Women politicians in India face a shocking scale of abuse on Twitter, reveals data from a new study titled “Troll Patrol India: Exposing Online Abuse Faced by Women Politicians in India.” The study was conducted using crowdsourcing, machine learning and data science. It reviewed hundreds of thousands of tweets sent to 95 women politicians — an analysis on an unprecedented scale in India, said Amnesty International India today.
Amnesty International India, in collaboration with Amnesty International – International Secretariat (AI-IS) conducted the study to measure the nature and scale of online abuse faced by women politicians in India. It found that women who express their opinions online are targeted with abuse not just for their opinion, but also for the various immutable identities – such as gender, religion, caste, marital status and many more.
“Touted as a ‘safe place for free expression’, Twitter was popularly envisioned to be a platform where marginalized populations, including women, Dalits and religious minorities, would have an equal opportunity to make their voices heard. While over the years the social media platform has evolved into an indispensable tool for political engagement, campaigning and activism, women are regularly and relentlessly subjected to abuse on the platform, which has a silencing effect on them,” said Avinash Kumar, Executive Director, Amnesty International India.
The study analyzed 114,716 tweets mentioning 95 Indian women politicians in the three-month period of March-May 2019 in the lead-up to, during and shortly after the 2019 General Elections in India. The women politicians in the study, represented a variety of political views spanning the ideological spectrum. The tweet mentions were decoded through the microsite Troll Patrol India by over 1,900 digital volunteers from 82 countries of which 1,095 were from India.
The analyses found that 13.8% of the tweets that mentioned 95 women politicians in the study were either “problematic” or “abusive.” This amounts to over 10,000 problematic or abusive tweets every day for all the women. The study defined problematic content as tweets that contain hurtful or hostile content, especially if repeated to an individual on multiple occasions, but do not necessarily meet the threshold of abuse. It also found that Muslim women politicians received 94.1% more ethnic or religious slurs than women politicians from other religions. The women politicians from political parties other than the ruling Bharatiya Janata Party also experienced more abuse. Other key findings are enclosed below.
Amnesty International India shared its findings with Twitter India on November 2019 and sought its response concerning reporting process, moderation and language detection. It also sought to understand whether any specific measures were taken by Twitter to decrease the online abuse during the 2019 General Elections in India.
In its response, Twitter said that “building a Twitter free of abuse, spam and other behaviors that distract from the public conversation is one of their top priorities. It has made strides in creating a healthier service and continues to further invest in proactive technology to positively and directly impact people’s experience on the service.”
However, many women politicians Amnesty International India spoke to described how Twitter is failing in its responsibility to respect women’s rights online.
Shazia Ilmi from Bharatiya Janata Party said: “More women should be entering politics. But the price that I pay is too much for what I choose to do. The price includes being trolled incessantly, being the victim of online harassment, having a lot of remarks passed about what I look like, my marital status, why I have or don’t have children, etc. — all the filthiest things you can think of. If they don’t like my strong opinions, they do not remark on my work but call me a ‘whore’ in every language that is used in India”.
Atishi from Aam Aadmi Party said: “It is not the role of each woman to individually ensure her safety in public space. For example, if a woman steps out in public transport, it is the government’s role to ensure that they are safe there. Similarly, if there is a woman who is accessing social media on Twitter, it is the responsibility of the platform to ensure that it is a safe and secure space for women”.
Kavita Krishnan, from Communist Party India (Marxist-Leninist) Liberation, told Amnesty International India: “It is particularly frustrating and mentally stressful when you report something, but Twitter or Facebook says that ‘this does not violate their norms’. I feel these platforms should either get rid of their reporting business or stop pretending that they have these norms; because if they are not going to act on them, then they should not have them at all”.
Amnesty International India’s Recommendations to Twitter
Twitter’s corporate responsibility requires the platform to ensure that its policies are transparent, uniform, and are based on human rights standards and gender-sensitive due diligence. The study outlines concrete recommendations to Twitter for fulfilling its responsibility. They include:
- Publicly sharing comprehensive, meaningful and disaggregated information about the nature and levels of online abuse against women on a country by country basis, as well as other groups, on the platform, and how they respond to it.
