could not determine the outcomes of the conflicts (27 observations out of 227).³ In addition, because it was difficult to rank the different outcomes and, more importantly, due to our interest in the effect of defeat in war, the remaining outcome categories were coded dichotomously: defeat = 0 and no defeat (namely Winner/Compromise/Tied and Stalemate) = 1.

The analysis did not include militarized interstate disputes, among them conflicts involving minimal casualties, because we assume that only major violent events will have a significant political impact on leaders.

Control variables

The first control variable is “War initiator.” The COW coding for whether a state initiated a war or not is yes (1) or no (2). This variable was therefore coded dichotomously in the research design. The underlying rationale of this variable is that political pressure on the leadership may be stronger in a state that initiates a war than in a state forced into war. As such, the probability of political liberalization as a means of reducing political pressure may be higher when the leader initiates the war.

The second control variable is “Casualties per Capita.” COW terms this “Battle Deaths,” referring to combat fatalities suffered by the state. However, battle deaths alone are meaningless because the price of casualties is felt more sharply when the nation is small, and the probability of outcry is higher when the public feels the price exacted by the war more keenly: the more intense the public outcry over the number of combatant fatalities, the greater the political pressure exerted on the leadership. Therefore, “battle death” in this study was measured as the number of battle deaths in the last year of the war relative to the whole population. The population data sources used for this purpose were Maddison Historical Statistics (Maddison, 2010) and the World Bank World Development Indicators for 2012.

The third control variable is “Change in GDP per capita between the last year of the war and the first year after the war.” The rationale for this variable was that because wars demand high public spending, both during the conflict and in the subsequent recovery period, they may cause an economic downturn, regardless of their outcome. This can lead to political pressure and to the leader’s attempts to mitigate it by liberalizing political rights. The sources for GDP per capita were Maddison Historical Statistics and the World Bank World Development Indicators 2012 (Maddison, 2010).

The fourth control variable was “Change of leader.” The underlying rationale was that political rights may change because the leader was replaced, not only because the state was defeated in war. The sources for this variable were the Statesman’s Yearbook and Zarate Political Collection (2013). If the leader of the state was replaced in the year following the end of the war, the code 1 was assigned, if not, the code was 0.

The fifth control variable is the measure of democracy in the year before the war started. This control variable was inserted for three reasons: First, we control for regime type because democratic leaders may not select de-liberalization due to their normative accountability. Second, the probability of structural change is low in these cases: if the state scores high for democracy and the change would be toward liberalization, or if it scores low and the change would be toward de-liberalization, the added value of such a policy would be smaller.

Third, it is reasonable that as the state is accorded a higher or lower score in terms of democracy, the probability of structural changes is lower because the possibility of liberalization or de-liberalization, respectively, declines and the added value of that policy will be less. Furthermore, political liberalization after a war may simply be the result of de-liberalization that occurred during the war. Hence, postwar liberalization is merely a return to the domestic status quo ante bellum.

Findings

Table 1 presents the statistics concerning changes in political liberties and involvement in an international crisis in general (not only after wars).

Table 1.

Statistics Describing Changes in Political Liberties.

Variables	Openness of Executive Recruitment	Competitiveness of Executive Recruitment	Competitiveness of Participation	Crisis Involvement
Total observations	9,122	9,120	9,104	9,122
Total observations of change	334	331	462	957
Percentage of total liberalization	3.6	3.6	5	10.4
Percentage of total	2	1.8	2.9

The observations in Tables 1 and 2 relate to policy changes that occurred from one year to the next for all states covered by Polity IV in the period 1918–2005. This does not include observations of foreign interruption, interregnum, anarchy, or cases of transition.

Table 2.

Distribution of Observations by Defeat/Success.

Variables	Openness of Executive Recruitment		Competitiveness of Participation		Competitiveness of Executive Recruitment
	Defeat	Victory	Defeat	Victory	Defeat	Victory
	47^a	107	56	109	46	108
No change	39	89	44	89	35	92
Liberalization	4	3	6	2	5	1
Crisis	4	15	4	15	4	15
De-liberalization	—	—	2	3	—	2
No change	0.83	0.832	0.786	0.817	0.761	0.852
Liberalization	0.085	0.028	0.107	0.018	0.109	0.009
Crisis	0.085	0.140	0.071	0.138	0.087	0.139
De-liberalization	—	—	0.036	0.028	—	0.019

^aCases of defeat are fewer than cases of success for two reasons: the “victory” category includes successes and stalemates and many of the states defeated in wars were occupied by foreign powers, which either terminated the old polity or else brought about a complete collapse of central political authority.

