Print
See related documents

Election observation report | Doc. 14327 | 29 May 2017

Observation of the referendum on the constitutional amendments in Turkey (16 April 2017)

Bureau of the Assembly

Rapporteur : Mr Cezar Florin PREDA, Romania, EPP/CD

1. Introduction

1. On 27 January 2017, subject to receiving an invitation, the Bureau of the Assembly decided to observe the referendum on constitutional amendments in Turkey and constituted an ad hoc committee for this purpose composed of 30 members (EPP/CD: 11; SOC: 10, EC: 4, ALDE: 4, UEL: 1 – in accordance with the D’Hondt system) as well as the co-rapporteurs on post-monitoring dialogue. It authorised the President of the Assembly to approve the list of members and appoint the chairperson.
2. At its meetings on 9 and 10 March 2017, the Bureau took note of the invitation to observe the referendum scheduled to take place on 16 April, sent on 23 February by the Deputy Speaker of the Grand National Assembly of Turkey, approved the list of members of the ad hoc committee to observe this referendum and appointed Mr Cezar Florin Preda (Romania, EPP/CD) as its Chairperson.
3. In line with the co-operation agreement signed between the Parliamentary Assembly and the European Commission for Democracy through Law (Venice Commission) on 4 October 2004, a representative of the Venice Commission was invited to join the ad hoc committee as a legal adviser. The invitation was declined by the Venice Commission on 20 March 2017 on the following terms: “As the Venice Commission adopted an opinion which was critical towards the content of the text submitted to the referendum as well as towards the procedure leading to the referendum, it would not be appropriate for the Venice Commission to take part in such a mission.”
4. For the observation of the referendum, the ad hoc committee (whose composition appears in Appendix 1) operated in the framework of an International Referendum Observation Mission (IROM), together with the Limited Referendum Observation Mission (LROM) of the Office for Democratic Institutions and Human Rights of the Organization for Security and Co-operation in Europe (OSCE/ODIHR).
5. The ad hoc committee met in Turkey from 14 to 17 April to observe the referendum on 16 April. The programme of the ad hoc committee’s meetings is set out in Appendix 2.
6. On the day of the referendum, the ad hoc committee split into 11 teams which observed the elections in Ankara and its surrounding areas as well as in the regions and municipalities of Istanbul, Antalya, Izmir and Diyarbakir.
7. The following day, the IROM held a joint press conference and issued a “statement of preliminary findings and conclusions” and a press release (Appendix 3).

2. Political background

8. According to the 1982 Constitution, Turkey is a parliamentary republic with executive power vested in the Council of Ministers headed by the Prime Minister. Legislative power is exercised by the 550-seat Turkish Grand National Assembly (the parliament), which, since the November 2015 elections, comprises four political parties: the Justice and Development Party (AKP) with a majority of 317 seats; the Republican People’s Party (CHP) with 134 seats; the People’s Democratic Party (HDP) with 59 seats; and the Nationalist Movement Party (MHP) with 40 seats. Under the present framework, the President of the Republic is the head of State and holds limited powers.
9. Constitutional reform has influenced the domestic political agenda during the past decade. The most recent constitutional referenda were held in 2007 and 2010. After a multi-party agreement could not be reached between 2012 and 2015, a parliamentary process to amend the Constitution was initiated by the governing AKP and supported by the MHP in December 2016. The manner in which the proposed amendments were passed in parliament, with what many criticised as a limited and not fully transparent debate, whilst a number of deputies from the HDP were in prison, as well as the absence of public consultation at an earlier stage of the process, diminished confidence in the constitutional reform process.
10. On 11 February, the Supreme Board of Elections (SBE) announced that the referendum would be held on 16 April.
11. The constitutional reform package contained 18 amendments which, among other things, proposed changes to the present parliamentary system, to abolish the office of the prime minister and transfer some of the parliament’s key oversight functions to an executive presidency, to increase the number of seats in the parliament to 600 and to empower the president to appoint high-level positions in the judiciary. The Venice Commission stated that the proposed amendments would result in a system where the separation of powers and the independence of judiciary were not assured, thus introducing a “presidential regime which lacks the necessary checks and balances required to safeguard against becoming an authoritarian one”.
12. A failed coup attempt on 15 July 2016 left at least 241 people dead and 2 194 injured. In response to the attempted coup, as well as a wave of terror attacks that shook the country in 2016, the government declared a state of emergency on 21 July 2016 and has extended it three times since (most recently for another three months starting on 19 April 2017). Under emergency decrees, over 100 000 people have been arrested and prosecuted, more than 40 000 remain in detention and 150 000 civil servants have been dismissed. Subsequently, some people were released from prison and reinstalled in their previous positions, and some organisations were allowed to re-open. The state of emergency that restricted fundamental freedoms, as well as the ongoing security operations in south-east Turkey that resulted in several hundred thousand people fleeing their homes, led to questions as to whether conditions were in place to enable a democratic referendum. In its opinion on the amendments to the Constitution, the Venice Commission noted that “the current state of emergency does not provide for the due democratic setting for a constitutional referendum”.

