Formation and breakdown of the alliance between state and civic forces in post-2013 Egypt: a state-in-society approach

Purpose – This paper aims to examine why the alliance formed between non-Islamist forces and state actors to oust MohamedMorsi from power in 2013 broke down quickly. Design/methodology/approach – This paper makes use of original data set derived from three waves of surveys fielded in 2011, 2014 and 2015 that ask questions about public threat perception. Around 10 elite interviews were also conducted to further test the study’s hypothesis. Findings – On the one hand, non-Islamists, civic forces challenged the status and interests of state actors in a way that made state actors view them with heightened distrust. On the other, many civic forces, in face of high threat perception, prioritized law during and order after Morsi’s removal, driven – at least partly – by shifts in public attitudes. Originality/value – Many views look at transitions in the Arab world from the angle of how Islamists interact with traditional power holders. Such an approach, however, could be reductionist in many ways because it disregards civic forces. This is a point this paper seeks to address.

. Multiple factors justify this approach. Islamists have traditionally been viewed as the force with the strongest mobilizational capabilities (Munson, 2001), regarded as representative of sizable conservative segments in Arab societies and therefore their integration was long held as a criterion to judge how much change has taken place.
Such an approach, however, could be reductionist in many ways. It fails to encompass the diverse socio-political map of these countries where Islamists are not one brand and where non-state actors are not confined to Islamists. Second, although there will always be disagreements on how far a democratic future in an Arab country would include Islamists (and what type of them and on what conditions), there is hardly any doubt that such a future should include non-Islamist, civic forces (although admittedly, not all no-Islamist forces could be termed as "civic"). International experiences and Egypt-specific factors make a strong case for this (Bernhard, 1993;Munck and Leff, 1999). This is the significance of civic forces. Egypt's post-Morsi developments, however, have shown that the interaction between state apparatus and these non-Islamist forces could also be as problematic as state-Islamists dynamics. This is a point this paper seeks to address. It examines what factors influenced the interaction between state and non-Islamist forces in post-2013 Egypt and why, after Morsi's removal, an alliance between both forces could not last long enough.
The paper adopts a wide definition of civic forces and thereby includes in the analysis groups that are diverse in both their forms and ideologies. With regard to forms, positions and discourse of political parties (e.g. Wafd, free Egyptians, Social Democratic Party, Karama, Ghad, Democratic Front, Popular Socialist Democratic Alliance and others), social movements and protest groups (April 6 t , Socialist Revolutionaries) and even individual figures are examined. When it comes to ideologies, civic forces include a mosaic so diverse to encompass leftists, Nasserists, liberals and centrists. Their common factor is their opposition to Islamist ideologywhich makes categorizing them as "non-Islamists" a more appropriate approach as is further discussed below. Whereas some of them tactically coordinated and even partnered with Islamists at different time periods, they all remainedwith varying degreesstrongly challenging to the Brotherhood ideology and hardly came out as supporters of religiously conservative politics at any time.
The main argument is that the alliance between state and non-Islamist forcesformed several times between 2011 and 2014 but most prominently in June 2013 to remove Morsieventually collapsed [1] because of two factors. On the one hand, non-Islamist forces had gone too far in challenging the status and interests of state actors in a way that made such actors view them with heightened distrust and, eventually, sidelined them when they had the social backing to do so. On the other hand, when some non-Islamist forces prioritized law and order at later stages of the transition out of perceiving threats of lack of security and rising extremism, they lost their agenda-setting power and henceusing the terminology of Migdal's (2001) state-in-society approachlost the fight over "social control" after Morsi's removal [2]. This is a fight they managed to be a powerful player of in most of 2011 and 2012 when they helped define the public agenda and were a force whose actions and reactions mattered. A state-in-society approach is used to focus on how alliances between actors transform quickly in a bid to increase social control (Migdal, 2001).
The paper makes use of original data set derived from three waves of surveys fielded in 2011, 2014 and 2015 that ask questions about public threat perception and support for authoritarianism. Around 13 elite interviews were also conducted to further test the study's hypothesis. Survey and interview data were supplemented by analysis of party literature and statements of non-Islamist parties before and after Morsi's removal. Through the employment of such a multi-method approach [3], the dynamics on both the levels of elites and masses could be captured in a way that provides further evidence to the argument. The Breakdown of the alliance purpose of analysing mass survey data is to show that non-Islamist elites were also driven by shifts in the attitudes of the public when their discourse underwent change post 2013.
