TUNISIA AND SYRIA ARMY IN THE REGIME CHANGE 

I.      Introduction and Context

The Arab World has recently experienced unprecedented popular mobilization that changed the political landscape of the region. From Tunisia to Syria, Arab societies have demanded political and economic freedom in a context of economic crisis. A crucial aspect of Arab transitions lies in the speed at which supposedly strong leaders were overthrown. Regime changes in Tunisia and Egypt showed the relative weaknesses of these governments dictated by Ben Ali and Moubarak, respectively. These two regimes had maintained the illusion that their power is rooted in all strata of the population, with an unquestionable loyalty of the security apparatus and de facto, of the pressured population. Ben Ali’s authoritarian regime successfully maintained the myth of a performing state until the 2008 food crisis and the global economic crisis. As Beatrice Hibou analyses in her book, Tunisia under Ben Ali could be defined as a “policing state”. That is a country in which there is an absolute oversight of the population. The taxi sector constitutes a good example of how the regime used the population as “watchdogs” to denounce criticism towards the regime. The force of obedience established by Ben Ali’s regime is not about police force or a strong regime. It is all the more about the control of economic policies to pressure certain groups of the population, such as taxi drivers, to keep an eye on potential threats (Hibou, 2006). However, this control has showed its limits in recent years, as the illusion of a performing state vanished.

First, the food crisis of 2008 resulted in unprecedented protests in the Gafsa Mining Basin. Expending geographically to the whole region, riots gathered demonstrators from different classes such as miners, students, unemployed and lawyers. Slogans in the streets directly criticized a corrupted regime and its inability to counter unemployment and inequality (Cloutier, 2012). Police forces started a violent repression that was unsuccessful in stopping the revolt. The army intervened thereafter, killing 3 protesters and injuring 10 others. Demonstrators immediately abandoned the struggle against the regime after the army’s intervention. Second, the suicide of the street vendor Mohamed Bouazizi on 17 December 2010 detonated anger in the populace. In two weeks, demonstrations spread nationally, affecting all major cities in Tunisia. While police forces struggled to contain  protesters in different regions, the army protected strategic infrastructure and   official

buildings. On 12 January 2011, the regime declared a curfew and the army was deployed in cities’ streets. After the population consistently refused to respect this curfew, the army was assigned to repress those who disrespect the law with the obligation to open fire. Military forces, led by its army chief of staff, decided to reject the Tunisian regime’s order to repress protesters (Murphy, 2011). Ben Ali escaped from his residence only two days after the army’s noncompliance. This crucial decision has mostly been omitted by the literature on regime change in the Arab World. This article acknowledges the importance of pressure from below to explain Ben Ali’s fall. However, popular mobilizations do not constitute a sufficient factor to explain the Tunisian’s rapid political transition. This article aims to elucidate a paradox in Tunisia’s regime change; namely, how such a perceived strong regime, with an absolute oversight of its population, broke down in the first few week of year 2011. Without neglecting the importance of the popular pressure, we argue that the army played a central role in the fall of Ben Ali. Why did the Tunisian army refused Ben Ali’s order to open fire at the demonstrators in 2011? What are the implications of such a decision? This paper is the first attempt to compare two consecutives events in which the army has decided to open fire at its own population, namely the 2008 protests in the Gafsa region and the massive revolutionary protests of early 2011. What is the rational thinking behind such decisions? Did the army have enough of the generalized corrupted regime organized around Ben Ali’s personality? Or, instead,   did it choose to back the 2011 protests because it was simply better-off for doing so? This paper will tackle questions on what consequences this decision had on the key actors of the uprising, namely Ben Ali, the demonstrators and foreign allies.

We find that the army was better-off to refuse Ben Ali’s order at this point in time because the probability that Ben Ali leaves power was high enough. We will show how this decision has been a turning point in the Tunisian uprising by altering motivations of the key players. After analyzing the evolution of the civil-military relationships, an extended model of game theory will be introduced to depict the interactions between Ben Ali’s regime and the army. This model focuses on one decisive aspect; namely the decision of the army to accept or refuse Ben Ali’s order to shoot at protesters. We compare two major demonstrations in which the army had the choice to shoot at people: the Revolt of the Gafsa Mining Basin in 2008 and the uprising of 2011 that lead to the Tunisian transition.

II.  Literature Review

Most recent studies explained that unprecedented pressure from below caused the end of Ben Ali’s regime (Joffe, 2011; Jdey. 2012; Bellin, 2012; Kuhn, 2012; Campante

and Chor, 2012; Aleya-Sghaier, 2012; Allal and Geisser, 2011-2012; Illan, 2012; El- Khawas, 2012). An interesting approach brings the middle class as an important feature of a democratic society and the process of democratization (Leventoglu, 2013). Critics of this precondition have argued that transitions in Latin American countries saw actors from the middle class supported coups. Nevertheless, this seems to be relevant to understand the Arab Spring. Tunisia, for instance, known as a middle-class society, has seen popular mobilization from every strata of the society to overthrown the dictator Ben Ali. In Egypt, the Muslim Brotherhood constituted a very organized opposition force, which gathered not only the poor but also democratic and secular young Egyptians (Way, 2011). The middle class is clearly an important precondition to the extension of popular protests that started Arab transitions but is certainly not necessary to explain variations of transitions outcomes in the Arab World.

The occurrence of democratic transitions may also be explained by the strategic interactions of principal agents, namely the elites and the civil society.

Joffe G. (2011) argues that Ben Ali’s fall was due to the organization of the  population after spontaneous demonstration and emphasizes on the role of the Tunisian Labor Union UGTT. Bellin E. (2012) focuses on the “power of contagion” and collective action from the population to put pressure on the regime. While Kuhn R. (2012) sees a direct link between the improvements of human development and social mobilization, Campante F. R. and Chor D (2012) show how the level of education of Tunisians affected the demand for economic opportunities. Economic crisis and the lack of job opportunities were, according to them, major to explain the national protests.

