Missed the event? Watch the video recording here


‘Online abuse is just part of the job if you are a journalist’. It is a throwaway phrase many have heard in their newsrooms, coupled by notions that some areas of reporting like crime or war require thick skin.

We also associate online abuse with very public, ‘face-of-the-news’ reporters: investigative and political journalists – especially women – have historically been the targets. And while reporting on violence is already highly traumatising, reporters in many other areas, including local journalism, are increasingly subjected to abuse. This has a very real impact on their work, as well as mental health.

Regional journalists have become vocal about the torrents of abuse they regularly face, with many editors speaking out about the issue. Others have taken more public measures, like the Yorkshire Post’s Call It Out Campaign, which acknowledges that while social media is a great tool for connecting with devoted readers, it is also an avenue for trolls and harassment. The campaign calls for readers to report abuse to social media platforms, or where necessary, to the police, providing instructions on how to do so.

Samantha Harman, the Oxford Mail editor, went as far as to survey her peers, finding that four in five journalists say abuse has become worse since the start of their careers. The survey goes on to say that Facebook is the most common space for abuse, and Harman is working with UK government body, the Department for Digital, Culture, Media and Sport, to tackle online abuse and harassment of journalists.

This was after she publicly tweeted an offensive email sent in by a reader. Harman clarified this was not the entire email because the rest was “more disgusting” and most emails are worse.



There is no single approach that social media platforms and governments have in place to protect journalists from abuse right now. Until there is some consensus on how to fight online harassment, journalists need to be equipped to deal with it and newsrooms need to create the culture, support mechanisms and policies to step in when it all gets too much.

The problem gets worse when you think about freelance journalists who are more isolated from the newsroom, and journalists who are working from home at the moment who feel similarly detached.

At a panel this week (6 October 2020) hosted by the International Press Institute, experts from different backgrounds discussed what strategies news organisations could implement to protect journalists in all parts of the industry.

Acknowledging the problem

Viktorya Vilk is the program director, digital safety and free expression for PEN America, a member-based organisation working to promote press freedom issues, with an emphasis on online abuse that leads to self-censorship and mental health problems. It does this through research, training and workshops, and other resources and guides for self-care.

The very first step, Vilk said, is to encourage staff to come forward about the abuse they face. Staff will not do this if they feel there will be negative professional consequences as a result.

A good way to do this is to offer a mixture of formal and informal channels they can action, added Sofia Diogo Mateus, audience development editor for POLITICO Europe. This is vital for newsrooms with a mixture of cultures, as some people may feel that an email process makes too much of a scene, or talking it over a coffee seems too casual. There would be options from there to receive external and internal support.

What proved efficient in the past when she worked for German broadcaster Deutsche Welle (DW) is to raise this early during the recruitment process, especially if journalists work on the audience side.

“We prepare them for the fact that, unfortunately, there will be a level of harassment or at least comments that are not pleasant to look at,” she says, adding that this is an opportunity to offer the candidates all the various areas of support and assure them of full job security if they need to take them up.

“I found in last two to three years it’s conducive to staff coming forward about emails, [online] comments or not being able to cope and needing time off.”

Vilk also suggested sending out anonymous surveys to staff just to gauge the scope of the problem internally and what staff may need. Diogo Mateus agreed, saying that it can be a useful way to signal to higher-ups new structures or protocols.

Message received

When a reporter is caught in the middle of a Twitter storm, what they need in that moment is not to feel isolated. They could have a full range of emotions at that time – from panic to fury – so whatever state of mind they are in, they must know where to go.

The New York Times (NYT) has a single, dedicated team for anything from going on dangerous assignments, to lost property, to online harassment.

“Don’t leave it to the employee to know who to contact,” says Jason Reich, vice president of corporate security, NYT.

“We make it very clear that there isn’t a specific definition of threats and harassment. There’s no bar for being under and over, because everyone’s bar is in a different place. If it makes you uncomfortable, just let us know, we’re well-staffed and we’ll address it.”