3. Legal framework

13. Amendments to the Constitution require either a two-thirds vote in the parliament to be ratified directly or a three-fifths vote in the parliament combined with the president submitting the amendments to a referendum. A referendum that results in a simple majority of votes in favour of the amendments effects the constitutional change. The 18 proposed amendments affecting 72 articles of the constitution were voted on as a single package. This did not provide voters with the opportunity to make a choice about each of the distinct issues featured in the amendments. There was no question on the ballot; voters were simply asked to vote “yes” or “no”.
14. The legal framework for referenda is inadequate for the conduct of a democratic referendum. It focuses on elections and is limited in regard to the specificities of referenda – in particular, it gives rights to the political parties but does not establish rights and equal opportunities for the camps of proponents and opponents of the proposal. Although aspects were addressed in regulations and instructions, the Supreme Board of Elections (SBE) did not exercise fully its authority to regulate the process to ensure a clear legal framework and declined to provide interpretations on campaign rules and the relocation of polling stations when formally requested to do so by stakeholders. Past Parliamentary Assembly recommendations on the legal framework have not been addressed, including those on suffrage rights, campaign financing, lack of judicial review and rights of observers.
15. The adoption of two emergency decrees that permanently amended election-related laws went beyond the exigencies of the emergency. Notwithstanding the constitutional provision, the SBE decided the changes to the law would take effect immediately. Furthermore, in response to appeals lodged by CHP members of parliament, the Constitutional Court decided that it does not have jurisdiction to consider appeals of emergency decrees, thus effectively barring challenges to the referendum-related decrees. In addition, the parliament did not consider the decrees prior to the referendum or within the 30-day legal deadline, leaving their legal status uncertain and further limiting the opportunity for appeal.
16. The fundamental freedoms of expression, assembly and association, which are already circumscribed by the Constitution and related legislation, were further restricted by the use of extraordinary powers under the state of emergency, hindering the conduct of a democratic referendum. This included provincial governors exercising their authority under the state of emergency to limit freedom of movement, association, assembly and expression.
17. The legal framework does not fully guarantee effective redress for referendum disputes. Decisions of lower electoral boards can be appealed to higher boards, up to the SBE. However, SBE decisions are not subject to judicial review. This leaves the process and results under the final authority of an administrative body, thus challenging the constitutionally guaranteed separation of powers.
18. The SBE received some 45 complaints, which were considered in a timely manner, but the dispute resolution process lacked transparency as hearings were closed and decisions were not published. The SBE satisfied appeals lodged by opposition parties against District Electoral Board (DEB) decisions to relocate polling stations in the south-east for security reasons. While many cases of campaign interference and misuse of administrative resources were noted by the IROM, few complaints were lodged due to diminished confidence in the dispute resolution process. The SBE and the courts did not provide effective redress in such cases, particularly for non-party stakeholders.
19. As the judiciary has primary responsibility for the referendum administration and adjudication of disputes, the recent dismissal of 3 979 judges and prosecutors, which represents almost one third of the judiciary, including five from the judicial oversight body and numerous high court judges, arguably impacted the independence of the judiciary in the referendum period. In April 2017, an additional 45 judicial officials were dismissed, and three judges and a prosecutor were suspended and put under investigation because of a decision to release 21 journalists detained following the coup attempt.