The following pages are divided into five sections. Section 1 outlines theoretically how a focus on coalescing actors, their behaviour and the prerequisites of alliance formation during transitions can contribute to our understanding of the alliance breakdown between state and non-Islamist in Egypt in 2013 and thereafter. Sections 2 and 3 analyze how, what was commonly referred to as the June 30th alliance between state and non-Islamist forces gradually crumbled. Section 4 shows how the Egyptian public opinion could no longer provide the social base for non-Islamist forces to prioritize democratic demands over law and order concerns. Section 5 presents the conclusion.

Alliances during transitions: significance and prerequisites
The question this paper seeks to answer is why an alliance between state and non-Islamist actors formed in Egypt in 2013 to oust Mohamed Morsi broke down quickly thereafter. The state-in-society approach provides some interesting insights (Migdal, 2001). In essence, it views societal and political developments as outcomes of constant conflict, struggle and alliances between multiple actorswith state actors among them. It does not regard state actors as omnipotent players but as forces that are in need to coalesce with societal forces at different times. Because these struggles happen almost on a daily basis, their outcomes differ, the alliances involved in them shift constantly and the surrounding structures do not always dictate who wins and who loses. Such perspective captures the dynamism of the struggles during periods of mass mobilization, uprisings and transitional periods as opposed to structural factors which are, by definition, more likely to be static over short-time periods. A state-in-society perspective therefore shows how states and societies, when interacting, transform and constitute one another.
In a sense, one way to look at the turbulent Egyptian scene post January 2011 protests is to view it as a series of formation and collapse of alliances. From a broad coalition between Islamists and civic forces during the January/February 2011 protests itself, to a breakdown of this alliance in favor of rapprochement between Islamists and the Supreme Council of Armed Forces (SCAF) starting from February 2011, to a rift of the Brotherhood-SCAF alignment against the backdrop of the latter's desire to influence the constitution-making process in the summer of 2011 [4], and a subsequent re-alignment between both to push forward parliamentary elections by the end of 2011. All these are just but few examples of the constant conflicts and alliances that governed the Egyptian transition only in 2011. In 2012 and 2013, these alternating alliances have, if anything, increased in the speed by which they emerged and in the diversity of their parties. The focus of this paper is on the June 30th alliance formed between non-Islamist forces and state actors to remove Morsi from power and why it collapsed quickly after Morsi's removal.
In general terms, alliances are quite important at critical junctures of transitional periods. They make it easier to generate concessions and convince actors to give upat least partlysome vested interests. They however entail some pre-requisites. This paper focuses on two of those. The first is that alliances need to be built on mutual trust that each party would have to make some concessions but alsoin returnwould maintain some gains. The proportionate distribution of benefits and costs during such times is quite important in making alliances work and endure (Przeworski, 1991). If the hardliner revolutionaries, however, threaten the vested interests of traditional elites altogether, any gainsif evercould be rolled back (Haggard and Kaufmann, 1992).
The reason is that even at moments where revolutionaries seem too triumphant over old actors, the leverage of the latter actually never get abolished completely. Despite temporal JHASS 3,2 retreat, such actors continue to hold substantial "social control" (Migdal, 2001), and it becomes only a matter of time till they re-group and enter into alliances that could guarantee that they maintain some of their traditional interests. Such a view makes compromise among political actors at the heart of process of change (Huntington, 1991). This paper argues that this is one prerequisite that was missing in the alliances formed between non-Islamist forces and state actors. Its first argument is that the discourse and behaviour of some Egyptian non-Islamist forces during critical periods of the post-2013 period challenged many of the views and interests of state actors. Such a condition made the June 30th alliance a fragile one.
The second factor this paper argues for pertains to the opposite discourse followed at different times by Egyptian non-Islamist forces, especially after Morsi's' removal. Whereasas outlined abovecivic forces are less likely to secure a democratic outcome if they move to completely challenge state forces, they are also likely to generate the same result if they lose focus of their original aim in their struggle to build functioning alliances; that is, if they can no longer set the agenda to ensure that democracy tops public and elite discourse. While the first argument outlines the pitfalls of going too far, the second asserts that it is equally perilous to fall short of what is required to have a democratic outcome. Such a change of priorities by the actors who are supposedly champions of democracy (although admittedly not all of them usually share the same commitment) could happen at times when threat perception is large enough to encourage trading liberties for security and prioritizing law and order over democracy.