NYT can afford to have a ticketing solution where the emails are prioritised by a specialist team. Smaller organisations could take the same approach simply with a different department like legal, human resources or IT.

Vilk added that it could even be a Slack channel or other private platforms. That way, for small organisations especially, the burden does not fall to just one person.

From a legal standpoint, Reich said he can understand the temptation not to respond immediately, but the priority is to assure the staff member that the problem is in the company’s hands. NYT aims to respond within an hour but smaller companies could realistically do this quicker.

Analysing the threat

What NYT has learned the hard way is to make the protocol invisible to the staff member. In other words, not make responses seem like auto-responses you would get on an emergency call. The responder is, of course, risk assessing for immediate threat, but doing so in a way that is conversational and human.

“Don’t make it seem like they are ticking boxes or reading from a list. If someone is being harassed, be that sexual or vulgar, they just want to have communication.”

In the event of the flag being raised, the security team does need to assess how real the threat is. For recurring abuse, Reich stressed the importance of threats not being looked at as isolated events but an ongoing picture. This means if you have in-house security teams, it needs to be the same person handling the situation. If you are going to the police, you must insist on the same investigator handling the case.

“If you can identify an issue of having a security risk, that ‘This is nasty but this person isn’t going to hurt you’, then you have tackled most of the problem,” he explains.

“But you can’t do a meaningful threat assessment, unless you [can look at] multiple threats over time.”

During her time with DW, Diogo Mateus said the company had a protocol to take staff offline and away from the comments if the situation was deemed potentially dangerous. If it did pose a personal threat, the security team prioritised the safety of the person and protected them as appropriate.

Basic safety is also communicated to all staff, which covers two-step authentication for social accounts and reminding them not to post anything which would give away their location, as to make it easier to be found physically.

Prioritise those most in need

If you are short of resources, it could be worth channelling those to the people who need them the most. Vilk said that in the US, it is women of colour who are disproportionally affected by perpetual abuse.

“Make sure they have the most support because people talk to each other and see how colleagues are treated – ad hoc support is not enough,” she explains.

Vilk is more specific for a US context; ensuring they have affordable mental health care and any reservations they may have, ensuring they have legal counsel even for the peace of mind, and escalating the issue with social media platforms with particularly egregious episodes of abuse.

“If we are having serious conversations in the US and elsewhere around the world about having newsrooms which are more equitable and inclusive, if you don’t support those who are most targeted, these will be the ones who will be shoved out of the newsroom and traumatised over time.”

Modelling down

Staff should have go-to resources available during a moment of need. But it is not always that simple as to supply some links and call it job done. Leaders must lead by example.

“Resources from leadership down is a very good signal that you take something seriously and people can come forward,” says Vilk.

“If there’s nothing in place, people won’t come forward. Model it at the top and then permeate all the way through, so that people will feel they will not be dismissed, mocked or told to get over it, but taken seriously.”

She added that leaders can also benefit from guidance themselves.

“Managers don’t always know how to be supportive when someone is traumatised or stressed.”

Peer-to-peer support

The issue with existing mental health resources, Reich said, is that they are not geared specifically towards online harassment and abuse. Many would rather speak to colleagues who have shared experiences, which is where peer-to-peer support is useful. While this is not a substitute for therapy, it can help put staff on the path to getting the help they need.

He and Vilk agree that Reuters is the “gold standard” of peer support networks. It is managed by a third party which provides professional training to staff and regulatory structure. This is beneficial when a staff member would rather speak to a colleague than a professional. Reuters has 48 journalists internationally making up its network.

Reuters says in part: “Reuters established the Peer Network in 2015 as an additional resource within its larger Global Trauma Program. The Global Trauma Program is centred around CiC, a London-based company of licensed clinicians that contracts with Reuters to provide 24/7 therapy services to Reuters staff and stringers struggling with work/life stress, anxiety, depression and trauma – this includes stress and trauma related to online harassment.

