A look back at the keynote session "From online violence to digital respect" at SIF 2019

On Thursday, 21 November 2019, the Safer Internet Forum (SIF) took place in Brussels, Belgium. With a theme of "From online violence to digital respect", it also celebrated 20 years of safer/better internet funding by the European Commission. Below, read the summary of the keynote session led by Thordis Elva, Icelandic writer, speaker and journalist.

Date 2019-12-17 Author BIK Team Section awareness

Thordis Elva is a writer, speaker and journalist from Iceland. She observed a lack of discussion in Iceland around issues related to harmful online content or intimate topics and so, to address this, she has deployed several initiatives. She has published two books on gender-based and sexual violence, which were published in 14 countries. She is also the author of three award-winning, educational short movies about online abuse and image-based exploitation, bodily integrity and sexual consent. In her country, she is a member of a parliamentary committee overseeing the national action plan against violence against women and children, and she was commissioned by the Icelandic government to make educational material for children and teenagers to prevent sexual abuse. From 2012 to 2014, she has been the Chair of the Board of the shelter for battered women and girls in Reykjavik – she has been a board member since 2010. She has been a prominent public speaker for two decades, delivering a TED talk about her personal connection to the issue of sexual violence – the video has been watched over 5 million times to date. Thordis has also given workshops to 18,000 people in five countries about image-based sexual abuse online. She has been working within the media for 15 years, including on television, on print and online.

Thordis Elva opened her keynote speech by providing an overview of the challenges faced by children and young people online, with a focus on the online abuse of women and girls. The internet has revolutionised how we communicate, including in areas such as banking, education, health and so on.

Gender equality is a persistent problem, both offline and online. It manifests itself in many forms. Women are underrepresented in the media and in the political arena – they hold only 24 per cent of the world's parliamentary seats, and only a quarter of the people featured in European news media are women. The gender pay gap is significant, since it remains in double digits even in the most gender-equal countries. Additionally, unpaid labour and housework is still largely in the hands of women.

The fact that women own less and have less influence than men is a result of laws and traditions that made it impossible for women to inherit property, educate themselves, vote or otherwise have full citizenship and civil liberties.

Globally, one in three women is subject to physical and/or sexual violence at the hands of a man in her lifetime. According to the UN Broadband Commission, women and girls are 27 times more likely than their male counterparts to be abused. 73 per cent of women have endured online violence, and this figure rises even more when we consider minority groups (LGBTI, ethnicity, disability). Moreover, women aged 18-24 are at an increased risk of being abused online. In the EU-28, 18 per cent of women have experienced a form of serious internet violence since the age of 15, which corresponds to about 9 million women.

The consequences of this online abuse are very real. 41 per cent of women who experienced online harassment felt that their physical safety was threatened. According to figures established by Amnesty International, 1 in 2 women experienced lower self-esteem or loss of self-confidence as well as stress, anxiety or panic attacks as a result of cyber violence and hate speech online. Even worse, the risk of suicide attempt is 2.3 times higher for a victim of cyber harassment compared to non-victims, according to UNICEF. Meanwhile, the FEMM Committee (the European Parliament's Committee on Women's Rights and Gender Equality) reports that illegal online hate speech targeting gender identity makes up only 3.1 per cent of reports to internet platforms in the EU, clearly indicating that the online abuse of women and girls is an underreported problem.

Thordis Elva conceptualised it as a continuum, defined as a "continuous series of elements or items that vary by such tiny differences that they do not seem to differ from each other". In that context, she introduced the pyramid of gender-based violence. The most extreme elements at the top (murder) are resting on and enabled by all the other layers – rape and assault; physical, emotional and financial abuse; threats, verbal abuse and harassment; rigid gender roles and stereotypes, glass ceiling; sexist jokes, language, objectification of women, and so on.

