Turkey held a referendum on April 16 on whether to approve the constitutional reforms pushed by President Erdoğan and his AKP (Justice and Development Party). With 51% of voters saying Yes, the constitutional amendments were accepted. These amendments will not only change the political system from parliamentary to presidential, but they will also finally institutionalize the broader political trend that has been transforming Turkey through the past years, most particularly since 2011. Pursued ambitiously by the ruling AKP, this change involved a dramatic centralization around the head of the executive branch and prioritizing of “majority will” represented by a strong leader, rather than a pluralist, deliberative, and consensual democracy. Thus, to some extent the reforms merely legalized the de facto form of governance, as the government authorities themselves acknowledged.
Nevertheless, the referendum has left the 49 % which said No feeling more marginalized and frustrated than ever before. Thus, the already deep polarization among the supporters and opponents of Erdoğan’s policies which had been tearing Turkish society apart reached a new level after the referendum.
Why is this not another election loss for Erdoğan’s opponents who had been losing since 2002? Why have insecurities and fears among the opposition elevated so much? Answers to these questions require considering why the referendum was held in the first place, what was actually voted, and what was experienced during the elections.
What was actually voted and why?
Since 2012 when the AKP officially proposed transitioning to a presidential system and the debates on the proposal started, the emphasis was not on the advantages and disadvantages of the various types of presidentialism. Both the supporters and opponents of the reform considered it as a move to empower Erdoğan against his rivals which have been throwing major obstacles at him over the last few years, including a corruption investigation, mass protests, and a coup attempt. Thus, before the actual proposal was circulated and examined by experts and citizens, opinions were mostly set. It was clear that the reforms would centralize the political system around Erdoğans’ agency and legalize his un-officially exercised powers over various state institutions. The Yes platform merely suggested that this new system would bring an end to “two-headedness” and thus, allow more efficient and effective governance. The managers of the Yes campaign might not have felt the need to elaborate on the specifics of the amendments as they were probably confident that Erdoğan’s followers will present their support anyway, largely independent of the content of the proposal. They would say Yes because they believed that this would help their leader whose future they saw tied with their own.
The No platform suggested that the reform package presented major risks for Turkey’ prospects for democratization, mainly because it could lead to one-manship. As the Venice Commission also noted the reforms would decrease the checks and balances, weaken the rule of law, and erode pluralist mechanisms in Turkey’s already fragile democracy. In this regard, the most direct consequence of the reforms would be the fusion of the powers of the President and the Prime Minister to produce a more powerful head of the executive branch. It also opened up possibilities for Erdoğan, who has been already in power since 2002, first as a Prime Minister and then as a President, to continue ruling for ten more years, and with possibilities to stay even longer. This prevents the “circulation of elites,” a requisite of democracy, and permits excessive use of incumbent advantages, as acknowledged even by some pro-government intellectuals.
Another critical concern regarding the proposal is that it grants new powers to the President over the judiciary, which is often the opposite of what needs to happen in a transition to presidentialism in order to achieve horizontal accountability. We know that in presidentialism where the executive branch of the state is typically more powerful than in parliamentary systems, a strong judiciary is key to check the government and resolve political controversies. The absence of a strong judiciary in a presidential system, especially in countries with fragile democracies and deep political cleavages can thus create major problems.
The proposed changes also leave the legislative branch with little power to balance the empowered executive branch, while providing the latter with important powers over the former. Accordingly, while the President will be able to use direct legislative powers by issuing decrees with the force of law during a state of emergency (Article 119) it can control the legislative agenda more indirectly thanks to its ability to appoint ministers from the parliament. Meanwhile the parliament will lose its powers on the executive branch, mainly due to its lack of veto power on the appointment of ministers. The proposal also suggests that the parliamentary and presidential elections will be held jointly and can be renewed bilaterally both by the President and the parliament. This creates two types of risks regarding horizontal accountability. Firstly, joint elections decrease the chances for the President and the members of the parliament to come from different parties and balance each other. Secondly, it allows the President to dissolve the parliament by going to elections at a time if s/he feels powerful enough to be reelected. While theoretically the opposite can also occur and the parliament can go to elections too in order to end the term of the President, due to the power asymmetry between the President and the parliament this seems less likely.