As Table 1 indicates, deviations in all variables are very small. The greatest percentage of change was about 10% (involvement in crisis), and the percentage of change in political rights (liberalization and de-liberalization) is very small (3.6–5%). Thus, the variables measured in this study show that changes in political liberties and involvement in international crises are both rare. It is therefore normal to find little alteration in measurements of change or involvement in international crises following wars. In addition, the percentage of liberalization cases is slightly higher than that of de-liberalization cases, though the difference is slight (for detailed data, see Table A1 in Appendix).

In cases of both defeat and victory, most observations show no change. However, the percentage of liberalization after defeat is much higher (round 3.5–5.5 times) than in the general population of countries. Moreover, in case of victory, the percentage of liberalization is lower than the general population in terms of competitiveness of participation and competitiveness of executive recruitment; with regard to openness of executive recruitment, the percentage is slightly lower.

In all measures of political rights, the percentage of liberalization is much higher following defeat in war than it is following victory. For involvement in international crises, the percentage is higher in cases of victory in war than in cases of defeat, though the variance is smaller than that in the liberalization policy. The comparison of de-liberalization cases indicates that the difference between defeat and victory is much smaller than for liberalization and involvement in an international crisis.

Under normal circumstances, the ideal way to test the influence of defeat on leadership decisions is to conduct a multinomial logistic regression, testing all three policy changes as parallel options available to the leader and discerning which was chosen. Yet, as Table 2 shows, in openness of executive recruitment, there are no observations of de-liberalization, and in competitiveness of executive recruitment, there are no instances of de-liberalization in case of a defeat.

Furthermore, in the multivariable regressions, the few observations of involvement in a crisis in cases of victory dropped out due to missing data in other variables. In addition, the cases of liberalization, de-liberalization, and involvement in international crisis do not overlap. Therefore, we conducted binary logistic regressions for each policy change separately; only with regard to competitiveness of participation where we conducted a multinomial regression, and only for liberalization versus de-liberalization.

As Table 3 shows, the only policy that was found to have a positive significant connection to war outcome is liberalization. De-liberalization and diversionary policies demonstrate no significant links with war outcomes. Furthermore, almost all the control variables have no significant connection with any of the three policies. Thus, defeat in war increases the probability of liberalization but not of de-liberalization or diversionary policies.

Table 3.

Binary Logistic Regressions for the War Outcome Variable, Control Variables, and Change in Policy.

	LiberalizationB (SE)			De-Liberalization B (SE)	Diversion B (SE)
Variables	XROPEN^a	XRCOMP^b	PARCOMP^c	PARCOMP	XROPEN	XRCOMP	PARCOMP
Initiator	−17.3 (5,789.4)	−0.38 (1.26)	−.79 (1.27)	−0.24 (1.25)	−.38 (.68)	−.35 (.69)	−.36 (.69)
GDP	0.001(0.002)	−0.001 (0.002)	.001 (.002)	−0.001 (0.001)	.001 (.001)	.001 (.001)	.001 (.001)
Death rate	0.00 (0.00)	0.00 (0.00)	.000 (.000)	0.000 (0.000)	.00 (.00)	.00 (.00)	.00 (.00)
Change of leader	−0.49 (1.31)	1.43 (1.22)	−.03 (1.3)	−1.07 (1.05)	−.41 (.83)	−.40 (.84)	−.30 (.85)
Prewar	−1.07	−1.18	−.40	−0.272	.35	.003	−.04
Liberties	(.40)***	(.62)*	(.38)	(.33)	(.51)	(.31)	(.19)
N	115	115	116	117	115	115	116

^aOpenness of executive recruitment. ^bCompetitiveness of executive recruitment. ^cCompetitiveness of participation.

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

Table 4 does not show a significant relationship between the control variables and change in competitiveness of participation. The liberalization category only has a positive, significant connection to war outcome. The de-liberalization policy change was not found to have a significant connection to war outcome. Consequently, defeat in war increases the probability of liberalization but not of de-liberalization. These findings increase the robustness of the findings in Table 3.

Table 4.

Multinomial Regression for Relationship between the War Outcome Variable, the Control Variables, and the Change in Competitiveness of Participation Variable.

Variables		B	SE	Wald
Liberalization	Outcome	2.11*	0.932	5.129
	Initiator	−1.151	1.252	0.845
	GDP	0.002	0.001	2.507
	Death rate	0.000	0.000	0.827
	Change of leader	−0.224	1.255	0.032
	Prewar political liberties	−0.333	0.342	0.952
De-Liberalization	Outcome	2.079	1.345	2.390
	Initiator	0.556	1.330	0.175
	GDP	0.001	0.002	0.295
	Death rate	0.000	0.000	1.054
	Change of leader	0.025	1.413	0.000
	Prewar political liberties	0.718	0.466	2.374

Note. N = 116.