Turbulent transitional periods provide the ideal conditions for such a shift in priorities to take place because they usually entail a sharp decrease in the provision of welfare and security as a result of regime breakdown. Perceiving significant threats to one'sand society'sphysical and economic security hinders adequate information processing to reduce the complexity of the situation and could changeone's and society'spriorities to ensure the immediate physical and material well-being over other goals, including protection of civil liberties. Sustained periods of uncertainty and instability have been shown to increase intolerance toward minorities (Stenner, 2005) and generate support for restrictions on civil liberties to maintain security (Huddy et al., 2005). The second argument, therefore, is that if transitional regimes are perceived to be failing at securing the immediate concerns of welfare and security, support for democracy wanes, intolerance increases and preferences for more authoritarian forms of government become more probable. This could happen at the level of elites as well as on the society's level.
Using these perspectives, Sections 2 and 3seek to analyze how the alliance between state and non-Islamist forces was formed in 2013 and then gradually broke down thereafterdiving deeper into actors, their micro-behaviour, their (sometimes very temporary) motives.
2. Alliance between the state and civic forces in 2013an ideal marriage of convenience Immediately after Mubarak's departure from power, and throughout much of 2011, a marriage of interests developed gradually between SCAF and Islamists. Both had an interest in reducing the number of protests and moving toward more structured, institutionalized type of politics for different reasons. For SCAF, it could not run the transitional period while always expecting masses' backlash against its decisions. Islamists, at the time, were viewed by SCAF as the force that could restore some street order. For Islamists, prominent among them the Brotherhood and the Salafis, institutional and formal politics lookedperhaps for the first time in their historyto serve their interests. They Breakdown of the alliance were by far the organizationally strongest, the financially best-funded and the ideologically most coherent political force. Ironically, both actors became to be status-quo actors. Their common foeat least temporarilywere non-Islamist, civic forces who either advocated a more radical change or pushed for a more secular outcome of events. Rational choice calculations (Downs, 1957), therefore, helped bring state and Islamist actors together. Things did turn out as initially imagined by the Brotherhood; their newly founded Freedom and Justice Party did come out as the largest force when parliamentary elections took place in 2011/2012. And when presidential election was held in May/June 2012, they did win the highest post in the country's political structure (Hassan, 2013). By the second half of 2013, however, another twist in Egyptian politics started to take shape. A second, but different, marriage of interests started to surface between state institutions and civic forces. In a remarkable turn of events, state institutions were transformed from a status-quo force to a pro-change force. Their most likely partner was then a mosaic alliance that included any force opposed to the Brotherhood, either ideologically (leftists, liberal, nationalists, old regime and others) or tactically (e.g. Salafists)and hence the term non-Islamist forces. The mass mobilization tool was the Tamarod (Rebellion) Movement which led a public petition campaign to withdraw confidence from Mohamed Morsi on the first anniversary of his inauguration.
The same rational calculations that brought state and Islamists forces together in 2011 and 2012 helped converging the interests and actions of state and non-Islamist forces in 2013. Although the full story has probably not yet been told on whether and how far the Tamarod movement received substantive, tacit or even custodial backing from state institutions, both actors had indeed the incentive to see it grow bigger. Yet again, each party had their own reasons. For state institutions, although Mohamed Morsi did not really move to threaten the institutional or economic interests of the military, a number of moves by Morsi and the Brotherhood were taken to challenge their long-held notion of "national security." Pressures were made to pardon detained Islamists, larger freedom of action was asked for a vast array of Islamist groups, and much closer and warmer relations were made to leaders of Hamasan organization Egypt's state institutions does not regard as an enemy but still has limits on how to deal with it. More importantly, core threats to national security (e.g. the Grand Ethiopian Renaissance Dam) were dealt with in ways that were seen quite unorthodoxto say the leastby state institutions (Hilal et al., 2017).
For non-Islamist forces, the drivers were different, but nevertheless leading to the same outcome; the need to challenge as strongly as possibleand as far as possiblethe authority of Mohamed Morsi. Not long after Morsi's swearing in office has a power-grab by the Brotherhood started to take shape. From appointing predominantly Islamist figures to the one-third appointed Shura Council, to issuing a constitutional declaration that put the President's decisions above judicial review, in addition to an increasingly antagonistic discourse by several Brotherhood figures against non-Islamist journalists, judges and intellectuals; to non-Islamist forces, most indicators pointed at an incumbent regime that was not going to tolerate them. Such power grab efforts by the Brotherhood in early 2013 (termed at the time as "Brotherhoodization" of the state or akhwana al-dawla) could, at least partially, be explained by Powell's (2000) analysis of the behavioral drivers of politicians. According to such drivers, politicianseven in a façadeare rationally motivated by an ever-increasing desire for power accumulation.