Although pressure from below was determinant to explain regime change in the Arab World and more precisely, in Tunisia, it does not seem to be sufficient to fully understand the rapidity of the Tunisian transition. A few articles have tried to identify the rupture between elites, and especially the role of the army to trigger change in Tunisia (Way, 2011; Barany, 2012; Gause, 2011; Sorenson, 2012; Dalacoura,   2012).

Barany Z. (2012) argues that the lack of support from the army was necessary to the success of popular mobilization. He explains that the regime’s lack of consideration, the disinclination of corruption organized by the executive power as well as the non- political status of a professional army justified the non-willingness of the Tunisian troops to repress its own people. However, even though tensions between the  Tunisian army and the regime have affected the army’s decision to refuse Ben Ali’s order to open fire on the protesters, Barany Z. (2012) omits the army’s decision to open fire on the protesters during the 2008 protest of Gafsa. Thus the disdain and the status of the Tunisian army towards the regime is not sufficient to explain the army’s decision.  Gause G. (2011) argues that the army has taken the opportunity to risk not  to back the regime in order to play a more important role after the transition (Dalacoura K., 2012). This argument has also shown its limitations since the Tunisian army has not been actively part of the political debate unlike the Egyptian army that organized a coup early July 2013. The literature on the role of the army in Ben Ali’s fall do not take into account historical and institutional factors to explain the reasons held by the army to not back Ben Ali’s regime. In addition to that, it appears that  these studies mainly used a descriptive analysis instead of an explanatory demonstration.

Instead, I find that the evolution of the civil-military relationships as well as the high institutionalization of the Tunisian army constituted the preconditions in the ability of the army to make a rational decision to back street protesters. Before demonstrating why the army refused the order to shoot at protesters – using an extended model of game theory – I will analyze key historical and institutional factors to explain current contentious relationships as well as the relative independency of the army to Ben  Ali’s regime.

III.             Analysis: the role of the Tunisian army in Ben Ali’s fall

3.1. The Evolution of the civil-military relationships

The uniqueness of Tunisia’s experience in Arab transitions is mainly due to the  special features of the military forces in the Tunisian society.

Ben Ali ruled Tunisia for 25 years through his party, the “Rassemblement constitutionnel démocratique” (RCD). About 2 million citizens (20% of the population) were official members of his party. Most opposition parties were controlled and cooperated with the RCD. Threats from any oppositional organizations were forced into exile such as the powerful Islamist party. Through his personal security forces as well as the growing influence of his step family on the Tunisian economy, Ben Ali built a powerful system of control in Tunisia. In December 2010, when protests erupted with the suicide of Mohammed Bouazizi in Sidi Bouzeid, popular protests quickly spread in the country. Demonstrators demanded the end of corruption and political liberalization reforms at first. However, the regime relied on his security forces, the police as well as the army to contain national protests. While the Tunisian police and Ben Ali’s personal forces opened fire on the demonstrators, the army deployed its troops into Tunisia’s key locations without intervening actively. In fact, newspapers in Tunisia notified that the army had a surprisingly sympathetic relationship with street protesters and actively helped them (Schraeder and Redissi (2011). How can we explain such a behavior of military forces, looking at Tunisia’s history and institutions?

Overall, the evolution of military forces in Tunisia from the 1980’s is characterized  by its professionalization as well as its depoliticization. Ben Ali as well as his predecessor Habib Bourgiba were opposed to the integration of the army into the political arena. Ben Ali was himself a General from the army that took power in a bloodless coup d’état in 1987. Tunisia can thus be described as an autocratic officer- politician regimes, in which a former military officer is head of state though  converted into a civilian politician. This differs from a military-ruled regime as the army doesn’t have a direct influence on politics (Appendix 1). Despite Ben Ali’s military background, he has constantly limited the political power of the Tunisian army by fear of a military coup. It is also worth noticing that the so-called ‘silent’ or

‘invisible’ army does not have economic power (Anderson, 2011), unlike the  Egyptian Army that led to a military regime. This further explains the lack of political ambition of the military forces.

It is necessary to consider the degree of institutionalization as a potential factor of regime change triggered by the army’s decision power (Bellin, 2004). Bellin argues that the more institutionalize military forces are, the more likely they will support political reforms and disconnect from political elites’ interests. Because the security apparatus is based on meritocracy and performance bounded by a clear set of rules, ties with the regime are not as embedded as in patrimonial relationships. High institutionalization of the military forces thus indicates that they are relatively committed to defend national interest overall instead of certain groups, and clearly identify themselves as an entity completely separated from the central political power. Instead, we can expect military forces based on patrimonialism (Institutionalism = Low in Appendix: Table 1) in a tribally dependent monarchy to oppose popular support because their strong ties with the regime. The Tunisian army forces became a highly professional force which did not interfere with political issues and were overall free of corruption (Lutterbeck, 2013). The high degree of institutionalization of the Tunisian army is therefore associated with a certain independency towards Ben Ali’s regime. Interestingly, Tunisia and Egypt offers two examples of military forces supporting the will of its people. With relatively institutionalized military forces, the army chose to disengage with the regime and join the opposition. This is in line with Bellin’s suggestions; namely that highly institutionalized armies are more likely to support popular demand because they are independent to the regime’s interest, professionals and mainly follows national interests.