*p < .05.

Discussion and Conclusions

This study examined whether defeat in war increases the probability that states will subsequently become involved in another international crisis, enter a process of political liberalization, or, alternatively, curtail political rights. We found that only one of these policies, undertaking a process of liberalization, is significantly connected to defeat in war.

Thus, defeat in war increases the probability that leaders will choose liberalization as a strategy to overcome domestic unrest. No such tendency was found regarding the two alternative strategies, namely involvement in another international crisis or de-liberalization of political rights. These findings support the hypothesis that, after defeat in war, leaders find it difficult to employ violent means; both involvement in an international crisis and de-liberalization of political rights require the use of force, or at least the threat of using of force, making these courses of action less appealing after a defeat. At the same time, liberalization does not require the use of force and, from the leader’s perspective, may even reduce the need to use it.

The current study provides several important insights that enrich the existing literature. The findings counter the arguments by some scholars that a leader’s weakness can push him toward diversionary or repressive actions in order to overcome domestic strife (Blomdahl, 2017; Mueller, 1970, 1973; Savun & Gineste, 2019). The findings support the commonly held claim in democratization literature that a leader’s weakness correlates positively with liberalization (Bratton & Van de Walle, 1992; Joseph, 1997). Furthermore, the findings indicate that changes in political liberties are rare. Indeed, we find only few examples of postwar liberalization. Nevertheless, the percentage of cases of postwar liberalization is much higher than the frequency of liberalization among the general population of states during the period 1918–2005. More importantly, for all measures of political liberties, the liberalization rate was higher after defeat in war than following victory.

Our findings improve our understanding of the connection between war and domestic politics in general and democratization processes in particular. Our theoretical model explains why short-term tactics may be employed after defeat in war. Nevertheless, defeat may have long-term consequences. In this respect, future studies can examine whether there is a link between defeat in war and public investments in welfare.

Footnotes

Appendix

Table A1.

Distribution of Observations.

	No Change	Liberalization	Crisis	De-Liberalization	Total
Openness of executive recruitment	128	7	19	—	154
Competitiveness of participation	133	8	19	5	165
Competitiveness of executive recruitment	127	6	19	2	154
Openness of executive recruitment	0.831	0.045	0.123	—
Competitiveness of participation	0.806	0.048	0.115	0.03
Competitiveness of executive recruitment	0.825	0.039	0.123	0.013

Note. Most of the observations revealed no policy change. The second largest category (much smaller than the no policy change category) is involvement in crisis, after that, liberalization; the smallest group of policy change is de-liberalization. In the openness of executive recruitment, there were no observations of de-liberalization.

Declaration of Conflicting Interests

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

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Shlomo O. Goldman

Notes

References

Baalen

Höglund

(2017). “So, the killings continued”: Wartime mobilization and post-war violence in Kwazulu-Natal, South Africa. Terrorism and Political Violence, 1–19.

Blomdahl

(2017). Diversionary theory of war and the case study design: President Clinton’s strikes on Iraq and Yugoslavia. Armed Forces & Society, 43, 545–565.

Boyle

M. J.

(2009). Explaining strategic violence after wars. Studies in Conflict & Terrorism, 32, 209–236.

Bratton

Van de Walle

(1992). Popular protest and political reform in Africa. Comparative Politics, 24, 419–442.

Bricher

Wilkenfeld

(2007). International crisis behavior project 1918–2006. Inter-university Consortium for Political and Social Research, 1990.

Bueno de Mesquita

(2005). The logic of political survival. Cambridge: MIT Press.

Bueno de Mesquita

Siverson

R. M.

(1995). War and the survival of political leaders: A comparative study of regime types and political accountability. American Political Science Review, 89, 841–855.

Bueno de Mesquita

Siverson

R. M.

Woller

(1992). War and the fate of regimes: A comparative analysis. The American Political Science Review, 86, 638–646.

Bunce

(2000). Comparative democratization big and bounded generalizations. Comparative Political Studies, 33, 703–734.

10.

Carey

S. C.

(2010). The use of repression as a response to domestic dissent. Political Studies, 58, 167–186.

11.

Carothers

(2002). The end of the transition paradigm. Journal of Democracy, 13, 5–21.

12.

Cederman

L.-E.

Hug

Krebs

L. F.

(2010). Democratization and civil war: Empirical evidence. Journal of Peace Research, 47, 377–394.

13.

Chiozza

Goemans

H. E.

(2004). International conflict and the tenure of leaders: Is war still ex post inefficient? American Journal of Political Science, 48, 604–619.