Image of the pyramid of gender-based violence

Thordis Elva then went on to examine how exactly gender-based violence is manifested online. First is verbal violence, which takes the form of sexist hate speech, defined as all the expressions which spread, incite, promote or justify hatred based on sex, as well as the use of sexist and insulting comments, slut-shaming and victim-blaming. It can also be mob attacks. Cyberbullying is also frequent; in this case, a person is targeted by repeated aggressive online behaviours with the objective of frightening and undermining that person's self-esteem or reputation – which sometimes pushes vulnerable individuals to depression and suicide. Cyberharassment is another common phenomenon, including unwanted sexually-explicit emails, text (or online) messages; inappropriate or offensive advances on social networking websites or internet chat rooms; threats of physical and/or sexual violence by email, text (or online) messages; hate speech, meaning language that denigrates, insults, threatens or targets an individual based on her identity (gender) and other traits (such as sexual orientation or disability). Cyberstalking consists of monitoring someone using electronic means with the intent to cause fear and/or distress. Finally, doxxing consists of publishing someone's personal information without their consent.

Next to verbal violence, there is graphic violence. This category manifests itself in the form of image-based sexual abuse (also known as revenge porn), including fakes and upskirting; unsolicited nudes (including "dick pics"); sextortion, which consists of using intimate material as a means of blackmail; hacking; impersonation (for example, stealing someone's identity and using it to advertise prostitution); trafficking and recruitment; and child sexual abuse (including "grooming").

Men and women can both experience violence online, but it affects women much more and has heavier consequences on the lives of women, possibly because the violence directed at women often has sexual undertones and runs deeper. According to the Swedish Centre of Crime Prevention, 70 per cent of online harassers are men. According to the Pew Research Centre, men are also affected by online abuse, but it is more at the surface level (with name-calling being the most frequently reported form). Men are less likely to find it upsetting (only 16 per cent of them find it "very" or "extremely upsetting"). Meanwhile, 63 per cent of women who had experienced online abuse said they had not been able to sleep well as a result of it. Well over half (56 per cent) of women said online abuse or harassment had meant that they had been unable to concentrate for long periods of time, according to a study by Amnesty International.

Amnesty International showed that this phenomenon leads women and girls to self-censor; they take precautions to prevent sexual violence, which affects their participation in society. To be precise, 76 per cent of women who said they had experienced abuse or harassment on a social media platform made changes to the way they use the platforms. This includes restricting what they post: 32 per cent of women said they had stopped posting content that expressed their opinion about certain issues.

The result of this is a loss of millions of euros, because gender-based online violence engenders lost wages; reduced participation in society; chronic physical conditions and loss of life expectancy; mental health conditions; sexual and sexual health issues, sometimes causing problems with reproductive health; substance abuse and organised crimes; social isolation; and individual and public expenditure on medical protection, judicial and social services. Regarding cyber violence happening in the context of intimate partner violence, researchers from the YWCA (Young Women's Christian Association) have estimated the cost associated with responding to technology-based victimisation to be "$1,200 compared to $500 for survivors of non-technological abuse".

Much more importantly, the future of the planet is at stake – women are much more affected by climate change now, and they will be too in the future. Simultaneously, they are silenced much more than before, so their voices are more needed more now than ever.

Thordis Elva then mentioned the famous internet expression "do not feed the troll", which normalises violence and leaves victims silent. Studies have shown that female users are "feeding the trolls" just by existing as women in the online environment.

  • Having a female name – a study by the University of Maryland's A. James Clark School of Engineering found that chatroom participants with female usernames received 25 times more threatening and/or sexually-explicit private messages than those with male or ambiguous usernames.
  • Expressing an opinion, especially on topics like politics, religion and feminism, as seen in higher levels of abuse towards female journalists, bloggers, politicians and activists.
  • Uploading a picture of yourself.

Image "Boys will be boys"

As such, for women, the only guaranteed way to avoid abuse is not existing at all. In addition, Thordis Elva questioned the whole conception of "trolls", saying that "by calling a person who shares hateful views towards women and girls online a 'troll', we are giving them a name we would not give them in the offline world. This creates the misconception that 'trolling', which involves trying to install fear in women and girls via threats, humiliation and silencing tactics, thereby undermining their agency and right to participate in democratic society, is an online phenomenon. But there are no trolls. What we have are people with sexist beliefs, misogynists who threaten women's rights online as well as offline."