Given that civic parties were unable to challenge the electoral machine of the Brotherhood at the ballot box, the only way to challenge the Brotherhood rule was going to be through a non-electoral, single-issue, non-party-based campaign. Admittedly, the non-Islamist camp started to surface as termand acquire an organizational dimensionwith JHASS 3,2 the formation of the National Salvation Front (NSF) in November 2012 as a response to the then Constitutional Declaration by Mohamed Morsi. The NSF continued to challenge Morsi's rule, playing its role in the June 30th protestsalbeit more of a media role rather than actual mobilization. Undeniably, the NSF comprised parties and forces that were certainly not all pro-democracy. Indeed, its coordinator -Amr Mousawas the leader of the Mutammar Party which along with at least two other parties were clearly linked to the Mubarak regime.
For the few months preceding Morsi's removal, both groups of actors (i.e. state actors and non-Islamist forces) gave clear signals to each other that they were in the same camp. The then Minister of Defense, Abdelfattah El-Sisi famously said in April 2013 that "the army never extended its hand to hurt Egyptians and any hand would be cut before it could hurt Egyptians" [5]. On the other side, Mohamed El-Baradei, an icon of the civic camp, said clearly that the "army has a national obligation to save the country from the chaos (it is in)" [6]. This marriage of convenience is the backdrop against which one of the largest mass mobilization efforts of contemporary Egyptian history took place in the period between June 30 and July 3, 2013. When Morsi was removed on July 3, 2013, the rapprochement between state and non-Islamist forces was perhaps unprecedented in Egyptian history.

Alliance breakdown
Although there is an agreement that the alliance between state institutions and civic forces eventually broke down, there are different views on why this happened and what this tells us about the alliance formation in Egyptian politics. While alliances could break down for multiple reasons, this paper argues that the 2013 temporary alliance between state and non-Islamist forces collapsed for three crucial reasons. The first is the significant challenge non-Islamists made to the status, views and interests of state actorsbefore and after 2013. The second decisive factor is that, by prioritizing law and order demands, non-Islamists lost the fight over "social control" in favor of state actors. The third is the many internal divisions that plagued the non-Islamist camp which fragmented their cohesiveness, bargaining power and public image.
3.1 Distrust of state actors in the more radical civic forces Trust is one of the central concepts in social sciences (Baier, 1986). It has been used to explain numerous dependent variables (Inglehart, 1999;Newton, 2001). Among the pioneering studies emphasizing the effect of trust on political dynamics are La Palombara's (1965) study of distrust and hostility in Italy and Rose's (1964) examination of a pervasive sense of trust in England. However, it was probably the rise of the social capital theory that has given a real push to studying the effect of trust (on early writings on social capital see Axelrod, 1984;Putnam, 1993;Taylor, 1987;Trivers, 1971). At its core, social capital theory envisions that dense networks of interactions and associations are likely to generate trust and cooperation among actors (Newton, 2001). Perhaps the most famous of those studies is Putnam's (1993) analysis of the Italian society.
Zooming in on the Egyptian scene between 2011 and 2013, there is evidence that the interaction between state institutions and non-Islamist forces involved significant distrust from the side of the state on particularly three dimensions; what civic forces would do to the 1952 governing formula if they are empowered to govern; civic forces' incompatible view of national security and their attitudes vis-à-vis inclusion of Islamists. The newness of the interactions between these two camps post-2011 might explain why a reservoir of trust levels between them did not exist (Newton, 2001). A detailed examination of each of these Breakdown of the alliance dimensions would outline how the rapid collapse of the alliance between the state and civic forces was the natural outcome. Ever since the late 1950s, Egypt's political system has had key structural pillars. Prominent among them is a strong presidency, a weak legislature and a strong political role for security institutions in domestic and foreign policy (Hilal et al., 2013). The discourse of many non-Islamist forces, however, was an uncompromising exact opposite of these pillars. Although they come from colorful ideological backgrounds, from early on in the transition, most of them shared calls for a more participatory accountable system where representative institutions are further empowered vis-à-vis unelected ones. Many non-Islamist forces have spent most of the years following Mubarak's departure propagating these views, putting them at odds with the state actorseven if temporary political circumstances brought them closer to one another at times.