14.

Colaresi

(2004). Aftershocks: Postwar leadership survival, rivalry, and regime dynamics. International Studies Quarterly, 48, 713–728.

15.

Croco

S. E.

(2011). The decider’s dilemma: Leader culpability, war outcomes, and domestic punishment. American Political Science Review, 105, 457–477.

16.

Dassel

Reinhardt

(1999). Domestic strife and the initiation of violence at home and abroad. American Journal of Political Science, 43, 56–85.

17.

Davies

G. A. M.

(2016). Policy selection in the face of political instability: Do states divert, repress, or make concessions? Journal of Conflict Resolution, 60, 118–142.

18.

Debs

Goemans

H. E.

(2010). Regime type, the fate of leaders, and war. American Political Science Review, 104, 430–445.

19.

Deegan

(1993). The middle east and problems of democracy. London, England: Open University Press.

20.

Downs

G. W.

Rocke

D. M.

(1994). Conflict, agency, and gambling for resurrection: The principal-agent problem goes to war. American Journal of Political Science, 38, 362–380.

21.

Enia

James

(2015). Regime type, peace, and reciprocal effects. Social Science Quarterly, 96, 523–539.

22.

Esposito

J. L.

Piscatori

J. P.

(1991). Democratization and Islam. Middle East Journal, 45, 427–440.

23.

Fearon

J. D.

(1994). Domestic political audiences and the escalation of international disputes. The American Political Science Review, 88, 577.

24.

Fearon

J. D.

(1997). Signaling foreign policy interests tying hands versus sinking costs. Journal of Conflict Resolution, 41, 68–90.

25.

Fortna

V. P.

Huang

(2012). Democratization after civil war: A brush-clearing exercise. International Studies Quarterly, 56, 801–808.

26.

Gates

Strand

(2008). Military intervention, democratization, and post-conflict political stability. In Oberg

Kaare

(Eds.), Resources, Governance and Civil Conflict (pp. 147–163). London, England: Routledge/ECPR. Retrieved from http://lawlib.shirazu.ac.ir:8080/pdfTemp/ebooksclub.org__Resources__Governance_and_Civil_Conflict__Routledge_Ecpr_Studies_in_European_Political_Science_pdf#page=168

27.

Gelpi

(1997). Democratic diversions governmental structure and the externalization of domestic conflict. Journal of Conflict Resolution, 41, 255–282.

28.

Goemans

H. E.

(2000). Fighting for survival the fate of leaders and the duration of war. Journal of Conflict Resolution, 44, 555–579.

29.

Grandi

(2013). New incentives and old organizations: The production of violence after war. Peace Economics, Peace Science and Public Policy, 19, 309–319.

30.

Grimm And

Merkel

(2008). War and democratization: Legality, legitimacy and effectiveness. Democratization, 15, 457–471.

31.

Grimm

(2008). External democratization after war: Success and failure. Democratization, 15, 525–549.

32.

Gurses

Mason

T. D.

(2008). Democracy out of anarchy: The prospects for post-civil-war democracy*. Social Science Quarterly, 89, 315–336.

33.

Harkavy

R. E.

(2000). Defeat, national humiliation, and the revenge motif in international politics. International Politics, 37, 345–368.

34.

Herreros

(2011). Peace of cemeteries: Civil war dynamics in postwar states’ repression. Politics & Society, 39, 175–202.

35.

Hippler

(2008). Democratization after civil wars—Key problems and experiences. Democratization, 15, 550–569.

36.

Huntington

S. P.

(1993). The third wave: Democratization in the late twentieth century. Norman: University of Oklahoma Press.

37.

Huntington

S. P.

(2006). Political order in changing societies. New Haven, CT: Yale University Press.

38.

Ibrahim

S. E.

(1993). Crises, elites, and democratization in the Arab world. Middle East Journal, 47, 292–305.

39.

Jarstad

A. K.

Sisk

T. D.

(2008). From war to democracy: Dilemmas of peacebuilding. Cambridge, England: Cambridge University Press.

40.

Jervis

(1976). Perception and misperception in international politics. Princeton, NJ: Princeton University Press.

41.

Joseph

(1997). Democratization in Africa after 1989: Comparative and theoretical perspectives. Comparative Politics, 23, 363–382.

42.

Joshi

(2010). Post-civil war democratization: Promotion of democracy in post-civil war states, 1946–2005. Democratization, 17, 826–855.

43.

Levitsky

Way

(2005). International linkage and democratization. Journal of Democracy, 16, 20–34.

44.

Löwenheim

Heimann

(2008). Revenge in international politics. Security Studies, 17, 685–724.

