Convergence through learning? Patterns of exclusion during the Arab uprisings

By Maria Josua, German Institute of Global and Area Studies, Hamburg

*This memo was prepared for the workshop, “Transnational Diffusion, Cooperation and Learning in the Middle East and North Africa,” held June 8-9, 2016.

During the Arab uprisings, the mass demonstrations that took place in many countries triggered different reactions by incumbent elites towards protesters. Regime reactions ranged from concessions and reform promises to defamation, negative framing and repression. The initial benevolence towards demands deemed legitimate, such as socioeconomic grievances or government reshuffles, subsided over time and was replaced by “exclusive-repressive policies” (Heydemann 2015) in the countries that slipped into protracted violence. However, in resilient, so-called “moderate,” autocracies such as Morocco, Jordan and Algeria, a negative attitude towards protest movements and the transnational ties between them also gained traction. Was this convergence towards the increasing delegitimation of transnationalism generally and of protesters more specifically due to learning by state actors in these countries? Learning is one crucial mechanism leading to convergence (Gilardi 2014), which can originate either in historical experience or the observation of external actors.

I look at various facets of exclusion, understood as “the neglect or rejection of the demands of certain groups and/or disregard for or discrimination against their identity” (Josua 2016a: 7). In the context of the Arab uprisings, the specific focus lies on the delegitimation and exclusion of protesters and different societal groups. From the multitude of mechanisms, I focus on three examples: negative and/or ethnical framing of protesters; banning dual nationals from public office; and physical exclusion. What these strategies have in common is that they seek to confine transnational ties and stop diffusion effects of activism from states undergoing profound change. As the exact mechanisms leading to convergence are difficult to trace, I outline some common strategies that deserve further study, building on evidence from recent literature on authoritarian learning (Heydemann and Leenders 2014, Bank and Edel 2015). The cases considered here include countries with a high degree of volatility, though most are resilient autocracies. This uncovers broader patterns that currently shape Arab politics and that might continue to do so over the next years.

Framing Protesters vs. The Good Citizen

Participants in mass protests of the Arab uprisings were targeted by a variety of exclusionary discourses seeking to discredit them. In most Arab countries, the dominance of state media and some degree of self-censorship helped enhance the frequency of such frames. Their aim was to deter risk-averse citizens who were unsure whether to join the protests.

Already in early 2011, beginning with Ben Ali in Tunisia, officials framed protesters as foreign infiltrators, trouble-makers, vandals or criminals in their discourse. An even blunter frame depicted all protesters as Islamists or even jihadists. Even in cases where protest mobilization was low, such as Jordan and Algeria, elites apparently felt the need to resort to these kinds of frames to delegitimize the protesters’ demands and to justify the security forces’ repressive actions.

Dehumanizing framing of protestors was employed infamously in Libya where it backfired and to a lesser degree in Syria (Heydemann and Leenders 2014: 82). Assad only once referred to conspiracies as germs then refrained from using this genocidal terminology (ibid.). Branding protesters as terrorists became the default strategy in Syria “to legitimate its use of force, demonize its opponents, and communicate to the West that it and the Assad regime shared a common foe” (ibid.). The anti-terrorism discourse neatly tied in with similar approaches in Western countries, where it symbolizes the ultimate justification for all measures sold as necessary, no matter who the group labeled “terrorist” actually is.

Related to the discourse about foreign influence, conspiracy theories loomed large, e.g. Yemen and Syria, where they were first brought up very prominently in Assad’s March 2011 speech. Allegations centered around Western democracy promotion schemes that were pictured as attacks on Arab states. Beyond the West, other enemies were suspected of meddling with domestic politics. Algeria probably even tried to create its own foreign conspiracy. An anonymous call for an uprising on 17 September 2011 was set up on Facebook, which appeared to have been faked. Also because it had no relation to domestic movements and political protests had long subsided, nobody followed it. The Minister of Interior was eager to blame “foreign parties related to the Zionist entity” in the newspaper Ennahar as the ultimate delegitimation.

A subtler, but more disturbing method of deterring the general population from taking to the streets was to frame protesters as belonging to ethnic or religious minorities. The most notorious case is Bahrain (Shiites), but it also happened in countries with moderate protests, such as Jordan (Palestinians), Morocco (converted Christians, sympathizers of the Polisario movement) and Algeria (Kabylians, Christians) (Desrues 2013: 417-418, Josua 2016a). Some of these ascriptions were ad hoc statements by individual officials in order to cater to resentments and prejudices by the “mainstream” population. In some cases, they have become entrenched, leading to societal divisions and suspicion of others. This strategy of othering not only brought up the delicate question of identity, but also exposed the minorities themselves to open challenges by their compatriots. The side effect of such labeling was to unsettle group members.