Dismantling the hyper-presidential regime has been one common factor of many civic party and non-party forces following the 2011 uprising, a position reiterated by many when writing the 2014 constitution. The Wafd, Social Democratic, Tagammu and Nasserist partieswith different degreescalled for a more empowered parliament and a less powerful presidency [7]. Some went as far as demanding the president to be a ceremonial figurehead post [8]. Limiting the power of military and security institutions has also been a demand by many civic forces. Most of the revolutionary movementsprominent among which the April 6 Movementwere strongly opposed to any exception that might allow trying civilians by military courts and opposed the 2014 constitutional referendum on this ground (among other issues). In his vision for Egypt's future, El-Baradei furthermore called for parliament to have the right to review the state budget in all its components, reforming the police through greater civilian control and oversight, a greater role for the civil formal institutions in foreign policy instead of "handing over the important foreign policy issues to security institutions" [9].
Other figures of the civic camp called for the state to completely give up ownership of broadcasting services [10]. Economically liberal civic forces preached for the dismantling of the vast and complex network of economic entities belonging to the state [11]. Privatization does not only decrease state control over the production and pricing of certain goods but also more importantly deprive the state from having hundreds of senior positions to award.
Another thorny issue generating huge distrust from state actors in non-Islamist forces is the latter's attitudes towards the inclusion of Islamists. Indeed, the interaction between state actors and non-Islamist forces in Egypt is constantly influenced by how these forces interact with Islamists. This is one other area where deep mistrust paralyzes such interaction. This mistrust was cemented by events before, during, and after the 2011 uprising, perhaps only seceded in the short moment of the June 30th protests but was revived shortly thereafter.
Close interactions between Islamists and non-Islamist forces were quite noticeable before Mubarak's departure. According to a Wafd Party member: [. . .] before 2011, all liberal and leftist political forces sympathized with the Muslim Brotherhood. Everyone looked at the Brotherhood as a group making sacrifices all along [. . .] Wafd has supported the Brotherhood since 1984. before Mubarak's ouster, Wafd collaborated with the Brotherhood in the (2011) protests. We came together to get rid of the regime. We also met the SCAF together (in 2011) [12]. This Islamist/civic interaction also included leftist forces. According to a member of the Hamdeen Sabahi's 2012 and 2014 Presidential Campaign: JHASS 3,2 [. . .] there was cooperation between al-Karama party and the Brotherhood before 2011. We sympathized with the group as they were constantly oppressed. Even more, many members of al-Karama Party defended the Brotherhood. The same can be said for al-Ghad Party [13].
Given this historical trajectory, the worry of a possible rapprochement that could happen any time between Islamists and non-Islamist forces is deep-rooted in state actors' minds. In fact, at times of elections it has been a constant challenge for the state to prevent non-Islamist forceseven those who are very close to its ideologyfrom forming coalitions with Islamists. Even before the 2011 uprising, when the National Democratic Party (NDP) was still the governing party, the main concern of its leaders at election times was to prevent hidden alliances between some of its candidates and Brotherhood candidates (Rabie, 2006).
When they do, non-Islamist forces usually collaborate with Islamists on pragmatic, utilitarian grounds. Out of a belief that the Islamists' mobilizational machine is the most efficient, non-Islamists usually seek to trade off some concessions in return for electoral backing of the Islamists' voting bloc. This has been the case under NDP. A similar motivation has also been the backdrop of the Fairmont declaration in 2012 in which secular figures agreed to support Morsi in his run-off presidential election battle against Ahmed Shafik. The memory of this declaration is unlikely to fade away from the minds of state institutions. It represents a constant reminder that non-Islamist forces are ready at certain times to coalesce with Islamists. The outcome is a conviction by state institutions that non-Islamists cannot really be trusted to be strong enough rivals of Islamists.
Despite the cooperation between both campsstate actors and non-Islamistsin removing Morsi, this distrust did not wither away. Several non-Islamist forces made frequent statements about the need to include the Brotherhood again in political life after Morsi's removal. El-Baradei came out when he was still Vice President to say that he "was ready to give the Brotherhood a rope as citizens" and that he deals with them "not as a political faction but as part of the Egyptian society that has political, economic and social rights" [14]. This is a view that was made publicly and privately by other non-Islamist forces. According to Khaled Dawood of The Constitution Party: [. . .] even if some people and some liberal or leftist political parties do not like the Political Islam Movement, we need to deal with them. These movements present a political reality, and we need to respect that [15]. These statementsand other similar onesare what made a senior former government figure talk of a constant concern on behalf of the state that if non-Islamist forces had more space, they wouldeven unintentionallyhand over the country to the Brotherhood in a "turnkey" style [16].