- Improving its reporting mechanisms to ensure consistent application and better response to complaints of violence and abuse.
- Providing more clarity about how it interprets and identifies violence and abuse on the platform and how it handles reports of such abuse.
“Online abuse has the power to belittle, demean, intimidate and eventually silence women. Twitter must reaffirm its commitment to providing a ‘safe space’ to women and marginalized communities. Until then, the silencing effect of abuse on the platform will continue to stand in the way of women’s right to expression and equality and Twitter will continue to fail to keep women safe from violence and abuse”, said Reena Tete, Manager, Gender and Identity-based Violence Program.
Below are the key findings:
1. 1 in every 7 tweets that mentioned women politicians in India was ‘problematic’ or ‘abusive’
13.8% of the tweets that mentioned 95 women politicians in the study were problematic (10.5%) or abusive (3.3%). This amounted to 1 million problematic or abusive mentions of the 95 women between March and May 2019, or over 10,000 problematic or abusive tweets every day across all women in the sample, or 113 per woman per day.
2. Indian women politicians experienced substantially higher abuse than their UK and USA counterparts
A similar study conducted by Amnesty International in 2018 to measure the online abuse faced by 323 women politicians in the UK and USA found that 7.1% tweets mentioning the politicians were problematic or abusive. The Troll Patrol India study, using a similar methodology but focusing on a shorter period during elections found that Indian women politicians experienced 13.8% problematic or abusive tweets, which is substantially higher.
3. Women politicians prominent on Twitter were targeted more
There is a clear correlation between the number of mentions and the proportion of abusive content received by women. The more visible the politician is, the more abuse she received.
We tested this by grouping the politicians by mentions. The top 10 most-mentioned politicians were found to have a mean of 14.8% problematic and abusive content versus 10.8% for the others. This meant that while the top 10 received 74.1% of all mentions, they received 79.9% of problematic or abusive mentions.
4. 1 in every 5 problematic or abusive tweets were sexist or misogynistic
Decoders were directed to label the abuse as containing sexism and/or misogyny, ethnic or religious slur, racism, casteism, homophobia or transphobia, sexual threats, physical threats or ‘Other’. Other was used when the problematic or abusive content did not fit in the categories provided. Notably, the most common selection was the “Other” category (74.1%) which indicates that most problematic content did not fit neatly into any of the suggested categories. Nevertheless, of all the tweets that were labeled as problematic or abusive, about 1 in 5 answers showed sexism or misogyny.
A deeper look at the tweets that demonstrated sexism or misogyny showed that sexism was experienced by women across all spectrums of political ideology and affiliation, religion, caste, race, age, marital status and election outcome.
5. Muslim women politicians received 94.1% more ethnic or religious slurs than those from other religions
Women politicians who are or perceived as Muslims received 55.5% more problematic or abusive content when compared to women from other religions. 26.4% of the problematic or abusive content Muslim women experienced contained to ethnic/religious slurs, nearly double the proportion for women who are or perceived as Hindus (13.7%).
6. Women politicians belonging to marginalized castes received 59% more caste-based abuse compared to women from other castes
Women from marginalized caste received 59% more caste-based abuse than women from general castes. In cases, where problematic or abusive content was identified, women belonging to marginalized castes, such as Scheduled Castes, Scheduled Tribes and Other Backward Classes (8.6%) received more caste-based slurs than those belonging to general (5.4%) or unknown castes (7.2%).
This indicates that caste identity is more often than not, a key element of problematic or abusive content for women belonging to marginalized castes.
7. Women politicians from ‘parties other than Bharatiya Janata Party’ experienced more abuse
Most of the politicians with tweet mentions (about 76%) belonged to either the Bharatiya Janata Party (BJP) or the Indian National Congress (INC), and about 62% of tweets in our sample mentioned someone from one of these parties. Compared to the ruling party BJP, women politicians from ‘other parties’experienced 56.7% more problematic or abusive content than BJP while INC politicians received 45.3% more abusive or problematic content than BJP.