Instead, how can women and girls react to online hate? Thordis Elva suggested building communities to defend one another and speak up. There is not one single way to respond to all attacks. Ignoring the abuse is an option, but it should not be the only one nor should it be the most common one.
Thordis Elva then moved on the topic of sexting – defined as consensual sexual communication – often photographs – that depict nudity or have a sexual undertone. Indeed, sexting is becoming more and more common place (54 per cent of young people in a US study have received or sent sexting messages before the age of 18).

Regarding the motivations for sexting, participants to the "Ungt Folk 2018" study of 10,450 Icelandic teenagers revealed that there are positive aspects to it. It is a way of expressing sexual desire in a consensual relationship; of having sexual relationships without the risks of sexual acts in person; of obtaining sexual pleasure easily; and of discovering one's sexuality. Some girls also mentioned that sending nude photos is a form of initiation into sexual maturity and girls often feel pressured into doing it. Indeed, girls experience six times more pressure than boys to share sexually-explicit content – this puts them at heightened risks of "revenge porn" (which is a misnomer and should be named as the assault that it actually is).

The sharing of nude and sexual imagery online poses great problems since 88 per cent of sexually-explicit material takes off when uploaded to the internet. 17.5 per cent of sexually-explicit photos depict children under the age of 15, and 7.5 per cent depict children under the age of 10. A growing number of children post such photos; the majority of them girls, according to the Internet Watch Foundation (IWF). The sharing of nude and sexual imagery without consent, also called "revenge porn" (although inappropriately, as is "non-consensual pornography") is the result of different things. It can stem from hacking or theft, or the photos may have been shared with consent, before being forwarded or distributed without consent or even without knowledge. It can also be the result of coercion, grooming, blackmail/sextortion, or violence. It can be taken/filmed without the photographed individual's knowledge or be the result of digital manipulation. It embodies the next generation of sexual violence.

When image-based sexual abuse meets doxxing, it very easily leads to a reputation hijack that can limit the victim's employment, social and economic status. Thordis Elva also mentioned fakes, emphasising that "not taking nudes" is simply not a solution anymore. When a person's intimate pictures (whether real or fake) are online, it also creates the possibility of the sexual assault being replicated, which makes it especially psychologically harmful for the victim. As such, such abuse has very real consequences, with some young girls going as far as committing suicide because of it.

Thordis Elva also said that we should address the elephant in the room that is the effect of pornography, especially the prevalence of violence in porn and the early contact children have with (violent) pornography. Educational materials about this exist, but they are not widely distributed.

So what can we do? Thordis Elva mentioned several avenues for action: education; local resources for victims of online violence; legal frameworks and law enforcement; and synergising treaties and collecting data.

In terms of education, children need to be educated on the rights and responsibilities that come with their digital citizenship, by putting it on the curriculum in schools. Words like "trolling" need to be replaced with terms that fully hold the assaulters accountable, like "hate speech" and "harassment". Media literacy should be integrated to school curricula, if that is not already the case. Traditional sex education should comprise age-specific digital sex education, with an emphasis on consent, if that is not the case already. Education stakeholders should invest in awareness campaigns for parents and for wider society in collaboration with stakeholders (NGOs, victim support groups, and so on) about what online abuse/harassment is; what cyber civil rights are; how to report online abuse/harassment; and where support is available in case of online abuse/harassment.

In terms of local resources for victims of online violence, governments should invest in these by:

  • Strengthening organisations/NGOs/initiatives that offer support services to victims, including psychological help.
  • Maintaining a hotline or 24/7 reporting service that can help victims of online abuse, with skilled staff who can react immediately to help stop the spread of harmful material.
  • Making removal help available in the local language for victims of image-based sexual abuse – in that regard, Thordis Elva pointed to the removal guides from the Cyber Civil Rights Initiative.