In a similar vein, government officials gradually employed and promoted a more nationalist discourse to deny the protesters’ loyalty to the nation. This was a counter-reaction to the protesters’ use of national symbols such as flags and their insistence that the demonstrations were directed against the regime, not against the state. Strategies like sectarian framing of protests as discord with strong religious undertones (in Arabic, fitna) were employed in Bahrain, Saudi Arabia and Syria (Heydemann and Leenders 2014: 87f.), but this wording was also used in Algeria and Jordan.

Codified Nationalism

Xenophobic sentiment also translated into legal policies. Legislators in several states enacted laws that prevent dual citizens from assuming high public offices. This insulation was supposed to demonstrate that transnational ties and “foreign” influence were undesired. Egypt had had such a stipulation for a longer time, but Jordan followed suit after the protests in fall 2011. The new regulation came into force retroactively, which meant that sitting members of parliament and ministers were forced to either revoke their citizenship of another state (such as the United States, Canada or Syria) or resign from their post (The Jordan Times, 13 October 2011). On the one hand, this measure reflected a general growing suspicion of foreigners and “foreign agents.” As shown above, this suspicion was partly evoked by state discourses. On the other hand, the new law posed “significant challenges for political inclusion for some of the most educated and well-trained Jordanians” (Tobin 2012: 97), not least those with a Palestinian background, who tend to hold dual citizenship more often than East Bank Jordanians do. Syria adopted a similar regulation in its 2012 constitution, Iraq attempted to codify a constitutional provision barring dual citizenship in 2009 and 2013, and most recently Algeria amended its constitution in 2016 with the same clause.

However, Jordan reversed its law in May 2016, as Egypt planned to do in 2015. Official newspapers have not given any particular reason for this change. One possible interpretation is that the signaling function of keeping non-exclusive nationals away from spoils was more important than implementing the actual policy. However, later the negative effect of excluding potential office holders who were competent or important for co-optation might have prevailed in the decision for revoking the regulation. In this case, the adoption of the law could result from transnational learning, while its abolishment would be due to historical (domestic) learning.

Denial of Public Space

Another widespread strategy was the physical exclusion of protesters by denying them public space. Authoritarian regimes in general have an ambivalent attitude towards public gatherings unless they are in the rulers’ support. In the Arab uprisings, the appropriation of public space by protesters was the crucial point of empowerment. The ouster of Mubarak on February 11, 2011 was certainly a strong motivation for implementing harsh policing measures in the Algiers demonstrations that took place on the following day. On both days, all traffic to Algiers was halted, security forces checked cars and busses, preventing demonstrators from reaching the central meeting point at the Place du 1er Mai. Public servants and students were advised not to pursue their work in the capital in order to keep the streets empty. At the protest site, people were split into small groups to avoid larger mobilization of the population, so that just more than 1,000 protesters were met by 30,000 security forces. Analysts of Algerian politics cited previous domestic experiences with riot control as crucial for containing the protests. Although the massive presence of police has been a characteristic of Algeria’s recent past, it would be plausible to attribute the high level of precaution at least partly to learning from the negative example of Egypt, probably combined with historical learning. Without the dangerous Egyptian precedent, the Algerian leadership might have been content with deploying less forces to counter the demonstration.

Following a similar logic of not allowing successful examples from abroad to gain hold domestically, in Bahrain the Pearl Roundabout’s importance was recognized as a symbolic and infrastructural center for the protest movement. Also in this case, a recognition of the crucial role that Tahrir Square in Cairo played probably led to authorities “learning from the losers.” Security forces therefore violently cleared Pearl Roundabout on the third day of its occupation in mid-February and for a second time in mid-March. Then they even destroyed the pearl monument in order to eliminate the protest movement’s main symbol (Bank & Edel 2015: 15), going far beyond challenging the protesters’ appropriation of public space.

Convergence Through Learning?

While the outcome of convergence towards exclusionary politics can be safely stated, tracing processes of learning is methodologically challenging. Learning in political science is understood as “a change of beliefs (or the degree of confidence in one’s beliefs) or the development of new beliefs, skills, or procedures as a result of the observation and interpretation of experience” (Levy 1994: 283). In order to show that learning actually took place, first the “change in individual beliefs” and second policy change as a direct result of the changed beliefs should be observed (Levy 1994: 291). However, learning does not always lead to policy change, but is likely to be successful when it reifies existing beliefs, thus impeding policy change (id.: 290). It is important to bear in mind that learning does not have to mean innovation. Learning might also be the confirmation of working strategies from an established repertoire under new circumstances.

The chronological sequence of events is vital for identifying actual learners. Another desideratum is identifying the sources of learning. One central finding in Bank and Edel’s study is that the models of learning are contingent upon “proximity – either in terms of geographical closeness or in terms of political similarity” (2015: 21). However, the threshold for establishing that learning has taken place is high because the cognitive processes among elites are a black box to outsiders who can only observe the policy outcomes. From a distance the evolution of individual decision-makers’ calculations is difficult to conjecture. As in-depth interviews with decision-makers are not available for these examples, in the following I sketch how learning as the mechanism at work could be studied, avoiding the trap of false positives judging only from the common outcome of converging strategies.

In the cases of physical exclusion, a thorough study would aim to empirically show whether learning from Egypt’s failure was a central part of the rationale behind massive protest proofing in Algeria and the destruction of the Pearl monument in Bahrain. Also counterfactual reasoning would help to strengthen an argument about the regional influence that played into the calculus of decision-makers.

Regarding the legal provisions that exclude dual nationality holders from access to office, interviews with parliamentarians might provide insight into where the initiatives came from and whether the laws in other countries set the example. Even without direct evidence of authoritarian learning, the convergence on the legal level is striking. It is safe to say that the legal modifications were a response to a diffusing perception, namely the frame that foreigners are disloyal and the nation should be shielded against all kinds of external influence, even by dual nationals. A contrasting study of the cases where these laws have been subsequently abolished would be insightful for showing the locus of historical learning.

The common pattern of othering dissenters is also noteworthy, though direct learning is most difficult to trace here. The important result from the framing examples is that local conditions shaped the specific forms that these transnational patterns assumed to adapt them to the domestic audiences, as can be seen in the specific minority framing used in different countries. Nonetheless, the tendency of these discourses again converged around exclusion.



What is striking about the exclusionary mechanisms described above is that they were present in countries belonging to different post-uprising trajectories, irrespective of the exact course of events. The patterns thus seem to reflect a general thrust towards more exclusion. This development also transcends the monarchy-republic gap. The variety and breadth of exclusionary strategies points at their relevance for autocracies in general. However, using them can backfire as they ultimately have far-reaching repercussions and develop self-fulfilling dynamics reinforcing prejudices. The normative dimension of such a shift of repertoires has worrying consequences, almost inevitably leading to an erosion of societal cohesion.

Finally, how distinctive are the phenomena described above? Such patterns of exclusion are not unique to the Arab world, as Lisel Hintz (2016) showed in her study of Turkey. But even going beyond the Middle East, there is nothing specifically regional about the strategies. Nationalist discourses and the repression of dissent below the threshold of violent force are trademarks of authoritarian regimes worldwide. Nonetheless, exclusionary policies have been specifically justified by weathering the crisis of the Arab uprisings. In this sense, a historical moment offered the chance for implementing the described policies. After providing a glimpse into selected mechanisms of exclusion, the ways in which such repressive actions are justified to domestic and international audiences deserves to be studied in more detail.

Bank, André and Edel, Mirjam 2015, ‘Authoritarian Regime Learning: Comparative Insights from the Arab Uprisings’, GIGA Working Paper 274, June, Hamburg: GIGA.

Desrues, Thierry 2013: ‘Mobilizations in a Hybrid Regime: The 20th February Movement and the Moroccan Regime’, in: Current Sociology 61 (4), 409–423.

Gilardi, Fabrizio 2014: ‘Methods for the Analysis of Policy Interdependence’, in: Engeli, Isabelle and Christine Rothmayr (eds.), Comparative Policy Studies. Conceptual and Methodological Challenges. Palgrave Macmillan: Houndsmill, 185–204.

Heydemann, Steven 2015, ‘Mass Politics and the Future of Authoritarian Governance in the Arab World’ in POMEPS Studies 11, The Arab Thermidor: The Resurgence of the Security State, 27 February (Washington, D.C.), 14–18.

Heydemann, Steven and Leenders, Reinoud 2014: ‘Authoritarian Learning and Counterrevolution’, in: Lynch, Marc (ed.) The Arab Uprisings Explained. New Contentious Politics in the Middle East. New York: Columbia University Press, 75–92.

Hintz, Lisel 2016: ‘Adding Insult to Injury: Vilification as Counter-Mobilization in Turkey’s Gezi Protests,’ Memo prepared for the POMEPS workshop, “From Mobilization to Counter-Revolution: The Arab Spring in Comparative Perspective,” held on May 3-4, 2016. Online:

Josua, Maria 2016a: ‘If You Can’t Include Them, Exclude Them: Countering the Arab Uprisings in Algeria and Jordan’, GIGA Working Paper 286, May, Hamburg: GIGA.

Josua, Maria 2016b: ‘Co-optation Reconsidered: Authoritarian Regime Legitimation in the Jordanian “Arab Spring”’, in: Middle East Law and Governance, DOI: 10.1163/18763375-00801001.

Levy, Jack 1994: ‘Learning and Foreign Policy: Sweeping a Conceptual Minefield.’, in: Inter-national Organization 48 (2), 279–312.

Tobin, Sarah A. 2012: ‘ Jordan’s Arab Spring: The Middle Class and Anti‐Revolution’, in: Middle East Policy 19 (1), 96–109.