4. Referendum administration and voters lists

20. The referendum was generally well administered by the four levels of electoral bodies: the SBE, 81 provincial election boards (PEBs), 1 080 DEBs, and some 175 000 ballot box committees (BBCs). All legal deadlines were met.
21. The SBE is a permanent body that consists of 11 members elected by and from judges of the Court of Cassation and the Council of State. All four parliamentary parties exercised their right to nominate non-voting members to the SBE. Out of 218 decisions adopted by the SBE, 180 were not published, including the one on the number of ballots printed. The meetings of the SBE and lower election boards were open only to non-voting political party members, which limited transparency.
22. PEBs have three members and are chaired by the most senior judge in the province. DEBs are chaired by a judge and include two civil servants and four representatives of political parties. BBCs are formed for each electoral process and consist of a chair and six members – two civil servants and five representatives of political parties. The law does not provide for equal gender representation in the electoral administration; women chaired 20% of lower-level electoral boards (41% at the district level) and there is only one female member of the SBE. Since the last parliamentary election, eight SBE members were replaced, all chosen by and from newly appointed judges: five due to the expiry of their terms, and three because they were in custody. A series of emergency decrees led to vast replacements at all levels of the referendum administration: 9 PEB chairpersons were dismissed and two others placed in custody, 143 DEB chairpersons were dismissed and 67 others placed in custody. Over 500 electoral board staff at all levels were also placed in custody.
23. The law does not provide for balanced representation of the proponents and opponents of the proposed amendments in the referendum administration. Of the BBC members nominated by political parties, 52% were put forward by parties supporting the “Yes” campaign and 48% by those supporting the “No’”campaign. For the first time, the SBE took a decision that provided guidance for DEBs on applying the “good reputation” requirement for the selection of BBC members. At least 170 BBC chairpersons nominated by the HDP were excluded due to their alleged “bad reputation”.
24. Every citizen who has reached the age of 18 by the day of the referendum has the right to vote. Active conscripts, military students, citizens declared legally incompetent or banned from civil service by a court and those serving prison sentences for intentional crimes cannot vote. The ban on military students and conscripts, and the blanket restrictions of voting rights for the latter three categories are disproportionate. On 15 February, the SBE adopted a decision that partially addressed the rulings of the European Court of Human Rights on suffrage rights and clarified that those with convictions who are not currently in prison are allowed to vote even if their sentence is not fully executed. At least 570 000 citizens were not eligible to vote.
25. Turkey has a passive voter registration system. The voters register is managed by the SBE based on personal data from the civil registry maintained and updated daily by the Ministry of the Interior. Voters were able to verify their entries in the voters lists both in person and through the SBE website, resulting in 467 984 changes. However, no changes to the voters lists were allowed after 10 March, contrary to good practice.
26. Special security zones were in place in parts of six provinces in the south-east, affecting some 670 000 voters. Local authorities in the south-east confirmed that police stationed near polling stations were instructed to check voters’ identification documents to identify those wanted for arrest, and a number of IROM interlocutors raised concerns that this may deter voters from voting. Concerns were also raised about the voter registration of those who had to flee their homes, whose number, according to various sources cited by the United Nations High Commissioner for Human Rights, is between 355 000 and 500 000 people. On referendum day, IROM observers were informed that some of these voters were not able to vote.