With a good understanding of the army forces and its power, Ben Ali found necessary to weaken the army’s power to not endure a similar fate than the previous regime (Murphy, 2011). To dissuade the population from contesting his power, Ben Ali based his power on a strong police apparatus, reaching 120 000 policemen in 2010, instead of on a national army. Military expenditures were very low compared to other countries (1.5% of GDP) in the region and the number of officers had not ceased to decrease until the Tunisian army became the smallest army in North Africa. Tunisian troops amounted to only 35 000 soldiers in 2010 and were underequipped (IISS, 2010). Distrust between these two actors reached a peak in April 2002, when “thirteen

Tunisian military officers, including the army chief of staff Brigadier General Abdelaziz Skik, were killed in a helicopter crash” (BBC, 2002). The army never believed this tragedy to be an accident but rather a strategic calculation by Ben Ali’s regime, which was suspicious and threatened by the militaries. Overall, the military forces in Tunisia were highly independent from the regime in power, highly institutionalized and had contentious relationships with the regime. These historical and institutional features of the Tunisian army are the preconditions of the ability of the army to actively support the street protests in 2011 by refusing to open fire on its own people and demanded Ben Ali and his family to leave the country.

I have shown the relative independent and contentious relationships between the army and Ben Ali’s regime over time. However, to explain Ben Ali’s fall, it is necessary to focus on a crucial single point in time, which will be analyzed by a model of game theory, to explain the strategic interactions between the army and the regime.

3.2   The army’s decision to support protesters: a model of Game theory

This paper emphasizes on the role of the army in regime transitions and, more specifically, on the strategic interactions between the Tunisian militaries and Ben  Ali’s regime to explain the fall of Ben Ali in 2011. The methodological goal of this article is to bring a more detailed and explanatory analysis of the interactions between regime elites, using game theory.

3.2.1  Approach and Methodology

The extended model of game theory used in this paper to explain the Tunisian transitions is mainly inspired by Przeworski’s model in Democracy and the Market (Przeworski, 1991) as well as an extension of this model presented by Blaydes and Lo in their article “One man, one vote, one time? A model of democratization in the

Middle East” (Blaydes and Lo, 2012)1.

1    Blayde and Lo (2012) presents their methodology as follow: we extend a canonical model of political transition developed by Adam Przeworski in Democracy and the Market (Przeworski, 1991) to include the possibility of two types of uncertainty. The first – discussed in the original Przeworski conceptualization – is uncertainty on the part of civil society regarding the willingness of regime liberalizers to repress; the second involves the uncertainty of regime liberalizers regarding civil society’s commitment to democracy”.

Przeworski demonstrates that a regime transition is a result of a choices and strategies between political and economic elites in a context of uncertainty in a given society.

Tunisia has seen this context of uncertainty with a period of high unemployment and economic crisis from 2008.  Przeworski claims that a transition may happen when  elite groups have an incentive to deviate from the status quo and impose a regime change, notably because of the non-establishment of strong ties between a regime and key elites in a society. Democracy emerges from a bargaining between elites in a society, and, more precisely, when a strong unity between competing democratic elites is created against the authoritarian regime.

In his model, at the specific moment when an authoritarian regime is considering political liberalization, Przeworski analyzes the choices and strategies of the “proto- liberalizers” in a given authoritarian regime and key actors within the civil society.

Blaydes and Lo extended Przeworski’s model and applied it to political transitions in the Middle East (Blaydes and Lo , 2012). These authors decided to test two assumptions made by Przeworski by focusing on the Middle Eastern democratic transitions. First, they integrated uncertainty or incomplete information, as the civil society may not know to what extent an authoritarian regime prefers repression over democratic transition. Second, they also cast doubt on the assumed commitment of  the civil society to democratic principles in the Middle Eastern countries. Their results suggest that democracy cannot emerge when the regime’s repressive capacity is too low.  Second,  they  emphasize  the  importance  of  uncertainty  and  beliefs which essentially determine the type of regime post-transition2. The model introduced in this paper aims to analyze the strategic interactions between the Tunisian army and Ben Ali’s regime, resulting in a regime transition. Our extended model mainly draws on the previous work of Przeworski and Blaydes and Lo described above.

2  Blaydes and Lo (2012) presented their argument as follow: “The model solution suggests a number of key findings. The first is that the existence of uncertainty is key to the possibility for democratic transition. Second, transition to democracy is only possible when the repressive capacity of a regime exceeds a certain threshold. Given these conditions, democracy occurs when regime liberalizers who prefer democracy to a narrowed dictatorship interact with a civil society that will honor democratic principles”.

This study acknowledges Blaydes and Lo’s contribution to Przeworski’s model of transitions in the sense that uncertainty is a crucial determinant that must be taken into account to study regime transitions. However, while Blaydes and Lo assumed imperfect information of the civil society to explain the resulting type of regime, our analysis will take into account uncertainty and, more precisely, asymmetry of information between Tunisian elites, namely Ben Ali and the militaries. This does not mean that the demonstrators had a complete understanding of the balance of power during the 2011 uprisings. As discussed in the last part of this paper, the Tunisian transition shows the crucial impact of strategic behaviors chosen by the elites to signal to the population the current balance of power, solving the incomplete information of the civil society before Ben Ali’s escape to Saudi Arabia.

The methodology of this paper substantially differs from most studies on the Arab Spring because of the strong focus of our analysis. Most authors have chosen to compare regime transitions in Arab countries and the role of the army in different transition outcomes. There is, by nature, a lack of focus in these analyzes. This results in a misconception of the political economy and inherent structural differences across countries. For these reasons, we focus our analysis only on the Tunisian transition, drawing on both historical intuitionalism and strategic interactions of key actors in Tunisia.

Unlike Przeworski and Blaydes and Lo’s model, one can notice that this study does not seek to predict the types of regimes which could result from the transition. Too many factors are in play and, as the Egyptian case has shown with the overthrown of previous President Morsi by the army in 2013, even short-term prediction in an uncertain context is quasi-impossible.

Considering the Tunisian case only, the main questions this paper tries to answer is: Why did the army refused Ben Ali’s order to open fire on its own population? How did this decision affect key actors of the Tunisian transition? Finally, how to explain that this decision resulted in Ben Ali’s fall?