45.

Maddison

(2010). Maddison historical statistics. Groningen, the Netherlands: Groningen Growth and Development Centre, University of Groningen.

46.

Mansfield

E. D.

Snyder

. (2010). Does War Influence Democratization?. In Kier

Krebs

R. R.

(Eds.), In War’s Wake: International Conflict and the Fate of Liberal Democracy (pp. 23–49). Cambridge, England: Cambridge University Press.

47.

Marshall

M. G.

Jaggers

Gurr

T. R

. (2011). Political regime characteristics and transitions, 1800–2010 Polity IV project. Center for Systemic Peace. Retrieved from http://www.systemicpeace.org/polity/polity4.htm

48.

McGillivray

Smith

(2008). Punishing the prince: A theory of interstate relations, political institutions, and leader change. Princeton, NJ: Princeton University Press.

49.

Miller

R. A.

(1995). Domestic structures and the diversionary use of force. American Journal of Political Science, 39, 760–785.

50.

Mitchell

S. M.

Gates

Hegre

(1999). Evolution in democracy-war dynamics. Journal of Conflict Resolution, 43, 771–792.

51.

Mueller

J. E.

(1970). Presidential popularity from Truman to Johnson. The American Political Science Review, 64, 18–34.

52.

Mueller

J. E.

(1973). War, presidents, and public opinion. Hoboken, NJ: Wiley.

53.

Oakes

(2006). Diversionary war and Argentina’s invasion of the Falkland Islands. Security Studies, 15, 431–463.

54.

Palma

G. D.

(1990). To craft democracies: An essay on democratic transitions. Berkley: University of California Press.

55.

Poe

S. C.

Tate

C. N.

(1994). Repression of human rights to personal integrity in the 1980s: A global analysis. American Political Science Review, 88, 853–872.

56.

Reiter

(2001). Does peace nature democracy? The Journal of Politics, 63, 935–948.

57.

Reiter

Stam

A. C.

(1998). Democracy, war initiation, and victory. American Political Science Review, 92, 377–389.

58.

Richards

Morgan

T. C.

Wilson

R. K.

Schwebach

V. L.

Young

G. D.

(1993). Good times, bad times, and the diversionary use of force a tale of some not-so-free agents. Journal of Conflict Resolution, 37, 504–535.

59.

Richards

(1998). A time of silence: Civil war and the culture of repression in Franco’s Spain, 1936–1945. Cambridge, England: Cambridge University Press.

60.

Sarkees

M. R.

Wayman

(2010). Resort to war: 1816–2007. Correlates of War (COW Version 4.0). Washington, DC: CQ Press.

61.

Savun

Gineste

(2019). From protection to persecution: Threat environment and refugee scapegoating. Journal of Peace Research, 56, 88–102.

62.

Schultz

K. A.

(1998). Domestic opposition and signaling in international crises. American Political Science Review, 92, 829–844.

63.

Schultz

K. A.

(2001). Looking for audience costs. Journal of Conflict Resolution, 45, 32–60.

64.

Sills

D. L.

Merton

R. K.

Laqueur

(1968). Revolution. In International Encyclopedia of the Social Sciences (Vol. 13, pp. 501–507). New York: Macmillan New York.

65.

Skocpol

(1979). States and social revolutions: A comparative analysis of France, Russia and China. Cambridge, England: Cambridge University Press.

66.

Stein

A. A.

Russett

(1980). Evaluating war: Outcomes and consequences. In Gurr

T. R.

(Ed.), Handbook of Political Conflict: Theory and Research (pp. 399–422). New York, NY: Free Press. Retrieved from http://www.grandstrategy.net/Publications/Articles/Articles/evaluating_war.pdf

67.

Ticchi

Vindigni

(2006). On wars and political development: The role of international conflicts in the democratization of the west. Institute of Governmental Studies Deposit. Retrieved from https://escholarship.org/uc/item/56x218x7

68.

Tomz

(2007). Domestic audience costs in international relations: An experimental approach. International Organization, 61, 821–840.

69.

Wantchekon

Neeman

(2002). A theory of post-civil war democratization. Journal of Theoretical Politics, 14, 439–464.

70.

Weeks

J. L.

(2008). Autocratic audience costs: Regime type and signaling resolve. International Organization, 62, 35–64.

71.

Zabad

. (2012). The political consequences of defeats in wars. Oakland: University of California. Retrieved from http://lalev.co/The%20Political%20Consequences%20of%20Defeats%20in%20Wars.pdf

72.

Zarate Political Collection. (2013). Retrieved from http://www.terraes/personal2/monolith/00index2.htm