A third issue contributing to the culture of distrust by state actors in non-Islamist forces is the latter's view of national security, seen by the state as incompatible to their own perceptions of the concept. There is a historical sense of skepticism among state actors on how any civilian force or actor would act on national security threats, if empowered to govern, either totally or partially. As per the literature on civil-military relations, a minimum level of trust by military institutions that civilians would act "rationally" on issues of national security is a prerequisite for the military to voluntarily accept civilian control of its affairs (Huntington, 1957). This is one other background why state actors often times viewed non-Islamist forces with amplified suspicion.
Much of the discourse of the non-Islamist elite has encouraged state actors to view them as immature in foreign relations to the level that their assumption of power could become menacing. In the 2011/2012 parliamentary election, many of the leading non-Islamist parties voiced real opposition to key pillars of Egypt's foreign policy such as its close relationship Breakdown of the alliance with the United States and peace with Israel. Al-Karama Party, for examplewhich was then led by Hamdeen Sabahy, one of the leaders of the June 30th Alliance later on and a challenger to the then candidate Abdelfattah Al-Sisi in the 2014 presidential electioncalled in 2011/2012 to "end (Egypt's) obligations according to the Peace Treaty and its relevant annexes, [. . .] to expel American forces and early warning systems out of Sinai and abolish the un-and-demilitarized zones" [17]. The Popular Socialist Alliance Party refused to recognize Israel and called for the "opposition of all forms of normalization with the Zionist entity all international agreements that violate national sovereignty, including the Camp David Accord." [18] The Social Democratic Party refused the "unjust terms of trade and economic cooperation" Egypt is involved in and the "submission to the decisions of the international institutions." [19] Certainly, part of the unorthodox views of national security issues by non-Islamist forces is not of their own making but instead a natural consequence of never being ever part of formal power.

Prioritizing law and order
In the aftermath of Morsi's removal, many non-Islamist forces, however, choseeither tactically or out of convictionto re-position themselves. They refrained from challenging outright many of the positions of state actors but either significantly delayed their opposition or completely changed their attitudes to align themselves with those of the state's. These actors mainly include formal political parties that agreed to continue to take part in electoral politics after Morsi's removal. Wafd, the free Egyptians, Tagammu, Nasserist Party and the Social Democratic Party are the prime actors of this bloc. For many of them, by accepting to prioritize law and order, they lost significant leverage in the fight over social control in post-2013 Egypt. Many of the developments post 2013 testify to this. The Protest Lawwhich outlawed protests that do not get formal government approvalwas a watershed on this trajectory. The law was drafted and discussed during the second half of 2013 and eventually promulgated in November of the same year. It was the response of state institutions to the continuous and repetitive waves of protests that captured Egypt's public space since the 2011 uprising. In 2013, the pro-state justification of the law was that it mainly targeted the Brotherhood protests. However, the law marked a major turning point in the elite and public debate on the balance between law and order on the one hand and political freedoms on the other. Several actors of the non-Islamist camp either supported the law or voiced partial objections to some of its articles but did not oppose it entirely. Almost all such statements emphasized that the Law was in response to security threats. The Nasserist Party, for example, said that the government was forced to issue the Protest Law because of the "unpeaceful" nature of protests that seek to "sabotage public properties and terrify people." The spokesman of the NSF said that the "political forces needed to understand that such a lawdespite any partial objection to its contenthas been issued to protect the people" and that "it should be marketed in a way that stresses how it would protect public and private properties" [20]. Although a number of the revolutionary movements did protest against the law (April 6 Movement and the Socialist Revolutionaries), it was eventually approved and what was viewed as a sacred cow in post-2011 Egypt at the timethe right to protest without any prior formal approvalwas no longer so.
Another significant turn was the 2014 constitution. Although, from a theoretical perspective, a progressive constitution when it comes to the many political, religious and civil liberties it includes, it was another acceptance by non-Islamist forces of an order that tips the balance of power in favor of state actors through its preservation of a strong president running a semi-presidential system. According to the 2014 constitution, Egypt JHASS 3,2 continued to havewith minor modificationsa semi-presidential system à la France's Fifth Republic. According to article 137, the president has the right to dissolve parliament. Although this move has to be approved by the electorate in a referendum or otherwise annulled, there is no political cost for the president if he or she loses such a referendum.