5. Referendum campaign, media environment and financial aspects

27. The law does not provide for broad stakeholder participation in the referendum process, as only eligible political parties are entitled to fully participate in the campaign, nominate observers, access the voters register, and enjoy other rights. To participate, a political party must be registered with the Supreme Court Chief Prosecutor’s Office (SCCPO) and have an organisational structure in at least half of the provinces and one third of the districts in those provinces, and have held a party congress not less than six months prior to the referendum. Such eligibility criteria unduly limit political pluralism.
28. Following an SCCPO investigation, 19 political parties that were eligible to compete in the November 2015 elections were found ineligible to participate in the referendum. The SBE approved the participation of 10 out of 92 registered political parties. The People’s Liberation Party and the Liberal Democrats Party lodged complaints with the SBE and SCCPO, respectively, claiming they fulfilled the eligibility criteria, but both claims were rejected. A civil society initiative advocating in favour of the “No” campaign tried to register as a political party to obtain full political participation rights in the process. Having filed its registration documents on 6 February, they remained unregistered.
29. The Law on Basic Provisions does not sufficiently regulate the conduct of referendum campaigns, and stricter campaign rules aimed at ensuring more equitable campaign opportunities apply only during the final seven days. Moreover, the law only protects the rights of and provides opportunities to campaign to eligible political parties. While the broader legal framework for freedom of assembly and expression applies to other stakeholders, including private individuals and civil society, the SBE decided that only eligible parties are entitled to hold campaign meetings and declined to clarify whether others may campaign through other means. Citing the state of emergency or concerns about public security, governors of some provinces either banned or introduced a permission requirement for campaign events organised by actors other than the 10 eligible political parties.
30. The campaign was especially visible in large population centres. Campaign means included posters, banners, billboards and vehicles clad in campaign slogans. Some stakeholders, including civil society organisations, engaged in door-to-door campaigning. Although large-scale rallies and smaller meetings were observed, most campaigners relied on the internet, and especially social media platforms. Voters did not formally receive information from the SBE or other State authorities on the amendments and their potential impact, leaving the eligible political parties to fill this gap, thus negatively affecting voters’ ability to form an informed opinion.
31. The campaign was characterised by the lack of a level playing field. The significantly more visible “Yes” campaign, led by the governing AKP and to some extent the MHP, was supported by several leading national officials, including the Prime Minister and the President, who under the Constitution is required to remain non-partisan and perform his duties without bias, as well as by many lower-level public officials.
32. Cases of misuse of administrative resources were observed countrywide by the OSCE/ODIHR LROM and widely reported in the media. Public ceremonies, such as those opening infrastructure projects, were used for campaigning, with some interlocutors alleging that public sector employees and university students were required to attend. Public transport was regularly free of charge for the day of the event in the cities concerned. Moreover, the President and other officials linked the outcome of the referendum to the government’s support for the regions hosting the events.
33. The “No” campaign was predominantly conducted by the main opposition parties CHP and HDP, with the latter significantly crippled in its ability to campaign given that hundreds of its party members remain behind bars, including its co-chairpersons and 83 HDP mayors. The “No” campaign was also supported by a number of civil society groups, smaller parties and former MHP parliamentarians.
34. Supporters of the “No” campaign faced a number of undue limitations on their freedom to campaign. Many “No” campaigners suffered physical attacks. A high number were arrested, most often on charges of organising unlawful public events or insulting the President. Some “No” campaigners faced difficulties renting premises for events or had their events cancelled by the authorities or venue proprietors, often on short notice. The HDP’s campaign poster and a song in Kurdish were banned by the authorities on the grounds that they violated the principles of the integrity of the State and Turkish as the official language.
35. Campaign financing is insufficiently regulated in the law, which restricts the amount and the nature of donations, but does not limit general party and campaign-related spending. Political parties must report their campaign expenses as part of their annual financial reports to the Constitutional Court, which has responsibility for oversight. Contrary to international commitments and good practice, these reports are not made public and only summarised audit reports are published online.
36. The Constitution provides for the right of freedom of expression but contains undue limitations and permits further restrictions in the Anti-Terrorism Law, the Criminal Code, the Press Law and other legislation. The vague provisions are often used as grounds for the prosecution and imprisonment of journalists. Furthermore, the Criminal Code contains broad defamation provisions, including with regard to the Turkish Nation and State, and provides special protection for public figures, including the President. These provisions limit freedom of expression, which has been further curtailed by the closure of numerous media outlets and the arrest of journalists following the failed coup attempt, as well as by the legal framework for the state of emergency.
37. The media landscape is dominated by outlets that are often owned by business groups that depend on public contracts. Since the July events, a total of 158 media outlets have been closed, including 60 television and radio stations, 19 newspapers, 29 publishing houses and five press agencies, which the Venice Commission described as a “mass liquidation of media outlets”. The majority of the 150 journalists currently in detention were arrested following the attempted coup, and arrests continued during the referendum period. This surge of closures, arrests and prosecutions has resulted in widespread self-censorship.
38. The legal framework did not provide equal access for the “Yes” and “No” sides of the campaign, nor did it guarantee eligible political parties equal access to the media or provide for impartial coverage. The law provides for paid political advertising during the campaign, but the lack of campaign expenditure limits led to the parties having unequal opportunities to reach the voters. The law grants each parliamentary party 20 minutes of free airtime on the public broadcaster, with an additional 10 minutes reserved for the ruling party. The president is also entitled to two 10-minute speeches on the public broadcaster including the last slot for a public appeal, which he officially renounced.
39. The Radio and Television Supreme Council (RTSC) submitted weekly media monitoring reports to the SBE. While the RTSC reports to the SBE included violations detected in their monitoring, no action was taken, as an emergency decree repealed the SBE’s authority to sanction private media if they fail to provide impartial coverage. This decree does not guarantee equal access to the media for political parties and limits voters’ ability to make an informed choice. Several political parties complained to the SBE and RTSC about their access to public and private media.
40. The OSCE/ODIHR LROM media monitoring findings showed that the campaign was visible in all national media. Three of the five monitored television stations, including the public TRT1, favoured the “Yes” campaign. The “Yes” campaign featured prominently in both the public and private media, with 76% of total airtime on television and 77.5% of space in the press, predominantly positive in tone, whereas the “No” campaign received only 23.5% of total airtime and space, mostly neutral in tone. The AKP was also given preferential treatment with 33.5% of total airtime/space, whereas the CHP, MHP and HDP were clearly covered to a lesser extent with 19%, 2.3% and 0.6% of total airtime/space respectively. The AKP received positive coverage on TRT1 and A Haber, and mostly positive on Show TV. The tone of CHP’s coverage was negative on A Haber, partly negative on TRT1, and partly positive on Show TV, CNN Türk, and Fox TV. The President and Prime Minister were overwhelmingly dominant in television coverage with 26% and 18% respectively, whereas opposition leaders were significantly less visible. The AKP took out 63% of all paid advertising time on monitored media outlets. The public broadcaster complied with the requirement to provide free airtime. Coverage of civil society was extremely limited on television. In the press, civil society organisations that supported the “No” campaign received more coverage (3.5%) than those supporting the “Yes” campaign (1.6%).
41. Contrary to Assembly recommendations and the Venice Commission’s Code of Good Practice, the legislation does not provide for international and non-partisan citizen observation. Only the eligible political parties are entitled to nominate observers, and their efforts varied in scope – the AKP and CHP observed widely, whereas the HDP reported difficulties recruiting observers due to a general atmosphere of fear of repercussions.
42. Following the attempted coup, 1 583 civil society organisations were dissolved, including at least three that supported observation efforts during the last elections. Some civil society organisations that were engaged in observation of past elections refrained from observation or significantly limited their efforts due to the overall political and security situation. The SBE rejected accreditation requests from two civil society organisations.