3.2.2  Model and Equilibrium

This sequential model of game theory analyzes the strategic interactions between two key players of the Tunisian transition; namely Ben Ali’s regime and the army. I will draw on two major popular threats – on the regime to explain the army’s decision to accept  orders  from  the  regime  to  repress  protesters.  We will  then  show  the

implications of the army’s decision on protest outcomes3.

3.2.2.1  Actions and Order of Play

First, Ben Ali’s regime responds to popular protests in 2008 and 2011 by ordering the army to repress demonstrators (Repress?) or to keep the status quo (SQ). In the two situations I analyze, the status quo can be described as high popular mobilizations in the streets. Police forces are trying to maintain order and stop  protests using violent methods. The army is present in key locations but does not intervene directly. Due to the two advanced protests of 2008 and 2011 and the police repression, this model assumes that Ben Ali’s regime is not able to make credible political liberalization reforms4. Second, the army will accept (Accept) or reject Ben Ali’s order to repress demonstrators. Given the contentious relationships with the regime and the army’s relative independence, this decision to follow orders will be made depending on the probability that Ben Ali’s regime survives, thus on the repressive capacity of the regime as well as the strength of the civil society that protests in the streets. 

Finally, Ben Ali’s regime remains in power (Stay) with a probability p if the status quo remains, a probability q if the army accepts to repress protesters and a probability r if the army does not follow Ben Ali’s order to open fire on demonstrators. We identified four potential outcomes in the sequential game presented below (Figure 1).While these scenarios help us to understand the army’s decision to refuse to open fire at its population, this article focuses on the decisive aspect of whether Ben Ali leaves power:

3 Other security chiefs, notably from the police and the president personal security, have played a role in influencing Ben Ali’s decisions. However, we assume here that the army was the decisive actor that made the difference in Ben Ali’s escape.

4 Even though Ben Ali has made enormous concessions the day before his escape, it cannot be said that

the population could take Ben Ali’s potential willingness to reform as a credible commitment. Interestingly, Ben Ali’s concessions seems to share many similarities with Ceausescu’s last speech, in which he made the same types of concessions to his people.

1.      Stronger Dictatorship (STRONG): If the Tunisian army accepts to shoot at its population and Ben Ali stays in power, the bloodbath’s success will certainly reinforce the complicity and the mutual interests between the regime and the army.

2.      Civil (WAR): If the army accepts to open fire and Ben Ali is overthrown in the long-run, the situation is likely to be highly unstable leading to division within the Tunisian society. Although the term of civil war is discussable, we might expect armed conflicts between the pro-regime and the anti-regime close to the Lybian repression of 2011 or the current civil war in Syria. We assume here that the Ben Ali’s departure will inevitably occur in the long-run.

3.      Weakened Dictatorship (WEAK): The scenario in which the army refuses Ben Ali’s order and Ben Ali stays in power will weaken the regime’s power. The desertion of the military’s forces might be seen as a strong signal that the regime’s security forces as well as national elites are divided.

4.      Regime Transition (TRANS): This scenario depicts what actually happened in Tunisia, i.e. a regime transition supported by a coalition between the army and the Tunisian people. If the army refuses Ben Ali’s order and Ben Ali leaves power, a regime transition is expected even though there is a high uncertainty on the type of regime. However, we only focus on the Tunisian regime transition itself, namely whether Ben Ali stays or leave power.

Figure 1. A sequential game between Ben Ali and the Army

3.2.2.2  Payoffs

Ben Ali’s payoff is determined by the probabilities that Ben Ali stays in power (p, q and r) relative to the utility of each actor when he stays in power. p, q and r describe these probabilities which depends on the army’s decision to repress (Repress?) or not (Status Quo).

This model assumes that Ben Ali’s payoffs equal zero, if the status quo prevails. Although this assumption follows conventional practices, the fact that Ben Ali would or should stay passive is discussable in this type of political instability. However, the regime would be willing to keep the status quo because it could reinforce its repressive capacity due to the provision of defense weaponry and training provided by foreign allies. As a result, while the strength of the civil society may grow marginally, it might be all the more true for the repressive capacity of the regime’s security apparatus. Overall, Ben Ali gets a payoff of zero because the balance of power between the regime and protests may not change substantially if the status quo is kept. In line with Przeworski’s liberalization model, the following equation gives a good picture of these variations when the status quo remains:

p = Repressive Force / Civil Society strengths5

The army’s payoff is also based on the probability that Ben Ali remains in power relative to the army’s expected utility when Ben Ali stays in power. However, this utility is conditional on whether military forces accepts or rejects Ben Ali’s order to repress demonstrators. The army’s payoff equals 0 if the status quo remains, which describes the situation in which the army occupies strategic points of the Tunisian territory but do not take direct action either in favour of or against the demonstrators.

3.2.2.3  Actor’s preferences

Ben Ali obviously prefers to stay in power rather than being overthrown. However,  the regime is certainly better-off if the transition is done peacefully than through an armed rebellion or a civil war. Therefore, he will get a payoff of -2 if the outcome is a

5 In his model, Przeworski assumes that protesters start to get organized only after the regime has engaged with political liberalization. It appears not to be the case in the 2011 Arab uprisings, including the Tunisian experience.

civil war (WAR) and -1 if a peaceful transition prevails (TRANS). Intuitively, he will prefer to stay in power conditional on the acceptance of the army to repress the demonstrators (STRONG) rather than suffering from a contentious situation with the military’s force. Therefore, a stronger government will give Ben Ali a payoff of 2 and a payoff of 1 if the game results in a weaken government (WEAK). The status quo, as explained above, equals zero since the spread of demonstrations might be counterbalanced by, for instance, the reception of new weapons and equipment sent Tunisian’s allies or the weakening of the street protesters.

We assume here that the army is indifferent between opening fire on the demonstrators conditional on Ben Ali to stay in power (STRONG) and refusing to open fire conditional on the regime to leave power (TRANS). In both cases, the army will get a payoff of 2 because we assume that their payoff is completely determined  by the identity of the regime post- protests.

