By: Everett Price, Senior Policy Advisor
With Contributions from Scott Rauland, Senior State Department Advisor

On June 24, Turkey held its first presidential and parliamentary elections since the passage of controversial constitutional amendments last year that began Turkey’s transformation from a parliamentary to a presidential system. The victors in this election are to preside over the transition to this new form of government and begin to shape the operation of its revamped institutions.

In accordance with its commitments as a participating State of the Organization for Security and Cooperation in Europe (OSCE), Turkey formally invited the OSCE to observe the vote. This invitation paved the way for the OSCE’s first-ever full-scale deployment of election observers to Turkey. Although the OSCE observed previous elections in Turkey—including last year’s constitutional referendum—it had never done so with a full complement of hundreds of short-term observers that deploy all over the country to record their observations on election day. In the absence of short-term observers, OSCE observation missions rely primarily on a smaller cohort of long-term observers who spend as much as a month in the country monitoring every dimension of the campaign period and balloting. (Learn more about OSCE election observation.)

Altogether, the OSCE’s Office for Democratic Institutions and Human Rights (ODIHR) deployed 22 long-term observers and more than 300 short-term observers to observe the election across the country. Most STOs are drawn from cadres of experienced volunteers offered by individual OSCE participating States. In addition, the parliamentary assemblies of the OSCE (OSCE PA) and Council of Europe (PACE) contributed 72 and 32 members of parliament and parliamentary staff, respectively, to serve as STOs.

The U.S. Helsinki Commission regularly participates in OSCE PA election observation missions. What follows is a first-person account from two U.S. Helsinki Commission staff who served as short-term observers during the Turkish elections. These observations are not an authoritative account of the conduct of the Turkish election, however. Readers interested in such an account should review the OSCE’s official statement of preliminary findings and conclusions.

In the days before the election, experts from the OSCE’s ODIHR and the OSCE PA organize a series of in-depth briefings in Ankara to acquaint short-term observers with the context and process for the coming vote. In opening these briefings, Mr. Ignacio Sanchez Amor, a Spanish parliamentarian tasked as the special coordinator and leader of the OSCE short-term observer mission, noted numerous ways in which this election was exceptional.

Turks would be voting under a nearly two-year-old state of emergency imposed by the government following a failed coup attempt in July 2016. The state of emergency gave Turkish President Erdogan sweeping powers to rule by decree and authorized provincial governors to curtail basic freedoms, such as the freedom of movement and freedom of assembly. Presidential decrees purged tens of thousands of civil servants from their work, shuttered over a hundred news outlets, blocked thousands of websites, and contributed to the arrest of scores of independent journalists, often on dubious national security charges.

Sanchez Amor further commented that the transition to a presidential system and the country’s newly-approved election laws made this election especially complex. Most of the constitutional amendments approved in last year’s referendum would take effect after the election. The victorious presidential candidate, for instance, would be the first to assume unprecedented executive powers that international monitors and the chairman of the U.S. Helsinki Commission, Sen. Roger Wicker (MS), criticized as undermining the separation of powers. Likewise, successful parliamentary candidates would take up seats in a somewhat neutered institution that is given no say in ministerial appointments and can be unilaterally dissolved by the president.

One of the constitutional amendments abolished a previous prohibition on new electoral laws taking effect less than 12 months before an election. This meant that the June election would be governed by election regulations passed in November and March that President Erdogan and his ruling Justice and Development Party (AKP) had muscled through parliament without any opposition support. Opposition leaders sharply criticized provisions in the laws that allowed the government to relocate voting locations on security grounds, loosened rules governing the police presence around polling stations, and weakened protections against election fraud by admitting ballots that are missing a required polling station stamp.

Sanchez Amor also expressed concern that one of the country’s major presidential candidates had been in pre-trial detention since November 2016 and was being forced to campaign from his jail cell. This treatment of Selahattin Demirtas, the presidential candidate for the pro-Kurdish Peoples’ Democratic Party (HDP), was one of many restrictions severely disadvantaging the HDP.

Campaign banners of two opposition parties stretched across a primary avenue in downtown Ankara.

The briefings we received from journalists, civil society organizations, and political parties largely focused on concerns that President Erdogan and the AKP enjoyed overwhelming and unfair advantages incompatible with a free and fair democratic process.