In sum, the constitutional reforms propose to shift the balance of power among Turkish state institutions in favor of the executive branch, limiting checks and balances among state institutions. This is concerning since democratic states need self-restrains to limit arbitrary use of power. Elections, assuming they are free and fair, can provide some vertical checks and thus they constitute the basic requirement of democracy. Yet, such checks have their limits. Elections only take place periodically, they tend to underrepresent minority interests, and they do not allow to check every single action of a government. Independent state institutions that can provide horizontal checks on the executive are thus necessary to represent the broader and long-term interests of the society. Hence, the constitutional amendments present challenges for Turkey’s prospects of democratization which need to be taken into account.
Free and fair elections
In addition to the institutional changes that the proposed constitution introduced, the limitations to the freedom and fairness of the referendum campaign and the actual election experience on April 16 also contributed to the fears and insecurities of the opposition.
A main issue that caused discontent among the opposition during the election campaign was the excessive use of incumbent advantage by the Yes platform. These included the use of public funds (such as turning official visits and meetings to campaign events) or media biases. According to the representatives of the CHP (Republican People’s Party), the main opposition party, the No platform had only 45 h of TV time during the campaign period, while the Yes platform enjoyed 485 hours.
The discourse embraced by some government authorities that criminalized No voters further alienated the opposition. These public figures’ emphasis that “terrorists say No” deeply hurt potential No voters and caused a major reaction. Finally Prime Minister Yıldırım felt the need to make a correction and explained that No vote would not make one a terrorist but it was important to note that terrorist groups, such as the PKK and FETÖ said No.
Finally, discontent peaked when the government announced that the Yes campaign won with 51% amid major dispute about election results. Opposition groups claimed the elections were fraudulent and held the Higher Court of Elections responsible for the violations of election procedures. President Erdoğan dismissed such objections immediately during his victory speech that night, suggesting that objections were futile and they had already moved on.
Observers of Turkish politics need to wait for the proposed reforms to be instituted to see their actual consequences. Yet, some of the impacts of the referendum process are visible even now. It seems that the process has strained the already fragile ties that kept the Turkish society together. Studies on Turkey suggest that the society has been dramatically polarized in recent years. OECD statistics demonstrate a major decline in the level of social cohesion in Turkey, which ranks lowest among all members. Akkoyunlu and Öktem state that “existential insecurities” among various social groups increased overwhelmingly. Ağır and Karcı highlight the rising number of Turkish citizens (about 40.000) who immigrated to Germany since 2011. Aktan argues that opposition groups are rapidly losing the sense of “being at home” as they no longer consider Turkey as a secure place for themselves. Certainly, this polarization has much older roots but it had been deepening since 2011. Finally, it seems to have reached a new level in the aftermaths of the referendum. Institutions are known as hard to change once they are established. Thus, the hope for reversing recent political trends which had been keeping the opposition standing on its ground may have declined after the constitutional referendum.
On the other hand, while the main cost of the prevailing political environment falls on the opposition, there are some hidden costs for the supposed winners as well. The chances for democratization are not high when there are such deep grievances and systemic disagreements in a society. In a political environment where democratic institutions are weak, the rule of law is absent, and the fears of the minority are high, even the majority cannot feel secure enough. This is because, in this kind of an environment the cost of losing would be too high and so will be pressure to maintain power. This can easily lead to a vicious cycle where the ruling party tightens its grip on power to avoid the risks of a possible ouster, but this only increases the cost of losing further. As the saying goes, he who rides a tiger is afraid to dismount. Thus, this is an urgent call for Turkish citizens to unite around mutually beneficial goals, which involve promoting tolerance, building trust in democracy, and restore the rule of law.
