In April, 2017, the first Armenian parliamentary elections under a new political system were held: Following a 2015 referendum, Armenia transitioned from a semi-presidential to a parliamentary political system. Constitutional amendments were introduced to reduce the powers of the president in favor of the prime minister and the parliament. Even though this sounds like the opposite of what was intended in the Turkey referendum just two weeks later, some of our interlocutors assumed this may be a way for the current president to extend his stay in power – as future prime minister.

As a consequence of these amendments, the Electoral Code was also changed and introduced a complex new way to vote, including new technologies: Voters first had to scan their passport or ID as well as their fingerprints on an electronic ‘voter authentication device’ (VAD), which then printed a receipt, have that receipt stamped, sign the voter list, and receive nine ballots. In the voting booth, they had to put one ballot into the envelope and discard the others. One corner of the envelopes was (intentionally) missing, so the official administering the ballot box would put a sticker on the ballot inside the envelope, and take back the receipt first issued from the VAD. Additionally, election officials were to change posts every two hours. What could possibly go wrong?

Individually, each of these steps had the potential to prevent some kind of fraud, for example multiple voting by the same person, impersonating other voters or stuffing the ballot box. Taken together, however, they resulted in an overly complicated system that was made worse by inadequate information and training for the election officials as well as for the voters. As a consequence, huge lines and crowds formed inside and outside of polling stations, leading to chaos. Some voters jumped ahead and missed a table (and thus a verification step), others simply left without voting even after having scanned their fingerprint: At the end of the day, the counting process clearly showed diverging numbers of voters according to the VADs, the voter list, and the ballots (which were fewer, not more, which would in turn indicate stuffing). 

Another measure taken was video surveillance of most of the polling stations, streamed online. Anybody with access to a screen and Internet could observe the voting process, as long as the stream was working. To a degree, this may have increased election officials’ motivation to demonstrate professional conduct. For example, officials at our polling station were quite adamant that we should not eat pizza for dinner inside the polling station, until I showed them on my phone (luckily, the polling station had excellent WiFi) that we were sitting outside the area covered by the cameras. However, witnessing some chaotic scenes might not have improved voters’ confidence.

Perhaps because anybody could observe the elections via video stream anyway, perhaps because it had already accredited over 28,000 citizen observers, the Central Election Commission officially refused all international NGOs an invitation and accreditation to observe the elections. On the one hand, the large number of citizen observers could point to the great interest of Armenian civil society in contributing to democratization and fair elections (for example, some citizen observers did not even vote because they were posted outside their home district, but thought holding election officials accountable and trying to ensure a correct voting process to be more important than casting a single vote that may be discounted during tabulation or watered down by bought votes). On the other hand, several interlocutors expected a large share to be ‘false-flag’ party proxies, in order to give parties even more opportunity to direct and possibly interfere with the voting process.

As a consequence, Silba STOs were accredited with the help of Transparency International’s Armenian branch and received accreditation as domestic observers through local NGOs. For election day, each Silba observer was paired with an Armenian STO from another NGO (since according to the new Electoral Code, only one observer per organization was allowed per polling station). I was fortunate enough to be paired with a great partner deeply invested in democratization and especially, pointing out the rules to the election officials and insisting they follow them – even if it meant suffering lots of (verbal) abuse for him. Spending a whole day at one polling station (instead of an hour or less in ten different places, as international observers usually do) provided a deeper insight into the workings of that polling station: Not only could (im-)proper procedures be observed more thoroughly (for example, when the OSCE team arrived, everything was fine – but they did not see how chaotic it was before and after their visit), but also the dynamics among the election commission, party proxies, and various observers.

Citizen observers play a more active role than international observers, and while far from interfering, may be more invested to drawing election officials’ attention to improper procedures, noting this in the protocol, and demanding its rectification instead of simply noting it on a multiple-choice form and reporting it later. On the other hand, international observers may get more attention from national authorities and (inter-)national media. They may also be freer to express their opinions and findings without fear of retribution. As a result, I hope Silba and other international election observation organizations can take into account the benefits of partnership – in whichever way – with credible local citizen observer organizations.