Overview
The information and communications operations of the Centre for Monitoring Election Violence (CMEV) for the Parliamentary Election held on 17 August 2015 was anchored to the foundations laid for the Presidential Election in January this year. Operational security from an information and communications perspective was radically different to the context in January. As subsequent reports have shown, the intelligence services, military and Police under former President Mahinda Rajapaksa had invested heavily in surveillance infrastructure to contain, control and censor information, including around elections. CMEV’s info-sec (information security) and op-sec (operational security) considerations in August were not hostage to these same considerations and were thus re-engineered to be less around secure and resilient communications, and more around information dissemination.

Platforms and apps
As has always been the case, CMEV’s central hub remains its website where press releases, statements, communiques, infographics and mainstream media coverage of operations and output were added on a daily basis in the lead up to the election, as well as on election day. Leading up to and on election day, CMEV’s website saw a surge in traffic. On election day, the website was viewed over 1,300 times. Designed to be resilient against denial of service attacks as well as traffic spikes (hosted on WordPress.com) CMEV’s website – the only fully responsive website of an election monitoring body in Sri Lanka (rendering perfectly on whatever browser or device it is viewed on) – coped well.

Continuing what was started in January for the Presidential Election as a means of putting out information around monitoring operations even if CMEV’s website came under attack, WhatsApp was employed extensively in the lead up to and on election day. Two groups, totalling over 300 individuals from Sri Lanka and abroad received over 36 updated which included audio, text and links to documents in Sinhala, Tamil and English plus images in the form of charts and infographics. Every single update put online was sent through WhatsApp. The Sinhala civic media website Vikalpa, anchored to the Centre for Policy Alternatives (CPA), also managed a WhatsApp based information service of its own, with its Editor – who was a CMEV monitor in Jaffna – providing updates in Sinhala throughout the day to over 100 individual subscribers.

CMEV’s Twitter account (@cmev) on election day alone published over 74 tweets. These included retweets of situation updates, engagement with other users, content and updates from CMEV plus links to pertinent news reports from the web. In contrast to the reach of CMEV’s website, CMEV’s Twitter feed alone, on election day, was read 23,534 times, over over the 7-day period ending 17 August, the account got 32,800 impressions – an average of around 5,000 impressions day. Over 230 clicked through to the web links the CMEV account tweeted, 177 retweeted our updates and 27 had favourited them. The level of engagement with CMEV’s content over social media was quantitatively and qualitatively higher than the Presidential Election in January.

In addition to CMEV, Groundviews (@groundviews) also put out updates on Twitter related to the election, complementing the output of CMEV. Whereas CMEV’s feeds stuck to information from CMEV itself or that which was from an official source, Groundviews tweeted opinion and news from other curated sources from Sri Lanka and abroad. Groundviews on Twitter over the 17 and 18 August was read over 80,000 times. Over the 7-day period ending 18 August, the account got 91,200 impressions, an average of around 19,500 a day. Also over the course of the week, 1,700 had clicked links tweeted by the account, over 360 had retweeted updates and close to a hundred had favourite them.

@CPASL, the Twitter account of the Centre for Policy Alternatives (CPA) also pushed out over 48 tweets on election day.

It must be noted that for all the accounts above, traffic was organic – in that there was no paid advertising pushing traffic to these accounts.

In addition to Twitter, CMEV, throughout election day, put out audio updates on Soundcloud.com, which were subsequently featured on CMEV’s Facebook, Twitter and sent over WhatsApp as well. Updates were in Tamil, Sinhala and English, at around the start of polling, mid-day and once polls closed. On the 17^th alone, over 370 listened to these updates.

CMEV’s Facebook page, on which all output from CMEV is featured along with curated content from other official sources, saw a surge in traffic over the week. Reaching over 15,200 accounts, over 4,500 ‘liked’ the page in the course of the week. Engagement with content on the page was also very high, with over 1,000 fans commenting on, liking or sharing the content published by CMEV. Organic readership on the 17^th was exceeded 2,000.

Continuing a tri-lingual campaign to encourage the exercise of franchise particularly amongst first time voters and those between 18 – 24 in particular, Groundviews, CPA and CMEV as well as Maatram (CPA’s Tamil language civic media eco-system) and Vikalpa re-featured the #iwillvote campaign across all web properties as well as, on the morning of the election, through an email sent to around 8,000 subscribers. The Editor of Groundviews Sanjana Hattotuwa posted an image at 7.12am on Groundviews noting that he had voted, using the virality of selfies particularly amongst the target demographic to promote the exercise of franchise. At 6.30am, the campaign was published over social media and sent out over email. By mid-day, this had resulted in hundreds uploading to Instagram, Facebook and Twitter images they had taken after voting. Though less in number overall than on the 8^th of January, the #iwillvote hashtag was re-ignited independent of CPA/CMEV/Groundviews, demonstrating the interest in the larger community around voter mobilisation through online and social media channels. The use of #iwillvote’s images (copyright free in tri-lingual) also reinforced the enduring validity of the campaign launched by CPA earlier this year.

