We’ve spent the past couple weeks looking at the clash of social networks and democracy in the United States. So let’s turn our attention abroad.
Assam is a state in India that is home to a large population of Bengali Muslims. The ruling party of India, Bharatiya Janata Party (BJP), are Hindu nationalists. In August, after six years of development, the government released a controversial national register of citizens
that omitted 1.9 million residents, many of them Muslim. The government has presented the project as part of an effort to expel “infiltrators,” but the overall effect has been to create an environment of fear for minorities in Assam, many of whom are poor.
None of that is Facebook’s doing. But as we have seen before in other countries where ethnic tensions are running high, the platform has become what the human rights group Avaaz
is calling a “megaphone for hate,” in a report released Tuesday. Here’s Pranav Dixit at BuzzFeed
Comments and posts that called Bengali Muslims “pigs,” “terrorists,” “dogs,” “rapists,” and “criminals,” — seemingly in violation of Facebook’s standards on hate speech — were shared nearly 100,000 times and viewed at least 5.4 million times, showed the Avaaz review, which covered 800 Facebook posts related to Assam. As of September, Facebook had removed just 96 of the 213 posts and comments that the organization reported, including calls to poison Hindu girls to prevent Muslims from raping them. […]
“Facebook is being used as a megaphone for hate, pointed directly at vulnerable minorities in Assam, many of whom could be made stateless within months,” Alaphia Zoyab, senior campaigner at Avaaz, said in a statement. “Despite the clear and present danger faced by these people, Facebook is refusing to dedicate the resources required to keep them safe. Through its inaction, Facebook is complicit in the persecution of some of the world’s most vulnerable people.”
The report is unfortunately not publicly available, and even if it were, I can’t read Assamese. It’s worth noting that Facebook does not agree with Avaaz’s contention that everything the group found is hate speech. (“We have clear rules against hate speech, which we define as attacks against people on the basis of things like caste, nationality, ethnicity and religion, and which reflect input we received from experts in India,” the company told Dixit.) And fortunately, as best as I can tell, nothing in the report links the spread of hate speech in Assam to real-world violence.
Before entering any new market, particularly those with volatile ethnic, religious or other social tensions, Facebook and other social media platforms, including messenger systems, should conduct in-depth human rights impact assessments for their products, policies and operations, based on the national context and take mitigating measures to reduce risks as much as possible.
I wonder what such an impact assessment might have said about Assam before Facebook opened shop there. Does Facebook employ enough content moderators who speak Assamese? How effectively can its machine-learning systems understand potential hate speech in that language?
That leads to a second thing I took from the UN report, which is that social platforms should provide country-specific reports about the hate speech they discover on their networks. As I wrote then:
Facebook ought to provide country-specific data on hate speech and other violations of the company’s community standards in Myanmar. We may not be able to say with certainty to what degree social networks contribute to ethnic violence — but we ought to be able to monitor flare-ups in hate speech on our largest social networks. Dehumanizing speech is so often the precursor to violence — and Facebook, if it took its role seriously, could help serve as an early-warning system.
And in Assam, it seems, that early-warning system is flashing red.