“Reuters makes it clear that Peers are not professional counselors, but Peers undergo training in active listening and self-care through the therapists at CiC. CiC also closely supervises to the Peer Network to ensure that Peers are responding appropriately and do not feel overburdened. The Peers are all volunteers who are vetted through a rigorous application process run by CiC. The Peers also sign a statement of principles, which addresses the issue of confidentiality and is taken very seriously.”

When to respond

We have all heard the phrase “do not feed the trolls”. Largely speaking, it is better to not respond to abuse because that can fan the flames.

“Countering almost always backfires,” adds Vilk. “Trolls want you to say something abusive back and to get fired.”

DW had also come up with a threshold for intervening, says Diogo Mateuswhich was ‘less of a science and more of an art form’. She is wary of amplifying false narratives, but where there is a very clear abuse of the journalist, the newsroom is prepared to rally behind its staff.

The tricky part, Reich said, is conveying to the recipient of the abuse that sometimes not responding is a strategy not just to protect the company but the individual too. Certainly against individual trolls, NYT would avoid response. It is when groups and organisations enter the discussion, it would consider making a statement in support of the journalist.

Counter-speech, however, is far from straightforward and he urged everybody to avoid doing it out of emotion, and only when there is a very clear and logical strategy. Trolls do not want to engage in discussion, they want to provoke a reaction. It is crucial to understand that countering is not about exchanging ideas but sending a message to other social media users.

“Trolls are not operating under the same playbook. Counter-speech is effective for the third person, for people watching the exchange happen and to recognise this is not acceptable,” Reich explains.

“‘We’re not going to validate the troll by acknowledging the specific claim’. The point is to stand behind the journalist and the cost of good journalism. You’re not rebutting an argument, you are supporting a person.”


When to moderate comments

“We cannot be in the business of going on offence, we play defence,” says Reich. This analogy means that NYT, even in the face of abuse, does not believe in limiting free speech through comment moderation. The approach is to foster resilience and trust in the support system NYT has created.

“The best scenario is they can brush it off their shoulder and get back to work.”

But Vilk said that robust and effective comment moderation signals to reporters and sources that you take harassment seriously. Social media platforms can block certain keywords and mitigate some of the damage but sophisticated trolls will still find ways to write around it.

Diogo Mateus said social media platforms still have more work to do in this department, and laws around it are not silver bullets either. A controversial German law, the Network Enforcement Act (known as NetzDG) passed three years ago, and it gave social media platforms 72 hours to remove incendiary comments. It was criticised for limiting free speech and removing comments which it should not have while providing no way for them to be retrieved. The law is currently being revised.

Peaceful resolutions

This is not to say your options are getting into a yelling match or being silent on the issue. Turning abusive hashtags upside-down is a positive strategy seen in the wider media.

Diogo Mateus also said that accusations of ‘fake news’ or ‘bias’ can be countered by simply providing a link to the story where the news organisation has in fact covered the issue extensively on all sides and reported the facts.

She added that sometimes you can acknowledge trolls which keep re-spawning under similar variations. Eventually, her team in the past started to directly address them by saying ‘Nice to see you again here’, as a signal to other users that this person is not to be taken seriously.

Beware of self-censorship

Reich made the point that, if pushed too far, abuse can contribute to an undercurrent of fear around taking assignments on certain topics.

“We might be moving in a direction where the harassment can have a greater impact because of the reputation that it has,” he suggests.

“We can’t have it be that the only people that want to cover far right movements are ones that have the stomach for it, that’s not [an ideal] arrangement because that brings us to the beginning where it is ‘just part of the job’.”

Further resources

When it comes down to it, journalists and newsrooms cannot rely on legislative and technical solutions just yet. They do need to be equipped to handle the current situation. Here are some resources to help:

Online harassment manual (PEN)

Measures for newsrooms and journalists to address online harassment (IPI)

Upcoming free webinars on online abuse (PEN)

Online abuse support for freelance journalists (IPI)

Safety of female journalists online (OSCE)