 Anadolu Ajansı, Başbakan Binali Yıldırım’dan başkanlık sistemi açıklaması, NTV, 13.10.2016, http://www.ntv.com.tr/turkiye/basbakan-binali-yildirimdan-baskanlik-sistemi-aciklamasi,d7hFSC8M3kmyuWtbEhUyQw [10.05.2017].
 According to the constitution proposal if the parliament decided to go to elections during the second term of the President, s/he can be reelected which would extend her/his rule beyond the official two-term period of ten years (Article 116).
 Emre Aköz, Başkanlık Sistemi: Tam olarak kaç yıl?, 10.02.2015, http://www.sabah.com.tr/yazarlar/akoz/2015/02/10/baskanlik-sistemi-tam-olarak-kac-yil [10.05.2017].
 The proposal suggests that the President will be appointing four out of thirteen judges of the High Council of Judges and Prosecutors (HSYK). In addition s/he will retain its previous powers over the judiciary and continue to appoint some of the judges of the Constitutional Court and the Council of State.
 Juan Linz and Alfred Stepan, Problems of Democratic Transition and Consolidation. Southern Europe, South America, and Post-Communist Europe, Baltimore: The Johns Hopkins University Press (1996).
 Government influence over the media does not necessarily take place through direct interventions of the government. The limits of media freedom in Turkey is to an important extent a consequence of the economic relations between the two. For the economic returns of media ownership in Turkey see Ceren Sözeri, tesev.org.tr/wp…/The_Political_Economy_Of_The_Media_A_Sectoral_Analysis.pdf.
 Demokrasi için Birlik Platformu’ndan Hayır toplantısı, “Hürriyet”, 31.03.2017, http://www.hurriyet.com.tr/demokrasi-icin-birlik-platformundan-hayir-topl-40413282 [10.05.2017].
 Main opposition in same boat as terror-supporting HDP: PM Yıldırım, “Hürriyet Daily News”, 07.02.2017, http://www.hurriyetdailynews.com/main-opposition-in-same-boat-as-terror-supporting-hdp-pm-yildirim.aspx?pageID=238&nID=109443&NewsCatID=338 [10.05.2017]; Mark Lowen, Turkey says ‘No’ to saying ‘No’, ahead of its referendum, BBC News, 24.02.2017, http://www.bbc.com/news/world-europe-39064657 [10.05.2017]; Washington Hattı, Turkish PM repeats claim that terror groups support ‘no’ campaign, http://washingtonhatti.com/2017/02/14/turkish-pm-repeats-claim-that-terror-groups-support-no-campaign/ [10.05.2017].
 Erdoğan ‘Hayır’ diyenler teröristtir derken, Binali Yıldırım ‘Hayır diyenler terörist değildir’ dedi, “Siyasi Haber”, 22.03.2017, http://siyasihaber3.org/erdogan-hayir-diyenler-teroristtir-derken-binali-yildirim-hayir-diyenler-terorist-degildir-dedi [10.05.2017].
 Karabekir Akkoyunlu and Kerem Öktem (2016), Existential insecurity and the making of a weak authoritarian regime in Turkey, “Southeast European and Black Sea Studies”, 16:4, 505-527, DOI: 10.1080/14683857.2016.1253225
 Volkan Ağır and İbrahim Karcı, Germany: New Generation Diaspora, Bianet, 19.04.2017, http://bianet.org/english/insan-haklari/185045-germany-new-generation-diaspora?bia_source=rss [10.05.2017].
 İrfan Aktan, Türkiye’den kaçma duygusu, “Gazete Duvar”, 01.05.2017, http://www.gazeteduvar.com.tr/yazarlar/2017/05/01/turkiyeden-kacma-duygusu/ [10.05.2017].
 Barbara Geddes (1999), “Authoritarian Breakdown: Empirical Test of a Game Theoretic Argument,” APSA.