6. Referendum day

43. In the limited number of polling stations visited by international observers, referendum day was generally organised in an efficient manner despite the majority of the BBCs visited not being fully staffed. During opening and voting, some IROM observers were impeded in their observation, access was either not granted or limited and decisions on access were often taken by persons who were not members of the BBC. Otherwise, IROM observers noted that BBC members followed procedures.
44. A few security incidents, affecting a BBC member and several voters, were widely reported and are pending investigation. In the pre-referendum period, local authorities confirmed that police stationed near polling stations would be instructed to check voters’ identification documents to identify those wanted for arrest. Civil society organisations reported three cases where voters were checked before accessing the polls, and it was directly observed by IROM observers in one case. A general police presence outside and inside polling stations was noted in most IROM observations.
45. While a few procedural errors were noted, the counting and tabulation were generally assessed positively by IROM observers. During referendum day, the SBE issued two instructions to consider ballots improperly stamped by the BBC and those without a BBC control stamp as valid, the latter given after the counting of votes in some BBCs had commenced. These instructions undermined an important safeguard and contradicted the law that explicitly states that such ballots should be considered invalid. The SBE was unable to provide the number of ballots affected and stated that since party-nominated BBC members signed the protocols the issue is closed; there is no opportunity for appealing the SBE decision. The HDP has made public claims that they detected discrepancies in 668 protocols.
46. An Assembly team in Ankara felt that many voters were not aware of what they were voting for. One of the teams in Istanbul was not well received by an AKP observer, who did not want them to be there. An Assembly team which visited a polling station set up in a prison in Izmir saw prisoners who were obliged to vote. The team in the Diyarbakir region reported that there was a massive police presence, as well as armed plain cloth individuals and that many people had been detained in the days before the referendum, not in prisons but in other premises, e.g. sports halls, without any official reasons and they were not able to vote. Moreover, that team noted that most of the internally displaced persons did not receive information on the referendum and had no real possibility to vote. It was also informed of pressure on the chairpersons of the BBCs and on civil servants and also that in small rural communities one person voted for everybody and that many people were in pretrial detention and could not vote. The team observed that people with disabilities had no genuine possibility to vote. In Diyarbakir, three people were killed in a polling station and on two occasions the police prevented the Assembly team from observing.
47. At 23:25, the SBE announced that the preliminary results were in favour of “yes” but did not provide any figures. In the media, turnout was reported as 83.7%.
48. The final results were announced by the SBE on 27 April 2017: 51.41% in favour of “yes” and 48.59% in favour of “no”, with a turnout of 85.43%.

