How the Myanmar coup d'état is playing out on Twitter
Updated: Feb 16, 2021
NEWS KEYWORD SEARCH: I searched for Tweets containing the words "Myanmar" or "Burma" (both names are used, but most English-language media outlets use "Myanmar").
CONTEXT: On February 1, citizens of the southeast Asian nation Myanmar woke up to a military coup. Armed forces had deposed the elected government. They had also detained Myanmar's de facto leader, Aung San Suu Kyi.
MY EXPECTATION: I predicted that the Tweets would recount the essential facts about the coup , with a majority of posts mentioning Aung San Suu Kyi.
Suu Kyi first gained international prominence in the 1980s, praised for leading a non-violent pro-democracy movement in Myanmar. Arrested by military leaders and placed under house arrest, she spent 15 years in detention - during which time she was awarded the Nobel Peace Prize.
However, in recent years, Suu Kyi has fallen from grace on the world stage - since she infamously refused to condemn the military's 2017 genocide against Rohingya Muslims. There have even been calls for her to be stripped of the Nobel Prize.
ROBOT'S RESULTS: The robot read 590-thousand English-language Tweets published in the days following the coup (February 3-4). Here are the most interesting groups of Tweets it identified:
1) Posts with an international focus, such as reactions from other countries, commentary on how the events in Myanmar could affect the broader region, etc.
2) Tweets focused specifically on U.S. reactions to the coup, many of them comparing Myanmar's politics to U.S. politics:
3) Mentions of Aung San Suu Kyi, many of them expressions of support for her:
4) References specifically to Myanmar's military. You will notice that there are many mentions of the military's decision to suspend Facebook (Myanmar citizens are avid users of Facebook, and the platform was infamously used to incite violence during the Rohingya genocide):
WHAT I LEARNED FROM THE ROBOT:
The largest group of Tweets (26%) identified by the robot was "international reactions and commentary", so I thought it was worth my time poring over these Tweets.
Reading the internationally-focused posts, I learned about the coup's potential economic and political impact on the region. Is it obvious that events in Myanmar could lead to regional stability, or interrupt markets? Of course! But without the robot's help, I wouldn't have found these Tweets. Most of them were published by people in Myanmar or in neighboring countries - accounts I don't follow and whose Tweets certainly would not show up in my feed.
And what, specifically, did I learn from taking a close look at these Tweets? I saw messages of support for Myanmar's people and deposed government - published by Twitter users in Vietnam and Cambodia. I read about fears that Myanmar's coup could spark similar events in Thailand and India. I saw divided opinions, from around the globe, on China's and Russia's support of Myanmar's military. Finally, I learned about a growing international call to boycott Japanese beer maker Kirin for its ties to the military junta. In short, the robot pushed me to learn and think about issues, groups, and perspectives far beyond my own U.S.-centric bubble.
TECHNICAL NOTES: Analysis based on Latent Dirichlet Allocation topic modeling. Tweets collected using Twarc API. Data set consists of approximately 590k English-language Tweets collected over an 18-hour period Feb. 3 - Feb. 4.
This is the largest Twitter data set I have analyzed using these methods, and I found that the large amount of data (over half a million Tweets) led to better results (the robot's breakdown of topics was the most accurate I have seen thus far).