Almost all non-Islamist parties supported the constitution, either zealously or out of a desire to move forward. The common understanding was that the draft constitution needed to be passed and with a majority that exceeds the majority that supported the 2012 passed during Morsi's rule (which eventually happened). Even some of the most controversial articles were eventually supported, or at least not strongly opposed, by several figures of the non-Islamist camp. Sameh Ashour, the head of the Bar Association and a member of the June 30th alliance, for example, was not opposed to trying civilians in front of military courts as an exceptional measure given Egypt's "special circumstances" but only voiced his opposition to its inclusion in the constitution [21]. The spokesman of the Popular Current refused the public scrutiny of the armed forces' budget because it is a "national security issue" whose details cannot be disclosed [22].
In interviews conducted at the time, many of the justifications for this shift in priorities stemmed from a perception of threat among leading non-Islamist actors. The fact that ISIS was making significant territorial gains in Syria and Iraq at the time, with the unprecedent wave of terrorist attacks that hit Egypt in 2013-2015 created a perception that Islamists had to be confronted. As per Abdel Ghafar Shokr, chairman of the Socialist Popular Alliance Party, the Brotherhood had to be confronted because they created a system of Wilayat al-Faqih or rule by the clergy [23]. According to Khaled Dawood, spokesman of the liberal Constitution Party, "the Muslim Brotherhood initiated a civil war by stating that other political movements are Kuffar [infidels]" [24]. Farid Zahran of the Egyptian Social Democratic Party also indicated that the "MB wanted to change the identity of the state, with Shari'a on top of the state" [25].
Fear of rise of extremism therefore topped the agenda and facilitated the prioritization of law and order over calls for democracy. Against such a backdrop, it was no surprise that a democratic pact as the ones in Southern and Eastern Europe, did not come out of the Egyptian experiment. According to O'Donnell et al. (1986, p. 72) "democracy is produced by stalemate and dissensus rather than by prior unity and consensus". Roeder (1998, p. 209) has put forward a similar point after examining post-communist transitions, saying that "the more heterogeneous in objectives and the more evenly balanced in relative leverage are the participants in the bargaining process of constitutional design, the more likely is the outcome to be a democratic constitution". The homogeneity of discourse and imbalance of power that happened in Egypt between state and non-Islamist actors after 2013 ran counter to these conditions [26].

Internal divisions within the non-Islamist camp
One additional factor for the alliance breakdown has been the internal divisions with the non-Islamists camp itself. Such divisions were clearly observable on multiple occasions. Perhaps one of the most visible incidents was the different responses to the clearing of the pro-Morsi camps in August 2013. On the one hand, the move was supported by a number of non-Islamists forces. Hamdeen Sabahy, a former presidential candidate at the time, one of the leaders of the NSF and long-time Nasserist leader, said that it has become clear that clearing the pro-Morsi sit-ins in Rabaa and Nahda has become a public demand because of what they represent, in their current status, in terms of threat to the nation's security and safety. On the other hand, Mohamed El-Baradei, the then Vice President, reportedly objected the clearance in the final hours before it took place' and resigned shortly thereafter [27].

Breakdown of the alliance
Although El-Baradei never actually proved to be a figure with significant mass following, he represented an icon of the non-Islamist camp and a character that gives cues for the Egyptian educated elite. Similar and stronger divisions were also reported vis-à-vis the Protests Law with Sabahi this time opposing it and even pledging to annul it as part of his 2014 presidential campaign manifesto whereas other forces embraced it as mentioned above [28]. Furthermore, on whether to integrate Islamist in the political life in the post-Morsi scene, divisions were also visible. According to a senior official of the Egyptian Social Democratic Party, interviewed in August 2015, non-Islamist parties were divided into two camps. The first camp viewed Islamists as "traitors that cannot be trusted", as "enemies of the state that have to be crushed". The second "progressive secular camp" saw that "Egypt cannot move on without a political solution" [29]. On the other hand, opposing views are adopted by other parties; according to a Wafd Party senior officer, "the MB are not legitimate and do not act like this" [30].