7. Conclusions

49. The 16 April constitutional referendum took place on an unlevel playing field and the two sides of the campaign did not have equal opportunities. Voters were not provided with impartial information about key aspects of the reform, and civil society organisations were not able to participate. Under the state of emergency put in place after the July 2016 failed coup attempt, fundamental freedoms essential to a genuinely democratic process were curtailed. The dismissal or detention of thousands of citizens negatively affected the political environment. The “Yes” campaign’s dominance in the coverage and restrictions on the media reduced voters’ access to a plurality of views. While the technical aspects of the referendum were well administered and referendum day took place in an orderly manner, late changes in counting procedures removed an important safeguard and were contested by the opposition.
50. The legal framework is focused on elections and is limited with regard to the specifics of referenda. Although the SBE adopted regulations and instructions to address some aspects of the process, the legal framework remained inadequate for the holding of a genuinely democratic referendum. Fundamental rights and freedoms that are already circumscribed by the Constitution and related legislation were further restricted by extraordinary state of emergency powers, and in particular by decisions by provincial governors to restrict freedom of assembly and expression. Emergency decrees that amended referendum-related legislation exceeded the exigencies of the state of emergency and were not subject to appeal.
51. The 18 proposed amendments affecting 72 articles of the Constitution were voted on as a single package. Voters did not have the opportunity to make a choice about each of the distinct issues featured in the amendments. None of the proposed amendments featured on the ballot; voters were simply asked to vote “yes” or “no”. The State authorities did not ensure that voters were provided with impartial or balanced information on the amendments and their potential impact, thus limiting their ability to make an informed choice. The referendum was generally well administered by four levels of electoral bodies. However, the work of the electoral boards lacked transparency – board sessions were closed to the public and observers, and only a limited number of decisions were published. Following the attempted coup in July 2016, three SBE members and 221 lower-level election board chairpersons, all judges, were replaced following their dismissals. The political party representation on BBCs was not fully balanced and was negatively affected by the rejection of over 170 chairpersons nominated by opposition parties. The law does not guarantee effective redress for electoral board decisions. While the SBE reviewed some 45 complaints in a timely manner, the hearings were closed and decisions were not published. SBE decisions are not subject to judicial review. The continued dismissals and suspensions of judges and prosecutors in the referendum period impact the independence of the judiciary.
52. More than 58 million voters were registered to vote, including over 2.9 million abroad. Voters were able to verify their entries on the voters lists and request changes.
53. The campaign framework was restrictive, and the campaign imbalanced due to the active involvement of the President and several leading national officials, as well as many local public officials, in the “Yes” campaign. The IROM observed the obstruction of efforts of several parties and civil society organisations to support the “No” campaign as well as the misuse of administrative resources. The campaign rhetoric was tarnished by a number of senior officials equating “No” supporters with terrorist sympathisers. In numerous cases, “No” supporters faced police intervention and violent scuffles at their events.
54. The legal framework for the referendum neither sufficiently provided for impartial coverage nor guaranteed eligible political parties equal access to public media. In addition, the law gives preference to the ruling party and the President in the allocation of free airtime, and the SBE’s authority to sanction biased coverage was repealed. Freedom of expression was further curtailed under the state of emergency; the arrest of an unprecedented number of journalists and the surge of media outlet closures led to widespread self-censorship. The “Yes” campaign dominated the media coverage.
55. The law does not provide for international and non-partisan citizen observation, contrary to Assembly recommendations and the Venice Commission’s Code of Good Practice. The efforts of political parties to observe the process varied, and civil society organisations significantly limited their support of observation efforts due to fear of repercussions. Following the attempted coup, 1 583 civil society organisations were dissolved, including some that previously supported observation efforts. A total of 73 international observers were registered to observe the referendum.
56. Referendum day took place in an orderly and efficient manner in the limited number of polling stations visited by international observers. Some IROM observers were impeded in their observation during opening and voting when access was either not granted or limited. Police presence was widely reported both in and outside polling stations and in some cases police were checking voters’ identification documents before granting access to the polls. The SBE issued instructions late in the day that significantly changed the ballot validity criteria, undermining an important safeguard and contradicting the law.
57. It is to be deplored that the President and the Minister for Foreign Affairs of Turkey publicly questioned the integrity and credibility of the observation mission. It is to be recalled that the mission was strictly carried out on the basis of the guidelines for the observation of elections by the Parliamentary Assembly.
58. The Parliamentary Assembly will continue to work alongside the authorities of Turkey in the field of elections and more generally on the reinforcement of democratic institutions.