We will assume that they are indifferent between the two outcomes in which they make a wrong prediction, namely in the scenario of a civil war (WAR) and of a weakened Government (WEAK), and will get a payoff of -1. While this assumption is indeed discussable, it does not change the final equilibrium and we only use it in our analysis as a matter of simplicity. The matrix presented below (Table 1) describes  Ben Ali and the army’s preferences in the sequential game. The choice of these specific numbers has been decided for simplification, notably to improve graphical clarity. Results do not change substantially when actors’ preferences vary.

3.2.2.4  Possible outcomes of the theoretical model

Actors’ preferences and payoff are conditional on the probability that Ben Ali stays in power. Therefore, this section aims to take into account the probabilities p, q and r to compute the threshold at which Ben Ali is indifferent between keeping the status quo (SQ) and repressing the population (Repress?), as well as the threshold at which the army is indifferent between following and refusing Ben Ali’s order to repress the population. The probability r is greater than q as the regime’s repressive capacity increases when militaries make use of their weapons. In other words, there is a greater chance that the regime stays longer in power if the army accepts to shoot at protesters, even though this doesn’t essentially determines the outcome, namely whether Ben Ali will fall. Thus: q r. The probability p, q and r are given by Ben Ali’s repressive capacity  relative  to  the  strength  of  civil  society  (Blaydes  and  Lo,  2011);  i.e.    p, q and r =

Given the actors’ preferences as well as the order of the p, q and r, we are now able to compute the four possible equilibrium of this game.

First, the equilibrium (SQ; Accept) describes the situation in which Ben Ali chooses  to keep the status quo while the army decides to accept to open fire. This condition is given by the following two inequalities: + (1 – q)

As Ben Ali gets a payoff of zero if the status quo remains (SQ), 2 if he represses the population, the army accepts the order and he remains in power (STRONG) and -2 if the army accepts to repress the population but Ben Ali leaves (WAR), the previous equation becomes:

+ (-1).(1 – q)

In other words, we can say that Ben Ali’s regime will choose the status quo when the probability that he leaves power is equal or lower than a half.

·   

Following the same logic of calculation, we have:

Therefore, the equilibrium (SQ; Accept) equals (                                     .

Second, there is an equilibrium (SQ; Reject) when Ben Ali chooses to keep the status quo while the army decides to reject Ben Ali’s order to shoot at the demonstrators. This situation is given by the following inequalities:

·    + (1 – q)

·   

Following the same logic of calculation as previously, we get:  Therefore, the equilibrium (SQ; Reject) equals (                                                             .

Third, the equilibrium (Repress; Accept) exists when Ben Ali’s regime decides to repress its people and the army accepts this order. The following inequalities define this situation:

·    + (1 – q)

·   

In this situation, we get:

Therefore, the equilibrium (Repress; Accept) equals (                                     .

Fourth, (Repress; Reject) is an equilibrium when Ben Ali’s regime make the order to repress the demonstrators and the army rejects Ben Ali’s order. This situation must satisfy the following inequalities:

·    + (1 – q)

·   

Following the same logic of calculation as previously, we get:  Therefore, the equilibrium (Repress; Reject) equals (                                                             .

Table 2. Equilibria and conditions of four potential decisional outcomes

Equilibria	Conditions
(SQ; Accept)	(
(SQ; Reject)	(
(Repress; Accept)	(
(Repress; Reject)	(

3.2.2.5  Results

Figure 2 describes our results and contains information about the four scenarios discussed above as well as the probabilities of Ben Ali to stay in power. Our results are presented below the straight line r = q, as we assumed previously that r is surely smaller than q, namely that the probability that the regime stays longer in power if the army accepts to shoot at protesters (q) is greater than this same probability conditional on the army to reject the order to open fire (r). We will focus on the area in which q≥, as we want to compare the 2008 and 2010 demonstrations. We can simply notice

here that there is no equilibrium (SQ; Accept). This might look surprising at first glance.

However, it suggests that no situation exists where the army would accept to repress the demonstrators and Ben Ali would prefer the status quo. In other words, whenever the army is ready to accept to open fire on the streets, Ben Ali will prefer to order a repression (Repress; Accept).

Figure 2. Equilibria of the game relative to the probabilities q and r

(Repress; Reject)

Proposition 1: When q is high and not bounded by a low probability r, Ben Ali’s regime will order to repress its population and the army will accept this order.

More  specifically,  when  q  ≥    and  q  , Ben Ali’s regime will choose the equilibrium (Repress; Accept), namely the dashed area in Figure 2.

In other words, when the probability that Ben Ali stays in power conditional on the army accepting to follow orders is high enough and the likelihood that the regime stays in power conditional on the army refusing orders is high enough, Ben Ali decides to order a repression of the population and the army accept this order.

The 2008 Gafsa protest can be represented within this dashed triangle.

Proposition 2: When q is high and bounded by a low probability r, Ben Ali’s regime will order to repress its population but the army will reject this order.

More  specifically,  when  q  ≥    and  q  , Ben Ali’s regime will choose the equilibrium (Repress; Reject), namely the dotted area in Figure 2.

In other words, when the probability that Ben Ali stays in power conditional on the army accepting to follow orders is high enough and the likelihood that the regime

stays in power conditional on the army refusing orders is low enough, Ben Ali decides to order a repression of the population but the army rejects this order.

The uprising of 2010, leading to a regime change, is represented within this triangle.

