Semtext #7: Disinformation and Indonesia’s 2019 Election

This SemText summarises the real and perceived challenges in Indonesia’s upcoming 2019 election, particularly regarding disinformation and candidate electability.

Throughout 2018 and early 2019, the Presidential Race was a doozy to follow. Political allegiances, according to the media and some observers, have shown to break up families and disharmonize supposedly “neutral” places like schools and places of worship. The stakes in this election are high, as some may put: continue with economic development or risk returning to New Order practices?

One thorny issue that has piqued the interest of the public is the widespread use of disinformation by both sides. How is disinformation used? What are the effects? These are some the questions that the panel addressed.

One note on format. In this SemText, I’ll move in and out from summary to commentary. While this may be confusing, I’ll do my best to denote which is the speaker’s voice and which is mine.

Social media and disinformation

This section highlights contributions made by two panelists, Septiaji and Isabel, who talked about how hoaxes propagated through social media and what can be done.

Septiaji, from Masyarakat Anti-Fitnah Indonesia [Anti-Hoax Community of Indonesia] (MAFINDO), presented quantitative data of the types and frequency of “hoaxes” which were propagated from 2015 until the second semester of 2018.

A short digression. The term “hoax” is used here because it is a common term used by people to describe disinformation. Of course, not all disinformation is the same as hoaxes, but since it would be difficult (and to some extent, pointless) to come up with a solid academic definition, this is what we will work with for now.

MAFINDO classifies hoaxes into several types, namely (1) religious-ethnic-social hoaxes, which may trigger horizontal conflict; (2) health hoaxes, related to misinformation of vaccines and other public health concerns; (3) financial hoaxes, especially related to financial scams and money games; (4) political hoaxes, targeted at political elites; and (5) terrorism-related hoaxes, which aim to delegitimize law enforcement and amplify terrorist acts. However, we will only be focusing on political hoaxes.

The data shows an accelerating upward trend in the amount of hoaxes from 2015 (I cannot show the graph as I do not have it). In 2018, there were around 998 hoaxes generated and propagated. The monthly average in 2018 was 100 hoaxes. Out of the hoaxes that were generated/propagated, a significantly large portion were political in nature; around 260 political hoaxes in the second semester of 2018. In the second semester of 2018, around 30% of the political hoaxes were targeted at Jokowi, while 21% were directed towards Prabowo.

Facebook remains the most popular outlet for hoax distribution. This coincides with the data showing most hoaxes were in the form of text accompanied with photos, as opposed to videos or text-only. I believe this may be perhaps due to the nature of the Facebook algorithm, which prioritizes visuals accompanied with text. Furthermore, such type of content tends to generate more engagement through comments and comment-replies, which further increase the probability of going viral. It could also be because text and photos could easily be generated and shared. Videos are also difficult to produce, however, this may be circumvented through the use of deep-fakes.

What about other social media platforms? Septiaji noted that propagating hoaxes via Instagram is much harder, due to the visual nature of the platform. Furthermore, it could be due to the Instagram community being more apolitical. Isabel Dunston (Chatham House) noted Twitter’s 140-character limit, which stifled hoaxes comprising of long text, along with the “call-out” culture inherent in the platform, which allows for fast debunking.

What is MAFINDO doing to curb the spread of hoaxes? As they are a community-driven organization, most of their activities are conducted online. They assembled in a Facebook group of over 50,000 members who actively patrol the internet to call out and debunk hoaxes. They also have an app in development, called the Hoax Buster, which will serve as a one-stop application to report hoaxes. To enhance the app, I would suggest each report would be filed in a repository, so we could have an archive of hoaxes used.

While MAFINDO is doing God’s work, they are still mostly driven by people with a passion, yet lacking in resources. Though they have partnered up with the government, law enforcement, and tech companies, for their work to be sustainable, they would need to be injected with professional and financial resources.

On the topic of social media, Isabel Dunston, a fellow at Chatham House, highlights the need for “cross-sector responses”, which I believe is jargon meaning “government and private sector cooperation”. Cooperation is a feel-good term, though. Increased enforcement and control of tech companies would be a much more ideal solution.

