“Voices from the Island”: Informational annexation of Crimea and transformations of journalistic practices

Abstract

After the annexation of Crimea by Russia in March 2014, the peninsula experienced a progressive transition of telecommunication and broadcasting infrastructure under Russian influence, followed by a wave of repression of Ukrainian media. Between 2014 and 2015, dozens of Ukrainian media organizations and independent journalists left the peninsula to continue working in exile. This paper explores the phenomenon of informational annexation using a mixed methods approach consisting of in-depth interviews with media and IT professionals as well as digital ethnography and network measurements. It argues that, besides pressure from pro-Russian authorities, journalistic work in the area is challenged by legal and infrastructural factors such as the absence of legal and financial protections for Ukrainian journalists traveling to Crimea, lack of holistic digital security within media organizations, and increased Internet censorship in Crimea. By analyzing the risk perceptions and digital security practices of exiled and Crimean civic journalists, this paper explores how informational annexation challenges journalistic work on the infrastructural and organizational level, enabling the rise of civic journalism, and how it affects journalists' individual digital security practices. In the context of the current Russian invasion of Ukraine, this research provides insights into some of the informational annexation tactics used by Russians in the occupied Ukrainian territories.

Introduction

Crimea, a peninsula located south of Ukraine and south-west of Russia, has been dependent on the Ukrainian mainland for supplies ranging from water and gas to electricity and communications. While Russia took over cable television in Crimea in March 2014 - using television broadcasts to generate support for actions in Crimea (Iasiello, 2017) - the occupation of Internet infrastructure has been much slower, and the effects of the informational annexation on Crimean informational space properly studied. Up until 2014, access for Crimeans to the rest of the Internet was predominantly handled through Ukrainian networks, held to Ukrainian law and oversight. However, in March of 2014, Russia’s Internet regulatory framework, with its legal and technical constraints including surveillance and online censorship, was required to be applied to Crimean networks. It took more than 3 years for Russia to completely take over Crimean Internet traffic.

Following this transition, Crimean media was heavily impacted on economic, legal and infrastructural levels. A wave of persecutions against journalists in 2014–2015 led to a massive exile of Ukrainian journalists from the peninsula. Moreover, new regulations pertaining to access to Crimean territory from both Russian and Ukrainian sides have deepened the informational isolation of the disputed peninsula.

Though the fieldwork for this paper was conducted before the full-scale Russian invasion of Ukraine in February 2022, some of its findings can help understand the “Russian way” of informational occupation currently being deployed in temporarily occupied Ukrainian territories such as the district of Kherson. Indeed, some of the technologies described further in this paper are currently being used on the occupied Ukrainian territories to maintain Russian informational influence and control of Internet traffic and media content (for instance, in Kherson the Crimean Internet service provider Miranda-Media was responsible for hijacking Internet traffic in May 2022).

Transformations of Crimean media in the context of post-annexation have been approached from different perspectives in recent literature. A few studies focus on the framing of the Crimean crisis by Russian and Ukrainian media (Krutikova et al., 2019), as well as on methods and means of propaganda and information operations in Crimean and Russian media (Iasiello, 2017; Ishсhuk, 2019). Zeveleva (2019, 2020) uses Bourdieu’s field theory to analyze how the Russian state imposes limits on the autonomy of news media fields during political transition, and how Crimean journalists use self-censorship in order to continue their professional activity in a new political context. Others explore legal and geopolitical aspects of Crimean annexation and subsequent attacks on freedom of the press from the perspective of international relations, international law and policy or Eastern European studies (Berg and Mölder, 2018; Coynash and Charron, 2019; Marxsen C, 2016; Bilych et al., 2015). In a more recent publication Voronova (2020) looks at how migrant Crimean journalists engage with the mediation of nationhood in conflict situations and how journalists imagine their audiences in their search for relocatory trajectories. However, no systematic academic studies have been conducted on the digital security practices of Crimean journalists in exile, nor on Internet censorship in Crimea.

This paper argues that, in order to understand the effects of annexation on hybrid media professionals (Pantti MK, 2016), it is necessary to analyze not only the individual “adaptive strategies” of exiled or displaced journalists (as in Zeveleva, 2019), but also the background sociotechnical dynamics, the “battle for infrastructures” (DeNardis, 2012) enabled by the parties involved in the conflict. We argue that media researchers must pay attention to legal and infrastructural factors when researching journalistic practices in war-torn areas. This paper analyzes how the “governance by infrastructures” of Crimean Internet and media (DeNardis and Musiani, 2014) affects the work of displaced journalists and reporters covering events in Crimea, the new digital threats they are facing, and the strategies they adopt in order to minimize risks associated with their work in this new sociotechnical context. Besides a more classical approach in journalist studies that analyzes individual practices of journalists, we argue that a thorough analysis of the subsequent infrastructural processes is needed in order to properly assess the impact of armed conflicts on journalistic work and security. We approach post-annexation journalism as an experimental practice “in-the-making” that is largely shaped by technological constraints (such as content blocking or restrictions) and affected by circumvention and digital security tools that appear as a necessity in the context of an armed conflict.

To grasp this process, we propose to introduce the notion of “informational annexation”. We understand “informational annexation” as a set of measures to control informational flows in a given territory that unfold on two analytically distinct layers: the infrastructural and the content layer. In our research we focused on both layers, analyzing first the infrastructural transitions in the occupied Crimea (ownership of cables, equipment, cell towers and other material bases of Internet infrastructure; quantitative analysis of traffic going from and to Crimea; and measurements of connectivity between providers within and outside Crimea) and second, access to online content (media websites, blogs, social networks) and means of circumvention of imposed content censorship. The term “informational annexation” refers to a broader framework of research on so-called “information control” (Deibert and Rohozinski, 2010), the term used to describe a variety of practices applied mostly by state-level actors to control access and circulation of information, as well as the infrastructure behind it, mostly for sociopolitical, geopolitical and economic reasons.

Following the Citizen Lab’s mixed methods approach, this paper combines ethnographic study with network measurements and quantitative analysis of Internet censorship and traffic dependencies. It is based on a 1 year field study (2017–2018) of information controls in the region of Crimea, followed by online observations and digital ethnography ¹ and desk research (2022) to update some of the findings in the context of the Russian invasion of Ukraine in 2022. In-depth semi-structured interviews ² were conducted with 45 persons, such as exiled journalists (15), NGO workers (15), digital security trainers (10), politicians (5) whose work is related to the Crimean region. ³ The respondents have all moved to live and work on the Ukrainian mainland and therefore can be classified as “internally displaced” (Voronova, 2020). Some of them used to regularly go to Crimea as reporters before the war. Now all of these journalists are refused entry to the peninsula.

Network measurements were conducted using the OONI probe application in the Ukrainian mainland, Russia and Crimea with the help of local testers in order to analyze how online media has been censored in Crimea. An analysis of historical BGP routing data from RIPE archives between 2013 and 2018 was conducted in collaboration with the Internet Health Report project ⁴ in order to track the evolution of dependency of Crimean Internet traffic on Russia as compared to Ukraine. Finally, digital ethnography was conducted on a corpus of selected forums and chat rooms of Crimean media and IT workers, in addition to analysis of “grey literature” such as NGO reports and relevant legal texts.

The paper is structured as follows. First, we analyze Internet censorship in Crimea and its consequences on the work of journalists, their access to information from Crimea and to their audiences in Crimea. Second, we analyze the consequent “informational isolation,” the rise of civic journalism and its role within the informational annexation. Third, we analyze risk perceptions and threats to exiled and civic journalists and highlight some of the self-defense strategies deployed by the journalists. We argue that the rise of freelance journalism and “journalistic activism” is the result of Crimean “state of exception” (Coynash and Charron, 2019) and the lack of protection from established Ukrainian and Russian media. While access to Crimea is limited by bureaucratic procedures from both Ukrainian and Russian sides, and reporters face new risks both digital and operational, communication and permanent contacts with local activists and an informal network of colleagues (including Russian independent journalists) help minimize these risks and continue gathering information from the annexed Crimea.

The “traffic wars”: effects of internet filtering on Crimean informational space

Before the annexation, local Crimean Internet service providers used Ukrainian upstream traffic and relied on the fiber optic cable infrastructures owned by Datagroup and Atracom, Ukrainian nation-wide telecommunication operators. Unlike in Russia, Crimean ISPs did not have special equipment for lawful interception and traffic filtering. Administrative censorship, mainly focused on copyright infringement cases, was applied occasionally. No state-regulated censorship mechanisms were enforced.

However, after the beginning of the armed conflict between Russia and Ukraine, blocking of websites was introduced progressively as a measure of information control, within the general doctrine of Ukrainian Cybersecurity policy. The first websites to be blocked by Ukrainian providers were Russian social networks, search engines and email services (such as Yandex or Mail.ru), widely used in Ukraine. Soon after, other blacklists were issued by the Ukrainian Bureau of Security (SBU), which included websites and online news media of the so-called Donetsk and Luhansk “Peoples’ Republics”, as well as websites and official media of the new Russian administration of Crimea.

According to our interviews with Crimean ISPs in exile, Russian takeover of Crimean Internet infrastructures happened progressively, following a “soft substitution” scenario: after the first underseas cable was built under the Kerch Strait in March 2014, Crimeans lived in what we call a “routing interregnum” for almost 3 years. ⁵ Transitions from Ukrainian to Russian administration of communication infrastructures relied on background legal and commercial negotiations, including creation of new legal entities for Ukrainian telecommunication companies, still using the same equipment and base stations. An exiled journalist comments:

“Many experts in Telecom noticed that very soon after the occupation of MTS base stations a new operator arrived, the so-called Win Mobile. It uses the same frequency […] and most likely same equipment. It seems it was not just an occupation, there were also some negotiations with Ukrainian commercial structures. Vodafone representatives pretended they were not part of it, saying something like “We don’t know what is happening, how can they take over our property…” but at the same time all this looked very slow and calm, and most probably some arrangements were made” [J5].

This particular character of Crimean informational annexation can be explained by strong dependencies of Crimea on Ukrainian infrastructure, which would make it impossible for the Russian Federation to substitute all necessary services at once without having an important period of disruptions and shutdowns. During that “routing interregnum” period, Crimeans had different experiences of Internet censorship depending on their providers. Journalists working in Crimea had to switch ISPs and SIM cards in order to find the least censored network. When Ukrainian SIM-cards were deactivated in August 2014, communications between Crimeans and the Ukrainian mainland became complicated and more expensive. As J1, an exiled journalist from Crimea explains, this transition resulted in temporary disruptions of communication between Crimeans and their family members, especially with older generations who were heavily relying on land telephony for daily communications: “My father has lost connection to his sister who lives in Odesa [Ukrainian mainland], because 1 day SIM-cards just stopped working” [J1].

Moreover, the controversial status of Crimea from the perspective of international law ⁶ and the resulting economic sanctions imposed by the US and European Union led to the creation of a “grey zone” where many international and Russian telecommunication companies refused to work officially, deploying complex techno-legal arrangements in order to continue providing connectivity to Crimea. Between 2014 and 2018, Crimeans were using mobile prefixes allocated to Krasnodar and were considered by the telecommunication operators as clients from the Krasnodar region in order to avoid sanctions. Since 2018, big Russian operators such as MTS have applied additional roaming fees for Crimeans. Interestingly, regardless of the Russian Federation’s claims on Crimean territory, telecommunication companies are concerned about international sanctions and restrain from operating under their “real” brand names on the territory of the peninsula. This makes the Crimean case comparable to other internationally disputed territories where international sanctions are applied, such as Transdniestria or Nagorny Karabakh.

Since the annexation, Crimea has experienced several Internet blackouts, some of them being orchestrated for political reasons. Our analysis of social networks, forums and the Telegram chatrooms of Ukrainian ISPs revealed several important blackout episodes that demonstrate the fragility of the Crimean infrastructure and the capacity of both the pro-Russian government and pro-Ukrainian activists to use blackouts as a means of information control and information warfare.

The “routing interregnum” period lasted for more than 3 years, ending after May 31, 2017, when users of a Crimean ISP, Telekommunikatsionnye systemy, reported having problems accessing Vk.com, Mail.ru or Yandex.ru. A blockpage in Ukrainian explained that the site was blocked upon the decision of Ukrainian president Petro Poroshenko. This incident made it obvious that many Crimean ISPs were still getting upstream traffic from Ukrainian operators. A wave of searches happened at the offices of Ukrainian ISPs accused of “collaborating with the occupants,” and on July 12, 2017 Ukrainian traffic to Crimea was officially stopped. Crimea transitioned under Russian Internet regulation, with two major upstream operators, Miranda-Media and UMLC, providing traffic from the Russian mainland using two cables, the Kerch Strait one and the newly built cable over the “Energetic bridge”.