Due to incomplete information, Ben Ali’s decision to order the shooting is not perfectly based on the army’s expected utility because he doesn’t know at which points the army will refuse to the order to open fire. In other words, Ben Ali has  vague information about the limit at which the army will accept to shoot. For simplification, we assume here that Ben Ali’s decision to order a repression is based on an expectation that the army will follow his order. The army, however, follows its preferred expected utility given that we assume that it has complete information about Ben Ali’s expected utility. Even though the Tunisian army has been seen as a ‘weak’ or even an ‘invisible’ actor in Tunisian affairs, it remains a key actor, notably because of this asymmetry of information that the army can take advantage of. The decision  of the two actors during the two different protests was highly influenced by asymmetry of information giving an advantage to the army. Because Ben Ali didn’t expect the army to consider rejecting the regime’s order, he only took his decision according to the probability that he stays in power when the army intervenes. This decision was likely to be the most effective to stay in power during the 2008 Gafsa protests. In fact, we can see that the army repressed the population because the probability that the regime stays in power, even if the army had refused to open fire,  is too high. Therefore, the regime made the “best” decision to remain in power.

However, the uprising of 2010 demonstrates that asymmetry of information between the regime and the army was crucial to determine a regime transition.

In our case, asymmetry of information allows the army to have a large control of the final equilibria because the Generals have more information than Ben Ali’s regime. During the 2011 uprising, the probability of Ben Ali remaining in power was below the line q = r + 1 because the spread of the population reached a high level that is close to the point where the probability of Ben Ali falling is more likely (towards q = and r = 0). The balance of power between Ben Ali’s repressive force and the strength of the  civil society is such that the army prefers to reject Ben Ali’s order to shoot at the population.

Let’s suppose that we have complete information. Ben Ali would have integrated the probability of the army to reject the order into his calculations. This would have substantially changed our equilibria and thus the outcome of the game.  A smaller  area would have represented the equilibrium representing Ben Ali’s decision to repress his people and the army to reject this order. Ben Ali would have preferred the status quo than to give the order to shoot. Therefore, we can speculate that Ben Ali would have been able to stay longer in power with complete information.

However, this equilibrium has not reflected Ben Ali’s decision, mainly because of asymmetry of information that destabilized the regime. In reality, Ben Ali gave the order to shoot and the army refused because he has incomplete information about the army’s preferences and payoffs. It is thus suggested that Ben Ali would have been better off and stay longer in power by choosing to keep the status quo. Critics of these interpretations may emerge because of the idea that it is easier to draw lessons after that the transitions actually happened. No one was able to predict such a fast regime change because of the illusion of domination that Ben Ali projected on his population and external actors. However, we have attempted to look at the strategic interactions of the elite’s behaviors to explain the Tunisian transition. The next part of this paper will give a detailed analysis of the crucial interactions and strategic behaviors that explain how the Tunisian army’s decision to reject Ben Ali’s order has been crucial to push Ben Ali out of the country.

IV.           Implications

4.1  Despite tensions, why has the Tunisian army never attempted a coup?

Barany Z. (2012) proposed to explain why the army has never taken actions against the regime can explain why the army has never taken substantive actions to overthrown the power in place. Ben Ali’s regime was careful in limiting the army’s power in terms of numbers, budget allocation and the scope of its role and its responsibility. Therefore, the army was largely overshadowed by more influential forces in the political economy of Tunisia. Ben Ali’s government conscientiously  used the police and other security agencies to secure and control its population. Another reason that can reveal the lack of ambition from the militaries to take power is the regime’s decision to send a significant number of officers to the United States in order to attend trainings and programs. The regime strategically made sure that the army stays away from political concerns and be distracted by other concerns than politics (Barany Z. 2012).

The weakness of the army’s political power in Tunisia constitutes an important factor to explain why the militaries have never attempted to overthrown Ben Ali’s regime. It is also argued here that the likelihood of Ben Ali’s regime to stay in power was always too high for the army to attempt any action against the regime. Beatrice Hibou (2006) explains how Ben Ali’s regime was involved in every strata of the Tunisian society and reinforced its power by satisfying key elites and civil society organizations. As the long as this long-term equilibrium sustained, the army had no feasible actions that could be undertaken against the regime even though tensions have always been present.

The 2008 regional protests of Gafsa demonstrated that the army was willing to back the regime and commit crime on its own population. The spread of the demonstrations was, unlike the 2011 uprisings, not as big of a threat for the regime. The army had no choice but to act severely against the Tunisian manners for the political power. Referring to the extended game presented above, we can observe that the probability that Ben Ali will remain in power was too high for the army to take the risk to take  the population’s side. Acting as such would have meant for them to accept retaliations from the regime, which is not in their interests. Timing was also crucial. The expansion of protests to the whole region was quick and Ben Ali’s order to the army came when the balance of power on the field was clearly in the regime’s advantage. One can fairly ask what the army’s decision would have been if the regime had waited a few weeks before ordering a severe repression.

4.2  The importance of uncertainty and beliefs in the Tunisian transition

The army believed that the results of the protests were too uncertain to take a step forward and act against Ben Ali’s regime. This paper argues that, on January 12th 2011, the militaries were better-off to refuse Ben Ali’s order to open fire at the demonstrators and, by doing so, they brought a decisive “coup” to change the regime in place, i.e. its preferred outcome. More precisely, the balance of power on the   field at that time was such that the probability of having a new regime in place was likely, even with the army on the regime’s side. Therefore the army believed that its action to back the population would make the transition almost certain. Also, the risk to shoot  at protesters was too high as the potential new government would have punished them harshly. Thus the army was determinant in ending the long-term equilibrium built by the regime over a quarter of a century.

Considering this paper’s results, Ben Ali’s regime would have been better-off to keep the status quo instead of ordering a repression led by the army. The regime miscalculated its army’s interest because of asymmetry of information. It is  not argued here that Ben Ali would have stayed in power if the status quo had remained. Instead, it is likely that the regime would have stayed much longer in place while the security apparatus, excluding the army, were fighting protesters in the streets.

In addition to that, there is high uncertainty that the army would have backed the population if the status quo had remained. The cost of protesting is increasingly  higher over time for the demonstrators because it is not affordable financially, physically and psychologically. Therefore, the likelihood that Ben Ali stays in power was increasing over time and the army may not have backed the population as it did, because of too an uncertain balance of power on the field.