Independent journalists noted that Turkey is the world’s largest jailer of reporters with approximately 150 behind bars. They further remarked on dramatic changes in Turkey’s media landscape in recent years that had seen nine out of 10 mainstream TV channels, and most of the print media, fall under the ownership of government aligned businessmen. As a result, the ruling party benefited from nearly wall-to-wall positive media coverage. Additionally, under recent legal changes the Supreme Election Board was stripped of its ability to impose penalties on broadcasters for violating regulations that mandate equal election coverage.

All opposition parties complained about Turkey’s 10 percent election threshold—the highest in the world—that requires a party to garner 10 percent of the national vote to secure seats in parliament. During this election, they feared that the ruling party would manipulate the vote to deprive the pro-Kudish HDP of a ten percent share, allowing AKP as the likely runner-up in Kurdish-dominated areas to assume the seats forfeited by HDP. This would be the AKP’s quickest route to a commanding majority in the legislature.   

Opposition party leaders warned that the government could use a combination of tactics to suppress the vote for HDP, particularly in the Kurdish southeast. The government had already announced that it had invoked its new authorities to relocate and merge nearly a thousand polling stations in the southeast for security reasons, affecting more than 150,000 voters. Election authorities could also use the admission of unstamped ballots to artificially diminish HDP’s share of the vote. In addition, several briefers noted that deadly violence could be used to intimidate voters. Indeed, less than two weeks before the election a campaign-related altercation in the south left three HDP supporters dead, along with the brother of an AKP candidate for parliament. Many briefers noted that the pro-government media had cast the AKP as the victim of the melee, where in reality the fight had been instigated and escalated by the ruling party’s side.

The opposition’s indictment of the fairness of the election was met by an AKP-led campaign to smear those who might tarnish perceptions of the credibility of the outcome. The OSCE observation mission was one of the targets of this campaign. The government denied two OSCE PA parliamentarians entry to the country to participate in the election observation mission, citing political opinions they had expressed in the past. A government spokesperson decried the OSCE’s interim report on the election on June 15 as “political.” Additionally, just two days before the election, Turkey’s semi-official news agency published a story citing anonymous security sources that claimed OSCE observers planned to create “chaos” in the country.

It was in this climate of heightened mutual suspicion between the ruling party and the opposition, and between the government and the OSCE observation mission, that we deployed to Istanbul to undertake our election day observation.

We were assigned to observe the election at precincts in and around Sisli, a mixed-income neighborhood in central Istanbul that historically supports the secular opposition. Each precinct in Turkey contains numerous voting rooms, with a maximum of 400 voters assigned to each. In all we visited nine so-called ballot box committees (BBCs) in five precincts throughout the day.

Sisli delivered its largest share of votes (48.7 percent) to Turkey’s leading secular opposition party CHP in the last parliamentary election in November 2015, and overwhelmingly opposed the transformational 2017 constitutional amendments (71.8 percent). Given the district’s political profile, it was unsurprising to find observers from CHP and other secular opposition parties deployed in full force at our first precinct where we observed the opening procedures for the polls.

We arrived just before 7:00 a.m. as the ballot box committee (BBC) was assembling to open the sealed election materials and prepare for voting to begin. The seven-person BBC, chaired by a civil servant and composed of bureaucrats and political party representatives, began to count and record the number of ballot envelopes and presidential and parliamentary ballots—a tally that is important for later confirming that no election materials are unaccounted for. 

The mood was serious but amiable and cooperative. With a solitary exception, BBC members worked together constructively without so much as a hint of their diverging political loyalties. Soon, an NGO observer (who was accredited as a political party observer, since there is no legal framework for NGO observation) appeared with a tray of traditional Turkish tea for everyone, observers and BBC members alike. (Tea stands were ubiquitous at the entrance to polling stations, fueling weary poll workers throughout the day and contributing to a generally festive atmosphere around the otherwise dreary school buildings.) 

Voters congregate outside a polling station in central Istanbul on election day.

Over steaming cups of black tea, the poll workers set about the onerous task of applying the BBC’s identifying ink stamp to each of the more than 300 envelopes and presidential and parliamentary ballots—nearly 1,000 stamps in all. Every BBC we visited that morning noted that they had labored well after the polls opened to complete this cumbersome but mandatory and important process.