In terms of legal framework and law enforcement, a gender perspective should be applied when writing and adapting existing laws that criminalise all forms of online violence, including re-sharing or harmful content, as well as of threats to do so. Criminal and civil causes of action should protect the privacy of victims, avoiding the secondary victimisation that comes with having their case made even more public. Image-based sexual abuse should be treated as a sex crime and anonymity should be extended to victims. As online abuse consists of cross-border crimes, synchronising the definitions of online abuse would be helpful in terms of collaborating in the pursuit of perpetrators, swapping best practices and conducting international comparisons. Authorities should ensure effective regulation on internet intermediaries that prohibit any forms of violence against women, in accordance with local law and international treaties. Governments should hold social media platforms to account when it comes to preventing and removing online abuse. They should also equip law enforcement with the funding, technical equipment, knowledge and personnel resources needed to effectively handle cases of online abuse.

In terms of synergising treaties and collecting data, Thordis Elva pointed to the different legal documents that mention online violence against women:

  • The Budapest Convention on Cybercrime and additional protocol (2001) – three articles of the Budapest Convention can apply to cyber violence against women, including article 4, 5 and 9.
  • The Istanbul Convention (2017) – several articles of the Convention can be applied to the specific topic of digital violence, including articles 33, 34 and 40.
  • The Lanzarote Convention – the Council of Europe Convention on Protection of Children against Sexual Exploitation and Sexual Abuse requires criminalisation of all forms of abuse against children, including online abuse.
  • Image-based sexual abuse could fall under the GDPR (General Data Protection Regulation) provision on "processing of personal data" and would consequently trigger application of the Regulation. The individual responsible for uploading image-based sexual abuse material as well as the publisher of such material could be considered joint data controllers, and hence fall under the obligations and sanctions imposed by the GDPR (154).
  • The Treaty on the Functioning of the European Union (TFEU) includes the possibility to develop legislation on violence against women in the framework of judicial cooperation.
  • Last but not least, both the Istanbul Convention and the Victims' Rights Directive require Member States to report statistical data and to produce gender-disaggregated data on cyber violence and hate speech.

After Thordis Elva's keynote speech, a panel discussion followed, involving Frida, BIK Youth Panel 2018 and 2019, Roger Loppacher, President, Consell de l'Audiovisual de Catalunya (CAC), Thomas Myrup Kristensen, Managing Director EU Affairs, Facebook, and Stephen Turner, Head of Public Policy, Government and Philanthropy, Twitter. The panel discussion was chaired by June Lowery, Head of Unit, Accessibility, Multilingualism and Safer Internet, DG CONNECT, European Commission.

Image of Roger Loppacher speaking at the Safer Internet Forum 2019

Roger Loppacher, President of the Consell de l'Audiovisual de Catalunya (CAC) thanked Thordis Elva and introduced the CAC's work on fighting online violence. The CAC began monitoring risk content on platforms, focusing on identifying content that is harmful to minors. They set up the eduCAC programme, offering educational resources to primary and high schools and to families. The topics identified as harmful content are content that promotes anorexia; incitement to the hatred of women; content that incites to suicide; content that promotes child sexual abuse; incitement to hatred of LGBTI people; content that promotes gambling and betting; content that promotes cyberstalking women; and content that sexualises videos of minors.

Roger Loppacher gave the audience a brief overview of the scale of the problem, and what the CAC usually does to respond, such as reporting to platforms, complaining to the public prosecutor's office, carrying out public presentations with partners, and drafting guidelines. He also talked about a programme the CAC created called eduCAC, developed in collaboration with the Department of Education of the Generalitat, the Professional Journalists Association of Catalonia, the Catalan Audiovisual Media Corporation, and "La Caixa" Bank Foundation. eduCAC offers educational resources to primary and high schools and to families, with the aim of fostering a critical attitude towards audiovisual content and promoting responsible use of mobile devices.

In this context, the CAC, the Catalan Audiovisual Corporation and "La Caixa" Bank Foundation developed the awareness campaign #AMiNoMenganyen (They don't fool me), a message to empower young people in their screen use. The overall aim is to encourage critical analysis and responsible technology use.

A group of Catalan influencers and YouTubers (together accumulating 1 million followers) will participate in the campaign, giving their take on the internet and social network use via Instagram, YouTube and TikTok stories.