Dunston mentions some ways which governments and the private sector may address and anticipate the problem of disinformation. She notes the government has typically responded to disinformation in three ways. First, building digital literacy. This has been done through the creation of SiberKreasi, a movement initially started by artists, which aims to increase overall digital literacy. The initiative, however, is hampered by a lack of funding. Second, through law enforcement. This however, is not a particularly effective approach as the preferred method is to arrest hoax propagators under the flimsy ITE Law. So much for freedom of speech. Legislation needs to be updated to prevent a clampdown on civil rights. Even if an arrest is warranted, it should only be a last resort and should be subject to proper due process. Third, the bluntest action is content-blocking. Honestly, this doesn’t do anything, as the Indonesian joke of a firewall can easily be bypassed through DNS spoofing or a simple free VPN.

As for the private sector, Dunston highlights some efforts made by Google, Facebook, and Twitter to curb the spread of misinformation. Facebook opened up its Indonesian office in 2014 and has since entered into partnerships with the local government to combat the spread of misinformation. Some include efforts like linking content to fact-checking websites. However, they face a barrage of fake accounts and still have a hard time understanding the cultural and linguistic context of content. Twitter has cracked down on bots, however, they have a hard time monitoring the so-called “buzzers” a.k.a real humans paid to generate content. For Google, a good start to see what they have been doing in regards to countering disinformation in general is the 2019 Google White Paper. Some measures include suspending AdSense revenue, especially for channels linked to radicalism.

Is disinformation distorting our perception of the election?

Perhaps the most interesting talk was delivered by Ross Tapsell (Australian National University). According to Tapsell, the 2019 election is a relatively “timid” election due to the lack of fervent ideological debate between Prabowo and Jokowi. However, it is perceived to be hostile and polarizing due to the noise generated by social media, especially by buzzers who have incentives to add to the cacophony. Hence, the illusion of polarization.

Tapsell argues the 2014 election was indeed polarizing because Jokowi and Prabowo appeared as polar opposites: one promised reform and advanced through sheer skill whereas the other brought with them New Order vibes. The election was framed as a critical juncture in Indonesian politics and democracy and with good reason.

However, in the “timid” 2019 election, Tapsell notes several things which makes it not as polarizing as the previous election.

First is the lack of contestation and debate on the future direction of the country. In the debates, both candidates have mostly been talking over each other: Jokowi emphasizes his programs, while Prabowo tries to find flaws in Jokowi’s programs but without offering a notable alternative. Related to this is a lack of a charged ideological stance offered by the candidates for this to qualify as a “polarizing election”. Most of the debates have centered on candidates proposing minor changes to already-existing policies, and less about their fundamental ideological stances.  Despite the lackluster debates, the Election Commission has established a “Peace Committee” to deal with debate-related grievances.

Second, despite the noise generated by buzzers, Tapsell notes they are actually cancelling each other out. He points out that electability has not changed; Jokowi’s electability remains high with Prabowo trailing behind. There are also indications that voters are either already entrenched (as I believe is the case for older voters) or they are weary of listening to the noise. Tapsell notes young voters have moved towards Instagram, which is more apolitical, and show enthusiasm for the satirical third candidate, Nurhadi-Aldo.

As such, the election now falls back into becoming a popularity contest and this is where Jokowi would be strongest. Despite recent polls showing a decline in electability, Jokowi’s strongest weapon is his own popularity, which he has show to leverage in many cases. As far as disinformation goes, there is little to suggest Jokowi’s popularity will be affected in a significant manner, although I do have reservations on issues regarding foreign worker influx.

Conclusion

In short, there have been a significant rise of disinformation tactics on social media, particularly political disinformation. There has also been responses from both government, the private sector, and civil society organizations, though the extent of the responses still need to be evaluated. As for the role of disinformation in the 2019 elections, while it has created a cacophony online, there is little evidence to show the tactics are indeed working. Electability polls have not seen significant changes, and it is difficult to pinpoint whether disinformation contributes directly to changes. In sum, disinformation may play a role, but in the 2019 election, it exaggerates polarization, which leads to misperceptions of the nature of the election. Is a humdrum election a good thing? I would agree.


Seminar: Disinformation and Indonesia’s 2019 Election

Convenor: CSIS

Seminar Venue: CSIS, Jakarta

Date of seminar: 15 March 2019

Original header image by Gerd Altmann from Pixabay

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Powered by WordPress.com.

Up ↑

%d bloggers like this: