Death and rape threats, accusations of treason, disfiguring memes, bots and automated agents, “black” public relation firms and hacking – these are only some of the features of state-sponsored trolling that have been used by governments around the world in recent years to attack journalists and other critics online.

This digital-age phenomenon was investigated in a new report published yesterday by the Institute for the Future (IFTF), a non-profit strategic futures organization, in co-operation with the International Press Institute (IPI), a global network of editors, media executives and leading journalists, and Global Voices, a citizen journalism outfit, and featuring original research from both organizations.

The report, State-Sponsored Trolling: How Governments Are Deploying Disinformation as Part of Broader Digital Harassment Campaigns, outlined comprehensively the features of state-sponsored trolling.

In the report, state-sponsored trolling is defined as online hate and harassment campaigns that are being used to intimidate and silence government critics, such as journalists. These campaigns combine disinformation and online harassment to defend state interests. Through the use of ordinary Internet users as well as the involvement of amateur and professional “cyber militias”, they are frequently camouflaged to appear “organic”. The report, however, emphasized the need to reveal the involvement of states so as to reduce the attacks’ harmful effects on democratic institutions.

The report compiled fifteen case studies of state-sponsored trolling campaigns in Azerbaijan, Bahrain, Ecuador, The Philippines, Turkey, Venezuela, and the United States.

In the case of Turkey, IPI provided research on three cases of state-sponsored trolling against journalists, conducted by using quantitative social network analyses of the attacks as well as interviews with affected journalists.

The analysis of the Turkish campaigns showed how state-sponsored trolling in the country gradually grew more sophisticated over the years. It also highlighted various factors that point to state involvement, including direct involvement and participation of high-profile politicians.

BBC World Service correspondent Selin Girit was subject to trolling after her reporting from the Gezi Park protests in 2013. The campaign against her was started by the then-mayor of Ankara, Melih Gökçek, a member of the ruling AKP party. On Twitter, Gökçek called on his followers to spread the campaign. During the attack, Girit received numerous rape and death threats.

Three years later, in the case of Nevşin Mengü, a television correspondent for CNN Türk, the links to government were much subtler. Mengü angered pro-government partisans with her coverage on the attempted coup in Turkey in 2016. The result was a state-sponsored trolling attack that lasted over a week.

The attacks on Mengü represented a refinement of the attack pattern, with government-linked accounts prompting the campaign by identifying the target and message but leaving the work of producing abusive and threatening message to a well-trained group of acolytes.

IPI’s research on state-sponsored online harassment of journalists was also featured yesterday in a report by Bloomberg, which highlighted the mechanics of an online “hate mob” against jailed Turkish-Kurdish journalist Nedim Türfent. The Bloomberg report also included IPI’s coverage of online attacks on the media in Austria, with a focus on the re-run of the 2016 presidential elections there.

Besides detailed examples of state-sponsored trolling campaigns, the report also offers a series of recommendations for solving this problem by implementing new policies for states as well as for technology companies.

As it may take long for legal changes to effectively stem the practice of state-sponsored trolling in the short term, the report suggested that tech companies have a responsibility to take action now. Tech companies, according to the report, could help fight state-sponsored trolling by detecting and identifying state-linked accounts and bots and by improving reporting mechanisms and responsiveness.