Image of the panel discussion during the main session at the Safer Internet Forum 2019

Frida, BIK Youth panellist in 2018 and 2019, said she recognised these challenges very well through the eyes of youth, since young people encounter these types of violence, as perpetrators or as victims. According to her, talking about such issues is fundamental, whether at home or with friends. In her Finnish high school, such topics are part of ICT classes. She believes education is the best way to tackle these problems, and it cannot start early enough – it should start from an early age and be lifelong.

Stephen Turner, Head of Public Policy, Government and Philanthropy at Twitter thanked Thordis Elva for the "reality check" she provided through her presentation. He reflected on how his own ICT education was very skill-based, and how necessary it is to take into account communication and interaction online.

Young people use Twitter quite differently from older people, and Twitter tries to keep the public conversation free of abuse, without silencing any voices as a result of receiving hatred. Twitter is trying more proactively to take the responsibility off the individual. He said that more accounts are being reported and removed, but Twitter is also trying to minimise individual users' exposure to malicious content and hate speech. For example, changes were made in 2017, such as notification filters and safe search, which aim to find that harmful one per cent of content and keeping it away from users.
The most important progress Twitter is currently seeing relates to the improvements made to the way in which the company communicates with people who violate the user agreements. Since the initiative commenced, among this group of users, 65 per cent have not violated the rules again. This goes to show that building a knowledge base and raising awareness work, as do providing education and resources for educators, and collaborating with other online services within the ICT Coalition.

Thomas Myrup Kristensen, Managing Director of EU Affairs at Facebook, said prevention and first aid are important, but education remains key. Facebook takes the fight against online violence seriously; it is embedded in the platform's Community Standards. Facebook is currently experimenting with the right way of tackling these issues, relying both on humans and artificial intelligence (AI). However, it is not possible for one company to do it all on their own, the whole community needs to be involved. Thomas Myrup Kristensen mentioned the example of the community helping to build features such as muting messages, or super-blocking people.

Thordis Elva replied that these testimonies create the hope that technology is evolving, but in order for such tools to work, the community has to have faith in such technologies. She stated that, currently, the numbers are not very promising, especially for females.

After these presentations, a Q&A session ensued, facilitated by June Lowery, Head of Unit, Accessibility, Multilingualism and Safer Internet at DG CONNECT. The panel was asked whether industry players communicate with one another to remove content across services. Stephen Turner replied that Twitter has been building up more collaboration, it is formalised for some topics – terrorist content and child sexual exploitation – but for others, such as harassment and safety, it needs to be improved. All platforms are different in their approaches, and each user is specific in its use of the service. However, this is evolving positively, with consultations between YouTube, Twitter and Facebook; while it can be difficult to involve smaller services, this is the next step. Thomas Myrup Kristensen added that there is no central database for illegal content, although it does exist for terrorism. Facebook also trains smaller companies to remove content in an adequate and efficient manner. Facebook has an interest for people to feel safe, as they want them to communicate freely on Facebook. Reporting is only one tool; AI has the potential to help so much. Frida agreed that harmful content is everywhere; it cannot be stopped, despite significant education efforts being made in Finland.

June Lowery then asked the panel whether the industry have an idea of why people engage in violent online behaviour. Stephen replied that it can be difficult to see patterns, especially for a platform that covers a broad variety of different regional contexts. However, some are evident, such as mob harassment. The panel was then asked whether platforms would consider blocking sexualised keywords from being sent to children and whether this would work across languages. Thomas Myrup Kristensen replied that Facebook has all languages covered and they do look at specific slurs, trying to figure out which ones cross the line – these are then detected automatically.

June Lowery then asked Thordis Elva about women who lose their confidence in the early teenage years, and how she builds their confidence in her workshops. Thordis Elva replied that there is no simple answer, but that she underlines that everyone has equal worth. She highlighted the fact that it is no coincidence that women's confidence levels drop at the exact age when they start being sexualised by the outside world. It is important to empower them to be active participants in a democratic society.

June Lowery thanked all panellists and concluded the session.

For more information about the Safer Internet Forum 2019 "From online violence to digital respect", you can read the full report on betterinternetforkids.eu and visit betterinternetforkids.eu/sif2019.

Related news