However, despite the centralization of traffic sources, censorship in Crimea was applied inconsistently (at least during our fieldwork for this study, 2017–2018). Our quantitative findings resulting from censorship measurements using OONI mobile application were also confirmed by interviews. Several interviewees, including a RIPE NCC officer responsible for Eastern Europe and Asia regions, indicate that Crimea has a separate blocklist that “depends on local administration” [I5] and on the decisions of Crimean governmental institutions:

General blocklists come from RosKomNadzor, but some URLs are communicated in emails or letters from the Ministry of Transport and Communications [of Crimea]. We can try to argue with them and not block, or we can block. From the point of view of our business, there’s no sense to argue with them. […] Locals want us to block [I8].

This “state of exception” in Crimean Internet censorship is also confirmed by journalists and activists who witness extra-judicial blockings of many exiled news media, and notice inconsistent behavior across providers:

One activist said that [they] could not open certain websites. We looked at it, and saw that [the access to] it was different everywhere. ⁷ It all looks DIY. ISPs behave in different ways. Sometimes there are explanations and blockpages, sometimes nothing at all, some websites are partly blocked, for example Krym.Realii – just some urls and some articles are blocked [A3].

To deepen our analysis of Crimean media censorship, we conducted research on media censorship during presidential elections (between February and April 2018) using the OONI application. We have seen that, on average, 25% more websites were blocked in Crimea compared to Russia, including websites that have never been included in official governmental blocklists (Valentovitch and Ermoshina, 2019).

The most censored content is related to Crimean Tatar media websites. Crimean Tatars, indigenous to the peninsula since the 13^th century, made up the majority of the population until the end of 19^th century. Today, however, they constitute just 12% of the area’s population. Most of the media outlets of this minority are critical of the Kremlin’s policies in the region and support restoration of Ukrainian authority on the peninsula. The Crimean Tatar community was most actively opposed to the annexation in its first months, and had a strong role in promoting an energetic and economic boycott of Crimea-an idea considered as extremist in Russia (Coynash and Charron, 2019).

A special campaign was launched by Crimean Human Rights Group in November 2018 to monitor blocked media in Crimea, with help from local testers. ⁸ This experiment, which was was repeated in 2019 ⁹ and 2020, ¹⁰ showed that censorship increased between 2018 and 2020. A similar study was conducted by Ukrainian NGO Digital Security Lab Ukraine focused on cybersecurity research and training for NGOs and NCOs. For these organizations, monitoring Internet censorship became part of the strategy against the informational annexation, in order to maintain communications with and from Crimea:

We defend freedom of speech and freedom of expression online, and this is one of the key components, the accessibility of sources. On the one hand, journalists have problems collecting information in Crimea, and on the other hand, they have problems with access to their Crimean audiences. Because most of the exiled media, who had to leave the region and establish in the mainland Ukraine, write about Crimea but are blocked in Crimea. So we have started to monitor these exiled media first, and then included other outlets in the list and we discovered that many Ukrainian resources were also blocked [A6].

Media organizations whose websites are blocked in Crimea use various tactics to circumvent censorship and deliver information to their Crimean audiences. Telegram and Facebook are widely used by blocked media to deliver information to Crimea. The shift towards social media as the main means for broadcasting changed media content production and promotion practices, with a stronger focus on shorter texts, live streams and multimedia. However, informational “bridging” of Crimea to Ukrainian mainland media does not solely rely on circumvention tools deployed by content providers, but also requires educating Crimean audiences. Blocked Ukrainian media and Human Rights organizations promote the usage of VPNs, and some even use advertisement billboards at the checkpoints to promote VPNs among those entering Crimea from Southern Ukraine.

Censorship had an important impact not only on Crimeans’ access to Ukrainian content, but also the quality of Internet connection as well as its price because Crimean ISPs were required by Russian Internet regulations to install expensive filtering and surveillance equipment on their networks. In 2018, expenses for filtering software and hardware equipment were estimated at between 15 and 20 billion rubles. ¹¹ The monopoly on upstream traffic by two operators created an unfavorable situation for the ISP market, leading to the disappearance of around 30% of Crimean ISPs and contributing to the degradation of connectivity, in turn affecting civic journalists, human rights observers and especially bloggers and streamers who document political and social events on the peninsula.