Appendix 1 – Composition of the ad hoc committee

(open)

Based on the proposals by the political groups of the Assembly, the ad hoc committee was composed as follows:

Chairperson: Cezar Florin Preda (Romania, EPP/CD)

Group of the European People’s Party (EPP/CD)

  • Nicole DURANTON, France
  • Vusal HUSEYNOV, Azerbaijan
  • Duarte MARQUES, Portugal
  • Cezar Florin PREDA, Romania

Socialist Group (SOC)

  • Josette DURRIEU, France
  • Pierre-Alain FRIDEZ, Switzerland
  • Predrag SEKULIĆ, Montenegro
  • Florian KRONBICHLER, Italy
  • Stefan SCHENNACH, Austria
  • Mechthild RAWERT, Germany
  • Alev KORUN, Austria

European Conservatives Group (EC)

  • Nigel EVANS, United Kingdom
  • Jaak MADISON, Estonia
  • Arkadiusz MULARCZYK, Poland

Alliance of Liberals and Democrats for Europe (ALDE)

  • Anne KALMARI, Finland
  • Andrea RIGONI, Italy

Group of the Unified European Left (UEL)

  • Nikolaj VILLUMSEN, Denmark
  • Andrej HUNKO, Germany

Co-rapporteurs of the Monitoring Committee (ex officio)

  • Marianne MIKKO, Estonia
  • Ingebjørg GODSKESEN, Norway

Secretariat

  • Bogdan TORCĂTORIU, Administrator, Election Observation and Interparliamentary Co-operation Division
  • Anne GODFREY, Assistant, Election Observation and Interparliamentary Co-operation Division
  • Nathalie BARGELLINI, Press Officer, Parliamentary Assembly
  • Arman DARBINYAN, Deputy Head of the Safety and Security Department/Head of Field Security Co-ordination, Council of Europe

Appendix 2 – Programme

(open)

Friday 14 April 2017

09:30-10:30 Meeting of the PACE Delegation:

  • Opening by Mr Cezar Florin Preda, Head of the Delegation
  • Briefing by Ms Marianne Mikko and Ms Ingebjørg Godskesen, Co-rapporteurs of the Monitoring Committee
  • Briefing by the Secretariat
  • Briefing by Mr Arman Darbinyan, Deputy Head of the Safety and Security Department / Head of Field Security Co-ordination of the Council of Europe

10:30-13:30 Briefing by the OSCE/ODIHR Limited Referendum Observation Mission:

Opening of the briefing:

  • Mr Cezar Florin Preda, Head of the Delegation of the Parliamentary Assembly of the Council of Europe
  • Ms Tana de Zulueta, Head of the OSCE/ODIHR Limited Referendum Observation Mission

Introduction of individual team members and moderation: Ms Meaghan Fitzgerald, Deputy Head of Mission

  • Political background and campaign: Mr Stefan Szwed, Political Analyst
  • Legal framework and complaints: Ms Marla Morry, Legal Analyst
  • Media: Mr Alain Chabod, Media Analyst
  • Referendum administration: Mr Ivan Tsikota, Election Analyst
  • Referendum day procedures: Mr Ivan Tsikota, Election Analyst
  • Briefing on security aspects: Mr Wayne Pilgrim, Security Expert

14:30-15:30 Meeting with leaders and representatives of political parties in favour of “Yes”

14:30-15:00 Mr Vedat Bilgin, Ankara MP and Head of the Turkish Delegation to the OSCE Parliamentary Assembly

15:30-17:00 Meeting with leaders and representatives of political parties in favour of “No”

15:30-16:00 Mr Erdal Aksünger, Deputy Chair, CHP

16:00-16:30 Ms Fatma Kurtulan, Deputy Co-Chair; Mr Evren Çevik, Foreign Affairs Commission Member; Ms Ceren Bayar, Press Committee Member, HDP

16:30-17:00 Mr Kürşat Ergün, Representative of MHP Opposition

Saturday 15 April 2017

09:30-10:00 Questions to Mr Wayne Pilgrim, OSCE/ODIHR LROM Security Expert

10:00-11:00 Panel discussion with representatives of civil society:

