Breaking the News of Government Takedown Requests Amid COVID in India: Conversations with Lumen Researchers

On 24 April 2021, Medianama, an Indian tech and policy online news platform, broke a story about the Indian government’s takedown requests to Twitter to have content critical of the government’s COVID handling removed from their platform. This story took the global media landscape by a storm — with almost every leading national and international news organization quoting the story to highlight the Indian government’s crackdown on freedom of speech. The story was centered around the disclosure of the notices in question made by Twitter on the Lumen Database because it was these government takedown requests that first caught the attention of the story’s author, Aroon Deep.

In the aftermath of the piece that made world news, we spoke with Aroon to further understand his motivation in writing the article and the process and importance of looking into the disclosures made by Twitter. Aroon has been working as a journalist with Medianama for over two years and he began by talking about why these government requests, in particular, were different from others that the Indian government had sent in the past, noting, “In February 2021, the Indian government had ordered Twitter to take down hundreds of tweets that used the hashtag #ModiPlanningFarmersGenocide. After some public back and forth, Twitter’s compromise was to take down tweets from smaller accounts but tweets by major journalistic outlets remained online.” He added, “Even though the ongoing in February was an attack on free expression, there was some level of deniability in saying that the hashtag ‘Modi Planning Farmer Genocide’ was patently untrue. However, this time around, that deniability did not exist.” The takedowns initiated in this instance by the Indian government did not have any hashtags in common.

“The notices we found on the Lumen Database revealed that the tweets were largely just criticism of the government, the kind of political speech that just flies under the radar in most industrialized parts of the world. The notices revealed that the nature of the censorship was clearly unprecedented.”

To understand more about the efforts that led him to find these disclosures, we asked him about the method that he used to trace these takedowns. He cheekily admitted to setting up a bot to help monitor the Lumen Search result webpage, “I landed on a custom result page on the Lumen Database after carrying out an advance search to find the latest government takedown requests in India. I then placed a bot on that page that refreshed the results every five seconds so that it would notify me if there was an update. With this, I slowly got into the rhythm of getting my hands on the latest disclosures made in the Lumen Database. When I found the takedown requests of content speaking critically of the government’s COVID handling, my team and I knew we had something big on our hand.”

Since Aroon’s findings centered around the willingness of an Online Service Provider –Twitter — to disclose the takedown notices it received, we were curious to know if there would have been an alternative way of being able to break the news if these notices had not been available. He was quick to respond:

“What would we have done if we weren’t able to access the notices? I don’t know. The availability of the notices is a really crucial workaround to a very opaque regime of censorship. To date, I have not been able to report on what posts were taken down on Facebook or Instagram, because they don’t share the notices they receive with the [Lumen] Database, or disclose them elsewhere.”

He added, “An alternate approach would be to publicly call for inputs asking people who have received emails from Twitter saying their account or some content on their account was restricted in India. However, the challenges with that would be enormous because it would be highly unlikely that every single such affected user would reach out to us. Second, there is an active debate on whether Indian law actually allows the takedown requests made by them to be kept confidential and this would make the users hesitant to share with us. Facebook and Instagram seem happy to let the interpretation for confidentiality prevail but it’s great to see that Twitter has shown some level of resistance to this interpretation by making disclosures to Lumen.”

This rings true for many countries around the world with authoritarian regimes. The public availability of takedown requests allows journalists to identify patterns of takedown initiated by the government even though the law and the governments by themselves may seek to withhold such information. Platforms that remain committed to transparency through disclosure actively contribute to helping identify such patterns.

Aroon then spoke about where and how the notices and disclosures made on Lumen featured in his research. He emphasized how Lumen had been able to contribute to the conversation and noted, “We sometimes use the database as a research aid when we want to dig deep into takedowns by particular companies. However, because of the nature of our profession as journalists, the notices available on Lumen do not amount to research aid as much as they are the story itself.”

“The notices made available on Lumen have standalone value. More often than not, they are the genesis of our stories about censorship.”

We then spoke about the challenges of navigating the database to find notices. Aroon mentioned that navigating the sheer volume of notices was a challenge in itself. As a journalist based outside of the US, he added, “Most notices are DMCA takedown requests and as a journalist in India, where the laws are different, finding the relevant notices requires time and patience. But the definite upside is having these notices available at all.”

Finally, Aroon concluded by highlighting how notice-sharing-based transparency helped in tech and policy journalism. He fittingly noted: “The usefulness of these disclosures is no longer limited to academic importance. Our article on the takedowns in India is suggestive of that.”

“The article took off the way it did because of what it says about India’s democracy at this time. It’s a narrative that we wouldn’t have been able to successfully weave without the public availability of the notices.”

We finished with a discussion on how helpful and important it would be for other platforms to disclose the notices they receive. Aroon noted, “We’d love to have a lot more visibility into what gets taken down on other platforms like Facebook. I suspect that is where the majority of the censorship happens, but I have no way of proving it because Facebook does not share with Lumen or does not make public notice-disclosures in other ways.”

“The urgency to really have these disclosures keeps increasing because there are explosive levels of control over dialogue hiding behind them.”

__________________________________

UPDATE, May 26, 2021: The Ministry of Electronics and Information Technology, in May 2021, asked all social media companies to remove any content referring to the coronavirus variant B.1.617, which the World Health Organization said was first identified in India, as the “Indian variant”. Social media executives reportedly flagged that this would lead to the takedown of hundreds of thousands of posts and would mean opening the floodgates to “keyword-based censorship going forward”. Since the removal request was made via a letter to all social media platforms and not as a formal takedown request, the Lumen database does not have access to it.

About the Series:

Lumen was conceived and founded in 2001–2002 by Wendy Seltzer, at the time one of the first Fellows at Harvard University’s Berkman Center for Internet & Society, in 2002. Then known as Chilling Effects, the project’s goal was to collect and study the takedown requests sent under the at the time still quite new Digital Millennium Copyright Act (DMCA). Over time, as the Internet evolved, so did the project and its scope. The database now holds copies of not only DMCA notices, but takedown requests of all sorts, including both domestic and international court orders.

Over the years, the database has been used by journalists, academics, policy analysts, and researchers the world over to break news, analyze trends and accuracy in takedown notices, and inform policymaking ideas for platform governance and content moderation online. In this series, we will talk to some of the people who have used the Lumen Database in their work, and learn about their research and the ways in which Lumen has helped inform the incredible work they do around internet law and policy. We hope you enjoy reading about these people and their work as much as we enjoyed collaborating with them! For discussing access to the database, potential notice-sharing, and related questions, please reach out to us at team@lumendatabase.org.

About the Author:

Shreya is an Employee Fellow at the Berkman Klein Center, where she works on the Lumen Project. She is a passionate digital rights activist and uses her research and writing to raise awareness about how digital rights are human rights.