Even though the end of the “routing interregnum” period was rather abrupt, our quantitative analysis of traffic dependencies showed that until May 2017 Ukrainian landline cables were still actively used by Crimean providers. The “unplugging” of Crimea from Ukraine turned out to be a relatively slow process. The long period of “routing interregnum” created, for more than 3 years, a peculiar infrastructural context where communication with and from Crimea was subject to frequent interferences, blackouts and disruptions. Moreover, the so-called “geoblocking” ¹² due to US and EU sanctions made several important services, such as Google Play, G-Suite or App Store, inaccessible for Crimeans. This impacted the access of local journalists and activists to privacy-enhancing apps, such as Signal, and added complexity to an already sophisticated techno-legal set up.

From a peninsula to an “informational island”: media exile and the rise of civic journalism

In her study of state control of contemporary Crimean journalism, Zeveleva (2019) analyzed several strategies employed by the Russian state in order to progressively gain control over Crimean media, such as direct violence against journalists, ownership appropriation, relocation of capital, and disenfranchising some outlets while favoring others. However, as our survey shows, Crimean media was never completely free from pro-Russian influences. Even before the annexation, the Ukrainian media presence in Crimea was said to be “just formal” (J12) and Crimean media were described as “flirting” with Russian-language audiences and promoting pro-Russian ideas ranging from supporting free trade with Russia to clearly appealing for “unification” with Russia. Russian-language media content was dominant in Crimea, and the pressure against independent journalists has always been higher in Crimea than on the Ukrainian mainland.

In Crimean pro-Russian media, the “language of war” rapidly developed following the annexation, targeting certain groups such as “Ukrainians,” “Crimean Tatars”, and sexual minorities. Ukrainian Human Rights groups perceive hate speech as one of the key means of informational annexation. A study conducted by Crimean Human Rights group ¹³ shows that hate speech was used widely on Crimean TV-channels as well as on the websites of new Crimean authorities.

The annexation led to a massive exile of Ukrainian media from Crimea. During the first wave in 2014, approximately 15 Crimean media organizations left to the Ukrainian mainland with their teams and equipment. The second wave corresponds to the end of the so-called “transition period,” a timeframe of roughly 1 year given to Crimean media organizations, telecommunication services and Internet Service Providers to change their legal statuses and obtain new Russian certificates. At the end of the transition period, licenses were not renewed for many Ukrainian media organizations, who then had to leave. After this second wave of exile in mid-2015, independent media organizations in Crimea almost disappeared:

“Certain journals have stayed (after annexation), for example “Sevastopolskaya gazeta”, “Slava Sevastopola”, and new projects were opened by rich Russians as part of the propaganda campaign […] Criticism in media is possible but only when it concerns local problems or when it’s between the elites, but not the criticism of the very fact of annexation” (J3).

As a result, the peninsula was transformed into a sort of “informational bubble” (to quote one of our respondents, J3), vulnerable to Russian propaganda. As J3 describes it, “there is no serious independent journalism in Crimea now, people are writing about birds and butterflies… My old friends started thinking and operating categories used by pro-Russian propaganda 3 years after the annexation.” As Coynash and Charron (2019: 36) point out, Crimean “media outlets were instructed to include a minimum of one story about Crimea per day, accentuating how conditions were improving and that people were happy to have become Russian citizens.” As J4 puts it, international journalism is also problematic, as it is much harder for foreign reporters to get to “the truth.”

Access to Crimea via the North of Ukraine has been impossible since 24^th February 2022, but even before the full-scale invasion it was strictly regulated. A special permit had to be issued by the Ukrainian Ministry of Temporary Occupied Territories and the Migration Service, which could take up to a few months. All journalists and human rights defenders had to cross Ukrainian and then Russian checkpoints. Crossing a Russian checkpoint involved interrogations, control and temporary seizure of devices. Moreover, Ukrainian journalists needed a valid press card from a media organization accredited in Russia, which is to be obtained through the Russian Ministry of Foreign Affairs. To circumvent this paperwork, curious collaborations between Ukrainian journalists and independent Russian media developed on an ad hoc basis:

No media in Ukraine gave me a press-card … they do not want to take responsibility for me when I am in Crimea. Then I tried another way. I wrote to Russia, to about 10 different liberal media, who are against the regime. I said, give me your Russian press-card, so I can work as a Russian reporter, and I will not have to ask the Ministry of Foreign Affairs. I will be your reporter. All liberal media in Russia refused. Only one small media group agreed [J6].