With two of their colleagues still engrossed in stamping and the eight o’clock opening just minutes away, the BBC’s other five members forged ahead with preparations, sealing the clear plastic ballot box with a wax stamp. More or less promptly at eight, the BBC chairwoman announced the opening of the polling station. 

An elderly gentleman who had already shuffled through the open doorway before the announcement immediately presented his ID to receive his ballot papers. His punctual appearance quickly revealed the disorganization of this particular BBC, which had failed to organize the somewhat convoluted voting procedure into an orderly workflow. 

Voters were to hand over their identification for confirmation against the printed voter rolls and then receive two oversized ballots for president and parliament, one undersized envelope, and a stamp to mark their choices. To cast their ballots, voters entered a curtained booth, marked one choice each for parliament and president, stuffed the large sheets into the small envelope, sealed it with a lick, and emerged to drop the envelope into the ballot box. Before departing, voters returned the stamp back to the BBC, signed the voter roll, and retrieved their identification and any bags or cellphones they left behind with the BBC.

Keeping track of identification cards, remembering to provide all four necessary voting materials (two ballots, envelope, and stamp), and managing the coming and going of voters proved difficult for our first BBC. In just the first twenty minutes the chairwoman twice pulled open a voting booth’s privacy curtain to locate a misplaced ID and missing stamp—an act that should rarely—if ever—occur, since it can compromise the secrecy of the vote. 

This procedural dysfunction may have slowed the vote and caused undue confusion for voters, but it had negligible if any implications for the outcome of the balloting. Indeed, the majority of other BBCs we visited were capably managed and all demonstrated admirable transparency. In each classroom we visited—and they were all classrooms—the BBC chair graciously welcomed us, answered our questions, and allowed us to review the voting materials.

Political party and NGO observers were common and none complained of being restricted in their work on election day. By the end of the day we had grown particularly accustomed to the sight of observers from the HDP party. These observers were almost uniformly impressive, assertive, young, and female. It became clear that what is generally described simply as a “pro-Kurdish party” has developed political purchase far beyond the Kurdish-dominated southeast, attracting many young, progressive Turks concerned with the rights of women and minorities. 

Presidential and parliamentary ballots and envelopes prepared for distribution to voters.

In our experience, the transparency protected by professional BBCs and capable local observers was only undermined by skittish security services who seemed uneasy about the role of international observers. Under new election laws passed earlier this year, Turkish police were allowed to patrol closer to polling stations and—for the first time—enter voting rooms at the request of any citizen. One instance of police involvement we witnessed was legal and appropriate; in several other cases it appeared to overstep the prescribed bounds.

At a polling station we visited in Gultepe, a more conservative neighborhood just outside Sisli, a heated argument erupted over a poll worker who allegedly exceeded his mandate in assisting a confused voter, sparking allegations of election interference. Consistent with their mandate, the police entered on at least three occasions during the prolonged shouting match to respond to the disturbance and to remove unauthorized people who had entered the voting room. These same police entered another time to exercise their prerogative to check our credentials but departed shortly after.

As the day wore on, however, our interactions with the police grew more frequent and contentious. At another polling station in Sisli, police greeted us almost immediately upon our arrival and insisted on escorting us throughout the building. When we entered a voting room to conduct our observations, the police followed us in without any discernible invitation and sat down to watch us until we were through. 

Arriving at our final polling station of the day, we were stopped at the entrance to have our credentials checked against a screenshot of approved individuals the policeman had received via the encrypted messaging application, WhatsApp. He informed us that several foreigners had been caught “posing as OSCE observers” so they were under orders to apply extra scrutiny. Although we were not on his screenshot, the officer relented after a few minutes’ delay and followed us inside while respecting the rules about entering the voting room.

After a short break, we returned to the same polling station to witness the closing and counting procedure, but this time the police refused us entry. They said they had still not been able to find us on their list, despite our accreditation by the Supreme Electoral Board. With the five o’clock closing swiftly approaching, we insisted on the importance of entering before the polls closed. They offered that we could observe the voting room from the hallway, but we were obligated to decline since the OSCE’s methodology requires unfettered access to the polling area. At the last minute, they said we could observe from a designated area inside the room. Once inside the room, it turned out no such area existed and the police displayed no interest in enforcing one. In contravention of the rules, however, they remained standing directly behind us inside the room nearly the entire time.