Therefore, it is demonstrated here that beliefs, asymmetry of information and timing played a central role in the Tunisian regime transition.

4.3  Refusing an order as a strong signal and commitment device

The army’s decision to refuse to shoot at demonstrators triggered strong signals to the population and Ben Ali’s foreign supports. The following arguments help to understand why the Tunisian regime collapsed in barely one month.

First, the army’s decision to reject Ben Ali’s order was seen as a strong signal by the population. It signaled to the population the current weakness of Ben Ali’s power, the division between elites and therefore, that the dictatorship is potentially close to break down. Demonstrations may have shown the true nature of the regime by providing a cascade of information, as described by Lohmann (1994). While demonstrations multiply interactions between Tunisian citizens, this has also been reinforced by the

predominance of social media. The army’s refusal to repress the demonstrators was a key information that spread quickly in Tunisia. Protesters received this signal that ended their illusion concerning the regime’s real power. Demonstrators’ beliefs are likely to have considerably changed after the army took position against Ben Ali’s regime repression.

Second, the army had no choice but to commit to its position of backing the population. After the army took the side of the demonstrators, the army had to take actions to foster Ben Ali’s fall. The army directly suggested Ben Ali to leave the country temporarily until the situation cools down. In the Art of War, Sun Zu mentioned such strategic actions to not “burn bridges” to give the enemies the opportunity to leave the battlefield and thus not commit until the end. The army’s decision to not burn bridges to Ben Ali is crucial to understand the rapidity of the Tunisian transition. In fact, Ben Ali was given an opportunity to leave the country immediately instead of fighting to defend his power. Ben Ali had no choice but to  stay on the Tunisian territory, he would have been committed to fight until the end, similarly to Qaddafi’s bloodbath in Libya. However, it is believed here that this new and immediate opportunity changed Ben Ali’s payoffs to stay in power and resulted in his departure to Saudi Arabia on 14 January 2011.

Third, the army’s decision to support protesters was also a strong signal send to Ben Ali’s foreign allies. Because the threat that Ben Ali will fall became credible, foreign allies, notably western countries, could not been associated with Ben Ali’s  dictatorship and against democratic movements. The provision of weapons by foreign allies became all of sudden very risky for two reasons. First, foreign allies would hurt their reputation by backing a falling and authoritarian regime. Second, a potentially post-Ben Ali era is unpredictable and lethal weapons may end up in the wrong ends, especially in a country located so close to Western Europe. Instead a few Arab countries, especially countries from the Gulf region, saw in Ben Ali’s departure an opportunity to advance their influence and their ideology in the country. Overall, foreign allies would not back Ben Ali’s regime after the army’s desertion. Ben Ali’s power was weak and ending. It became too risky for allies to consistently back the falling regime.

4.4. Implications in the post-regime change period

In the post-Ben Ali’s departure, we are now interested in how Tunisia managed this period. After democratic elections in October 2011, the Islamic Party Al-Nahda was elected to run the country and write a new constitution with the consent of opposition parties. However a political crisis arose when two opposition leaders, Mohamed Brahmi and Chokri Belaid, were assassinated. Massive protests arose once again and political and societal instability prevailed for at least a year (Sprusansky, 2014). The main task of the army forces in the transition period was to fight terrorists in the mountainous region of Tunisia. Military generals did not show any political ambitions to take over the country despite high political instability. The fact that it became an apolitical force without any economic interest in the country can explain his non- inference in public affairs. Despite the relative division of the Tunisian society between pro- and anti-Al Nahda, Tunisian people took advantage of the newly obtained freedom of expression to debate and exchange their ideas and opinions. Although a few incidents arose during the transitions, a civil war was never expected because the armed forces were highly independent from the political arena. Way (2011) didn’t show optimistic views on the Tunisian transition because of the weakness of democratic forces. However, he gave too much importance on the political parties and undermined the role of the Labor Union in Tunisia. In fact, the relative smooth political transition was managed by the Tunisian General  Labor Union (UGTT). This labor organization was used to bring political actors on the

negotiation table in order to end the political crisis and finish the redaction of the new Tunisian constitution, voted with a large majority on January 27th, 2014.

V.   Conclusion

Despite the Tunisian army’s lack of involvement in the political and institutional sphere, this article concludes that it has changed the rules of the game in the political transition in Tunisia.

The recent literature on Tunisia has mostly focused on the role of popular mobilization. It must be emphasized that we do not neglect the role of the Tunisian population. However, we argue that the army’s decision to reject Ben Ali’s order to shoot at the population has been pivotal and has generated very strong signals to key actors that changed their beliefs about the evolution and the outcomes of the popular protests. It is true that some articles have seen that the role of the militaries was  crucial to explain Ben Ali’s departure, but their analysis have been more descriptive than explanatory.

Instead the extended game introduced here enables us to see what type of interactions, strategies and outcomes influenced the army’s decision. Bayles and Lo (2012) rightly thought that uncertainty is key in regime transition and, for this reason, it is taken into account here as well.

More research is needed on business power in Tunisia and how this had influenced  the Tunisian transition. It is true that Trabelsi’s family (the President’s wife family) have multiplied unconventional behaviors before the 2011 uprising, such as imposing the bank or businesses to give (“lend”) them money. However, it would be crucial to know to what extend businesses discontent is related to the army’s decision to reject Ben Ali’s regime order, by researching how business and military elites has interacted pre-transition.

VI. References

Aleya-Sghaier A. (2012). “The Tunisian Revolution: The Revolution of Dignity”,

Journal of  the Middle East and Africa.

Allal Amin and Vincent Geisser (2012). “TUNISIE : “Revolution de Jasmin ou Intifada?”, La Découverte | Mouvements.