Social control won over by state playerssurvey evidence
Social and political forces are unlikely to act in complete isolation from the social milieu in which they exist. Certainly, a sizable part of the support and mobilization for the 2011 uprising was the existence of a genuine demand for change. Fast forward to 2013-2014 to apply the same logic, there is ample evidence that the ground has shifted by then and that demands for political freedoms have seceded in favor of demands for stability, law and order, and prioritization of bread and butter issues.
Methodologically, it is difficult to infer whether this has been because of the misrule of the Brotherhood, the mistakes of the non-Islamist forces, or the effective discourse and behaviour of state actors. The safest answer of course is that all factors played a role. What is more certain however is the outcome; change was no longer the salient issue, but rather stability; political freedoms were no longer the priority but law and order. This is hardly surprising for a people traditionally accustomed to a certain level of security and certainty both of which were threatened in the period between 2011 and 2013 and perhaps even shattered by the terrorist attacks that intensified since 2013. One way thus to look at why non-Islamist forces started to decrease their attention to issues of political freedoms is that they wereat least in partreacting to a popular mood that was shifting and easy-to-read. To test this, one needs to examine how far there was a genuine change in the public support for imposing restrictions on democratic rule driven by perception of increased lack of security and fear of rising extremism.
To test the argument on the level of the masses, original survey data that come from nationally representative surveys carried out in Egypt over three waves in 2011, 2014 and 2015, are used. The significance of these surveys is that they capture potential shifts in public attitudes at critical junctures in the country's stumbling transitional path. Whereas the first survey was face-to-face, the two surveys in 2014 and 2015 were phone surveys [31] ( Table 1). All three surveys however had similar questions tracing support for imposing some restrictions on democratic rule. The face-to-face survey comprised a questionnaire of roughly 35 min with a nationally representative sample of 2,000 individuals drawn from the electorate. The sampling procedure was a combination of multistage cluster random sampling and systematic random sampling. The phone surveys were much shorter (average of 10 min). Questions on support for imposing restrictions on democratic rule were included in all waves and questions tracing perception of the biggest problems facing the country over multiple waves in 2011, 2014 and 2015 (Hassan et al., 2018). In all surveys, the sample was determined using census data and targeted only those citizens aged 18 or above. The results of the different waves show that when it comes to the "biggest problems facing the country", a usual indicator of what constitutes major threats, "poverty" went down from being by far the biggest problem in 2011 to be replaced by "security and crime" which moved to be the biggest problem in 2014, a trend that continued in 2015 (Figure 1). Religious extremism became the second biggest problem in 2015, almost on a par with security and crime.
A stronger picture is painted by Figure 2 which shows the change between 2011 and 2015 in support for military intervention in politics. The relevant question asked respondents to express which one of the two following statements comes closest to their views: "The military should withdraw entirely from political life for good", or "The military should continue to intervene when it thinks necessary". Although such support started unusually high in the first year after Mubarak's departure (61.9% supported such intervention with varying degrees, because of the SCAF' support of the protesters), it increased even further to 79.4% in 2014 and 89.8% in 2015.
One interesting question is whether support for a military's role in politics is driven by perceptions of threat to security and rising extremism. To test this hypothesis, two measures of threat are used; views of crime and lack of security as the biggest problem and views of rising extremism as the biggest problem. To examine how far welcoming the military's intervention in politics is driven by these two threat perceptions, two ordinal logistic regressions were conducted to trace the effect immediately after the uprising (2011) and after Morsi's removal (2014 and with support for military intervention in politics as the dependent variable. The results are shown in Table 2. The model controls for age, education, income and religion. The three waves taken together show no effect. Separating the data into before and after Morsi's removal however (which corresponds to before and after the June 30th alliance) produces interesting results in support of the paper's hypothesis. Whereas perceiving security and crime as well as religious extremism as, the biggest threats facing the country, significantly decreased support for military intervention in politics in 2011, both cause the exact opposite effect in the aftermath of Morsi's removal. Both effects are significant at the 99% confidence level. Education appears to act in the traditionally hypothesized direction in all models, decreasing support for restricting democracy. Males also seem to be less supportive of military intervention than women. Time was also statistically positively significant in the full sample, indicating that attitudes to restrict democracy increased over time as we moved from 2011 to 2015. These results provide evidence that perceptions of threats pertaining to security/crime and a fear of an Islamist takeover did affect popularand not just eliteattitudes.  Notes: Standard errors in parentheses; ***p < 0.01, **p < 0.05, *p < 0.1 JHASS 3,2