  • Ms Başak Yavçan, Assistant Professor, Ankara Representative, Vote and Beyond
  • Ms Dilek Ertükel, Country Director, NDI
  • Ms Feray Salman, Co-ordinator of the Platform for Human Rights (IHOP)
  • Mr Öztürk Türkdoğan, Chairman of the Human Rights Association (IHD)

11:00-12:00 Panel discussion with representatives of the media:

  • Ms Duygu Güvenç, Diplomatic Correspondent, Cumhuriyet Daily
  • Mr Fatih Şahingöz, Deputy Head of the News Department, TRT
  • Mr Turgut Dedeoğlu, Deputy Chair, Progressive Journalists Association
  • Mr Hüseyin Likoğlu, Ankara Editor, Yeni Şafak Daily

12:00-12:30 Meeting with the OSCE/ODIHR LROM long-term observers deployed in Ankara, Ms Tereza Lewis and Mr Ingo Buettner

12:30-13:00 Meeting with interpreters and drivers

Sunday 16 April 2017

Observation of the referendum

Monday 17 April 2017

08:00-10:30 Meeting of the PACE delegation (debriefing and general discussion)

10:40-12:00 Meeting of the PACE delegation with the OSCE/ODIHR LROM

15:00 Press conference

Appendix 3 – Press release of the International Referendum Observation Mission

(open)

Lack of equal opportunities, one-sided media coverage and limitations on fundamental freedoms created unlevel playing field in Turkey’s constitutional referendum, international observers say

Strasbourg, 17 April 2017 – The 16 April constitutional referendum in Turkey was contested on an unlevel playing field, and the two sides in the campaign did not have equal opportunities, the international observers concluded in a statement released today. While the technical aspects of the process were well administered, voters were not provided with impartial information about key aspects of the reform, and limitations on fundamental freedoms had a negative effect, the statement says.

“On referendum day there were no major problems, except in some regions, however we can only regret the absence of civil society observers in polling stations”, said Cezar Florin Preda, Head of the delegation from the Parliamentary Assembly of the Council of Europe. “In general, the referendum did not live up to Council of Europe standards. The legal framework was inadequate for the holding of a genuinely democratic process.”

“The referendum took place in a political environment in which fundamental freedoms essential to a genuinely democratic process were curtailed under the state of emergency, and the two sides did not have equal opportunities to make their case to the voters”, said Tana de Zulueta, Head of the ODIHR limited election observation mission. “Our monitoring showed the ‘Yes’ campaign dominated the media coverage and this, along with restrictions on the media, the arrests of journalists and the closure of media outlets, reduced voters’ access to a plurality of views.”

Although the Supreme Board of Elections (SBE) adopted regulations and instructions to address some aspects of the process, the legal framework, which is focused on elections, remained inadequate for the holding of a genuinely democratic referendum, the observers said. Provincial governors used state-of-emergency powers to further restrict the freedom of assembly and expression.

“A state of emergency should never be used to undermine the rule of law”, Preda said.

The legal framework for the referendum neither sufficiently provides for impartial coverage nor guarantees eligible political parties equal access to public media, and gives preference to the ruling party and the president in the allocation of free airtime, while the SBE’s authority to sanction for biased coverage was repealed, the statement says.

The law limits full participation in the referendum to eligible political parties and does not regulate the involvement of other stakeholders, the statement says. Further, the SBE decided that civil society organisations and professional associations were not permitted to hold campaign events.

“The campaign framework was restrictive and the campaign imbalanced due to the active involvement of several leading national officials, as well as many local public officials, in the ‘Yes’ campaign”, de Zulueta said. “We observed the misuse of State resources, as well as the obstruction of ‘No’ campaign events. The campaign rhetoric was tarnished by some senior officials equating ‘No’ supporters with terrorist sympathisers, and in numerous cases ‘No’ supporters faced police interventions and violent scuffles at their events.”

Referendum day proceeded in an orderly and efficient manner in the limited number of polling stations visited by international observers. In some cases, access for ODIHR observers during the opening and voting in polling stations was either denied or limited. Police presence was widely reported both in and outside polling station and, in some cases, police were checking voters’ identification documents before granting access to the polls. The SBE issued instructions late in the day that significantly changed the ballot validity criteria, undermining an important safeguard and contradicting the law.