General observations around social media
In addition to the information eco-systems of CPA, Groundviews, CMEV, Maatram and Vikalpa – reaching hundreds of thousands – the larger social media ecosystems around the Parliamentary Election is worth further study. Snapshots of this landscape on and leading up to election day follows.

For the first time, Facebook encouraged all its users based in Sri Lanka to update their status messages around the election. While opt-in (users could choose to ignore the Facebook prompt) this move by Facebook around voter mobilisation resulted in thousands updating their Facebook status to reflect the fact they were going to vote, had voted and various shades of political opinion. The status message also linked to the Election Department’s official website, which promptly crashed on account of the traffic. While re-instated later in the day, the engineering into such a critical website – which has repeatedly gone down at every single major election – remains hugely suspect and in need of urgent reform. Most citizens were delivered information around the results of the Parliamentary Elections via SMS news services or election results dashboards set up on private media websites, in addition to of course radio and TV broadcasts.

Another key feature around the Parliamentary Elections was the use of Facebook in particular to promote candidates and political parties through individual accounts (i.e. through voluntary disclosure and without paid advertising). Many on Facebook used the platform to publicly reveal who they were voting for, why and encouraged others to vote for the same candidates or party. Others were more discreet and said they were voting to continue the change brought around on the 8 of January.

The open publication of political preferences – not just to Facebook friends, but as public posts anyone even without a Facebook account, or outside one’s friend’s network could see – is a major development in the discursive landscape of social media, and was reflected in updates in Sinhala and Tamil too (though your author’s primary focus was content in English). One possible reason for this is the freedom of expression, felt across the political spectrum, after the change of Presidency in January. Whereas in January there was a groundswell of opinion over social media around regime change, just a handful expressed openly the desire to be rid of the Rajapaksa’s and there was not a single update across any social media observed that to the extent evident on and around 17 August openly stated voting preferences, political opinion and political opposition.

Tacitly acknowledging the power of social media and its reach, the former President Mahinda Rajapaksa took to Twitter at 7.29 on the morning of the 18^th to counter a story on AFP noting that he had conceded defeat and that his dreams of becoming PM were dashed. The story grew virally across local and international mainstream media as well as social media. @groundviews over Twitter and Facebook first tweeted Udaya Gammanpila’s Facebook update that the UPFA had lost the election, refuting directly the former President’s comment on Twitter that he couldn’t accept victory or concede defeat until the final results were officially released. This interplay of social media and web based mainstream media news resulted in heightened traffic to sites and accounts with updates around what the former President said, or didn’t say, along with others from the UPFA contradicting him or seemingly conceding defeat as well. It is unclear if TV and radio broadcasts picked up on this traffic spike. Reflecting this heightened exchange, at around 7.30am on the 18^th, when official election results were trickling in, Twitter updates around the elections reached their peak. Groundviews monitored two hashtags #SLGE15 and #GenElecSL – which were by Sri Lanka’s Twitter community writ large accepted as the two ‘official’ hashtags to use around election updates. #SLGE15 at around 7.30am was peaking at around 9 tweets a minute. #GenElecSL at around 24 tweets a minute. As long as Groundviews has archived public tweets around events in Sri Lanka (Twitter Q&A, UN debates, elections) this was the most amount of tweets ever published in the same span of time.

Social media – Facebook and Twitter in particular – were also used to fan discord and sow unrest by spreading false rumours. The LTTE and UNP were frequently joined, and others claimed the result of the election was “an international conspiracy”. Yet others on Twitter said the new Government would lead to Eelam’s fruition, and many published derogatory comments about Ranil Wickremesinghe, the Leader of UNP, and surprisingly, also on President Sirisena. Given reports that former President Mahinda Rajapaksa employed vast numbers of individuals in strategic, carefully planned media operations including over social media to construct his image, and shape the discourse (including by planting false information and rumours), the effect of these disinformation and misinformation was minimal. One reason for this could be, as noted early, the opening up of the discursive spaces online resulting in the former President’s modus operandi – controlling the meta-narrative in a tightly censored space – falling apart. With hundreds of independent voices all publishing at the same time and calling to question the record of the former President and those in his government, attempts to incite hatred and the use of incendiary language did not have the same impact as intended, and would have been possible under a repressive political context.

Overall, the information and communications operations of CMEV, complemented by the updates put out by Groundviews, Maatram, Vikalpa and CPA itself, remain unparalleled. From field based information gathering to communicating in real time to CMEV’s HQ, from the use of social media to the coverage CMEV operations received across Tamil, Sinhala and English mainstream print media – including the international press – was qualitatively and quantitatively more than the Presidential Election. CMEV’s information generation, dissemination and storage remains a model for other electoral processes to emulate, especially those operating in austere, censorious and authoritarian contexts.

All CMEV output is in the public domain, and copyright free.