Anderson, L. (2011). ”Demystifying the Arab Spring Parsing the Differences Between Tunisia, Egypt, and Libya”, Foreign Affairs. 3 Apr. 2011. Web. 22 Apr. 2014. http://www.foreignaffairs.com/articles/67693/lisa- anderson/demystifying-the-arab-spring.

Badie, B (2011). “Printemps arabe : un commencement”, Etudes, 7 Tome 415, 7-18.

Barany Z. (2012). “Comparing the Arab Revolts, the role of the military”, Journal of Democracy, October 2011, Vol 22, Issue 4, pp 24-35.

BBC News Middle East. “Tunisian army chief dies in air crash”, Retrieved 1 May2002 from http://news.bbc.co.uk/1/hi/world/middle_east/1962415.stm

Bellin, E. (2004). “The Robustness of Authoritarianism in the Middle East: Exceptionalism in Comparative Perspective”, Comparative Politics, Vol. 36, No. 2 (Jan., 2004), pp. 139-157.

Bellin E. (2012). “Reconsidering the Robustness of Authoritarianism in the Middle East, Lessons from the Arab Spring”, Comparative Politics, Vol. 44, No. 2,

(Jan., 2012), pp. 127-149(23).

Blaydes, L., and Lo, J. (2011). “One man, one vote, one time? A model of democratization in the Middle East”, Journal of Theoretical Politics 2012 24: 110 originally published online 14 November 2011.

Campante F. R. and Chor D (2012). “Why was the Arab World Poised for Revolution? Schooling, Economic Opportunities, and theArab Spring”, The Journal of Economic Perspectives.

Cavatorta, F (2012). “Le printemps arabe : le réveil de la société civile. Aperçu general”. Barcelone: Institut Européen de la Méditerranée (IEMed).

Cloutier, J-D. (2012). “Mobilisations populaires et changements politiques en Afrique du Nord: Une analyse du « Printemps arabe » en Égypte et Tunisie”.

Unpublished Ph.D. thesis, University of Ottawa.

Dalacoura K. (2012). “The 2011 uprisings in the Arab Middle East: political change and geopolitical implications”, International Affairs, 88: 1 (2012) 63–79.

El-Khawas M.A. (2012). “Tunisia’s Jasmine Revolution: Causes and Impact”,

Mediterranean Quarterly, Duke University Press.

Gause G. (2011). “Why Middle East Studies Missed the Arab Spring, The Myth of Authoritarian Stability”, Foreign Affairs, 1 July 2011. Web. 22 Apr. 2014.

<http://www.foreignaffairs.com/articles/67932/f-gregory-gause-iii/why- middle-east-studies-missed-the-arab-spring>.

Hess S. (2013). “From the Arab Spring to the Chinese Winter: The institutional sources of authoritarian vulnerability and resilience in Egypt, Tunisia, and China”, International Political Science Review, 0(0) 1–19.

Beatrice Hibou (2006). “The Force of Obedience: The Political Economy of Repression in Tunisia”. Polity. June 2011.

IISS (International Institute for Strategic Studies) (2010). “The Military Balance 2009-2010”. Oxford: Oxford University Press.

Illan R.W. (2012). “Tunisia and the Critical Legal Theory of Dissensus”, Law Critique.

Jdey A. (2012). “A History of Tunisia, January 14, 2011: The End of a Dictator and the Beginning of Democratic Construction”, Boundary 2 Duke University Press Journals

Joffe G. (2011). “The Arab Spring in North Africa: origins and prospects”, The Journal of North African Studies.

Kuhn R. (2012). “On the Role of Human Development in the Arab Spring”,

Population and Development Review.

Leventoglu, Bahar (2013). “Social Mobility, Middle Class, and Political Transitions”,

forthcoming in Journal of Conflict Resolution.

Lohmann, S. (1994). "The Dynamics of Informational Cascades: the Monday Demonstrations in Leipzig, East Germany, 1989-1991." World Politics (1994) 47: 42--101.

Murphy E. (2011). “The Tunisian Uprising and the Precarious Path to Democracy”,

Mediterranean Politics, 16(2), 299-305.

Przeworski, A. (1991). “Democracy and the Market: Political and Economic Reforms in Eastern Europe and Latin America”, Cambridge: Cambridge University Press.

Schraeder P. (2012). “Tunisia’s Jasmine Revolution, International Intervention, and Popular Sovereignty”, The Whitehead Journal of Diplomacy and International Relations.

Schraeder P. and Redissi H (2011). “Ben Ali’s Fall”, Journal of Democracy, 22.3 : 5 19.

Sorenson D. (2012). “Revisiting Transitions in the Arab World, Spring or Fall?”,

ASPJ Africa & Francophonie.

Sprusansky, D. (2014). "Comparing national dialogues in Tunisia, Yemen and Libya",

The Washington report on Middle East affairs (8755-4917), 33 (1), p. 55.

Way, L. (2011). “Comparing the Arab Revolts: The Lessons of 1989”, Journal of Democracy 22 (no 4): 17-27.

VI. Appendix

Appendix 1. Features of Arab Armies6

Type of civil-

Countries

military relations     Institutionalization

Economic

Power

Support to

mobilization

Regime

Change

Tunisia	Autocratic officer- politician regime	High	No	Yes	Yes
Egypt	Autocratic officer- politician regime	Medium	Yes	Yes – passive	Yes
Yemen	Autocratic officer- politician regime	Very low	No	Divided	Yes (relative)
Jordan	Tribally dependent monarchy	Medium	No	No	No
Oman	Tribally dependent monarchy	Low	Yes	No	No
Bahrain	Tribally dependent monarchy	Low	Yes	No	No
Kuweit	Tribally dependent monarchy	Low	Yes	No	No
Morrocco	Tribally dependent monarchy	Low	No	No	No
Algeria	Autocratic officer- politician regime	Low	Yes	No	No
Syria	Autocratic officer- politician regime	Low	Yes	No	No
Lybia	Dual military regime	Very low: fragmented military	Yes	No	No