Consistent with OSCE observations across the country, vote counting at our BBC began promptly once the polls closed. In much the same sprit of cooperation we witnessed at the opening in another precinct, the members of this BBC worked smoothly together to perform the critical, final procedures: securing the voting materials and counting and recording the results. 

The BBC’s genuine effort to conduct this process fairly and transparently was marred by some critical procedural errors and the persistent presence of the police, which risked undue oversight by the security services of a sensitive political process. Most procedural faults took place early on and introduced avoidable opportunities for mistakes or manipulation. 

Rather than count the unused voting materials after the polls closed, for instance, the BBC departed from the prescribed procedure and counted them before the room was open to the public to observe the count. By depriving observers of the opportunity to verify this tally, the BBC undermined a safeguard that confirms the number of votes cast matches exactly the number of voters who participated. 

Another significant oversight involved the BBC’s failure to enter crucial figures directly into the official register, known as a “protocol.” By having one member of the committee simply jot down the tallies of voting materials on a scrap piece of paper, the BBC failed to guard against subsequent mistakes in transcription or intentional alterations. 

Oddly, the otherwise attentive and assertive political party observers in the room did not raise these issues with the BBC, possibly out of ignorance of the procedures or disinterest in the importance of these steps. They seemed most focused when it came to the centerpiece of the process: the all-important counting and adjudication of ballots.  In this, the BBC acquitted itself quite well—holding up each ballot in full view of all present, loudly announcing the vote, and recording it only once all were satisfied with the chairman’s judgment (i.e. valid, invalid, or blank). 

U.S. Helsinki Commission Senior State Department Advisor Scott Rauland reviews voting materials with Ballot Box Committee members.

Given the considerable pre-election controversy about the admission of unstamped ballots, it was surprising that no observer raised a question about whether the ballots or envelopes were appropriately imprinted with the BBC’s seal, which was often faint and on the reverse side of the papers. Late on the day of Turkey’s controversial 2017 constitutional referendum, the government unilaterally decided to count unstamped ballots despite the widespread understanding that the stamps protected against fraud. The number of admitted unstamped ballots last year allegedly accounted for the government’s slim margin of victory in that vote. As a result, opposition leaders protested earlier this year when the government used its absolute majority in the parliament to codify the validity of unstamped votes beginning with the 2018 presidential and parliamentary election. This decision created frustrating ambiguity about the need for the elaborate stamping process that tied up BBCs in the morning, sometimes for more than an hour. 

The last steps of the vote count turned out to be the most cumbersome. The astonishingly analog voting process created numerous frustrations, significantly delaying delivery of the ballots to the District Electoral Board responsible for tabulating all the votes in Sisli before forwarding them to the Provincial Electoral Board that oversees a third of Istanbul. The chairperson was consumed for almost an hour manually copying detailed voting results onto nearly a dozen copies of the official protocol for distribution to political party representatives and observers. Another time-consuming process involved sealing all the ballots and sensitive voting materials in a cloth sack using twine and a wax seal. All present watched in quiet agony as the chairperson struggled to melt the nub of wax with a lighter, singing his fingers and nearly setting fire to the bag in the process.

Out of the 250 votes counted in our BBC, leading opposition presidential candidate Muharram Ince prevailed with 65 percent of the vote and his party, CHP, took 50 percent of the parliamentary ballots. The simultaneous presidential and parliamentary election afforded voters the opportunity to split their votes between the two ballots. Specifically, many analysts speculated that opposition supporters would endorse Ince as the favored presidential candidate while casting a vote for HDP in the parliamentary election to help the party clear the ten percent threshold. The outcome in our BBC seemed to bear this theory out: Ince received 15 percent more support for president than his party did in the parliamentary vote, while HDP’s presidential candidate Demirtas secured only 4 percent in the presidential but his party garnered 24 percent in the parliamentary.

Once counting was complete in all the precincts’ voting rooms, members of the BBCs boarded a municipal vehicle with the sealed sacks and official protocols for delivery to the District Electoral Board. Per OSCE instructions, we jumped into a separate vehicle to tail the municipal van through the narrow streets of Istanbul to the DEB to confirm the official results were delivered directly without interference. 

A long line of vans packed with other BBCs was in front of the District Electoral Board waiting their turn to offload. When it came our turn I—accompanied by a police escort—followed the voting materials past heavily-armed guards and crowd control fencing into the building. It was a cramped but sprawling high-rise divided into a warren of small, austere rooms. A crush of poll workers pressed into the building’s narrow corridors trying to reach their designated room. In each room were half a dozen election workers waiting to receive election materials from every corner of the district, double-check the calculations in the protocol, and forward the results for district-level tabulation. 

After verifying the secure delivery of our BBC’s materials, I sought to follow the process a step further. Instead, I was offered a meeting with the judge who chairs Sisli’s electoral board. 

Supporters of President Erdogan and AKP celebrate their election victory in Taksim Square.

It was now well past 9:00 p.m., more than four hours since the polls closed. The judge sat in his office watching two sets of election returns roll in: semi-official results were being broadcast via cable news on a large television across the room while a map on his computer screen that read “Supreme Election Board” was being populated with the official numbers. Although it was impossible for me to tell what discrepancy might have existed between the figures at that moment, opposition leaders were simultaneously turning to social media to reassure their supporters that pro-government media were broadcasting premature results to discourage them. These hopeful claims appear to have been inspired more by optimism than reality—the official results released the next day differed little from what the media was reporting in the evening. At least in central Istanbul, the election results at that time of night were still in the early stages of being compiled at the district level.

The judge explained how in the coming hours the district’s protocols would be digitized, loaded onto a public website, and used to generate a district-level protocol of official election results. As chairman, his role would be to adjudicate disputes and discrepancies in the tabulation and certify the final results. Satisfied that I had followed the process as far as I could, our observation ended. 

Around 10:00 p.m., President Erdogan declared victory. With 52.6 percent of the vote, he had won outright in the presidential election, avoiding a runoff with the leading secular opposition candidate by a comfortable margin. In parliament, AKP fell just short of an absolute majority for only the second time in its 16 years in power. The AKP’s election coalition partner, the nationalist MHP party, surprised many with its strong performance, earning 49 seats in the 600-seat parliament and easily supplying the six seats AKP needs to reach 301 votes in the legislature. Importantly, HDP cleared the ten percent threshold and will be the third-largest party in parliament with 67 seats behind CHP’s 146. Altogether, an impressive 86.2 percent of the population had participated in the vote.

Over a late night dinner in a gentrifying secular neighborhood of Istanbul, I could hear some nearby diners discussing the election results with resignation over glasses of wine. Further off in the distance, the blaring of car horns announced the beginning of celebrations by the President’s supporters. I followed lines of cars festooned with Turkish flags and AKP banners as they streamed toward centrally-located Taksim Square. There, a spontaneous victory party had broken out. A jubilant AKP loyalist was being carried aloft, leading the gathering crowd in chants of “Allahu Akbar!” and “Recep Tayyip Erdogan!” 

The evening stroll between these two contrasting scenes was a journey across a wide social and political chasm in Turkey—a chasm the president may choose to widen or narrow in his new mandate. Recent studies have revealed acute polarization within Turkish society that reflects high levels of social distrust and political intolerance. These ills present critical challenges for governance.

During the campaign, President Erdogan pledged to lift the nearly two-year-old state of emergency upon his reelection. AKP statements since the election suggest that Erdogan may decline to renew the state of emergency when it expires on July 18. This would be an appropriate first step toward rebuilding trust and one the U.S. Helsinki Commission called for in an October 2017 letter to President Erdogan. But lifting the state of emergency might only be a superficial gesture if it is not accompanied by significant prisoner releases and amnesties—particularly for human rights defenders and journalists—as well as meaningful judicial reform to restore the credibility and independence of Turkey’s politicized justice system.

In accordance with its mandate, the U.S. Helsinki Commission will continue to monitor Turkey’s implementation of its commitments as an OSCE participating State to respect human rights and democratic principles. In this most recent election the Turkish people demonstrated formidable levels of political participation and civic engagement. Now and in the future, the government must succeed where it has recently failed to ensure that all its citizens have an opportunity to participate in Turkish society and institutions on the basis of fundamental equality.  

The morning after the election, a woman crosses Taksim Square.