All interviewed human rights activists and journalists mentioned a general feeling of isolation mixed with financial difficulties and complicated paperwork. The feeling of unjustified risk is shared by all interviewed journalists who are still traveling to the peninsula:

“Very few journalists work on the topic of Crimea and the annexation […] Sometimes when I have made an investigation and prepared an article, I don’t know where to publish it. The Ukrainian media is not prioritizing Crimea at all. Some editors propose 1000 hryvnia (35 euro) ¹⁴ for an article about arrests of Crimean Tatar activists, but it involves weeks of work and lots of risk” [J8].

As the Russian war against Ukraine influenced the topics that are more highly demanded and better paid for, Crimea turned out to not be a priority for Ukrainian media, as our interviews and media analysis show: “We have conducted a monitoring in autumn [2017], analyzing 10 Ukrainian News TV channels: only 2% of air time is dedicated to Crimea. But if we take 6 biggest Ukrainian TV channels, such as 1+1, ICTV, Inter – only 0.4% of air time is dedicated to Crimea” [P3].

Interviewed journalists criticized the informational politics of the Ukrainian government in relation to Crimea, as it disrupts their access to information and lowers the quality of journalism:

We have a real problem with finding objective information from Crimea. Our Ministry of Information and Ministry of Occupied territories are not doing their job to help people from Crimea broadcast information here, and to bring more Ukrainian journalists there [J3].

However, some important steps were made toward developing Ukrainian broadcasting in the territory of Crimea. In August 2018, the Ministry of Information of Ukraine announced they had developed the Strategy of Informational Reintegration of Crimea. ¹⁵ In 2017, a radio tower was installed on the Ukrainian border with four different radios broadcasting in Ukrainian, Crimean Tatar and Russian.

One strategy developed by Ukrainian journalists in response to the crisis of institutional support was to become freelancers. All interviewed journalists who still traveled to Crimea in the times of the study were freelancers and work for many different media outlets. According to the interviewees, freelancing offers more independence in the choice of topics to cover, providing a possibility to circumvent some bureaucratic constraints and conduct longer field trips. The role of freelance reporters in conflict zones has been recently discussed by Høiby and Ottosen (2019) who state that “freelancers carry the burden of getting facts from the frontline to the public because, for safety and financial reasons, traditional Western media avoid sending their own reporters to the frontline”. Indeed, interviewed freelance journalists share a feeling of insecurity, as Ukrainian editors do not offer proper legal protection for reporters who go to hot spots such as Crimea or Eastern Ukraine:

“The main problem is that when I am in Crimea, I am like one of the most wanted journalists for all these media. They take everything that I send to them. But the editor is not ready to take responsibility for you. You’re told openly: ‘Work for us, we’re happy to publish your texts, but if something happens to you, sorry for that. We can help to spread information about you. But we won’t engage in any kind of legal work to release you” [J6].

Lack of governmental or institutional support has changed the way freelance journalists perceive their work. Some of them have almost become “activists,” while human rights defenders, in turn, have engaged in important reporting and analytical writing from the field. The border between the two groups has become blurred (Budivska and Orlova, 2017), as this interviewee, a member of a prominent Ukrainian NGO focused on Crimea, witnesses:

I am myself an ex-journalist, I was writing a lot about human rights problems, and at some point I understood that, while we can freely talk about these problems in Ukrainian media, they are still not solved […]. At some point I crossed the border of journalistic work and became a human rights defender [A6].

During the first year after annexation, many Ukrainian NGOs were still operating in Crimea and were making regular reports about human rights violations in the region:

In March 2014 there were many cases of violent kidnapping of journalists and activists in Crimea. People were held in the basement, tortured. At the same time, Crimean journalists, our friends and colleagues, were leaving the region. So we, the NGO people, had to go there and document the situation. We created Crimean Field Mission for Human Rights together with Russian colleagues, and for 1.5 years, until summer 2015, CFMHR was the only permanent international monitoring organization in Crimea that was documenting and collecting information, and transferring everything to us. We analyzed it, made reports, and informed international structures. This was important, as no international monitoring mission could go there [A6].

Ukraine developed the so-called “stop list” to monitor everyone who entered Crimea through the Russian side without a permit from the Ukrainian Ministry of Temporarily Occupied Territories. This made work in the region risky for both Russian and Ukrainian NGOs. As a result, only a few journalists and human rights defenders were still going to Crimea from the Ukrainian mainland, sometimes as volunteers:

People who work in Crimea now are enthusiasts, who risk their health and safety but they can not sell their texts. A person organizes a risky trip, spends their own money to pay for it, and this money never returns to them, because they will be able to sell their text for 100 dollars at best. So people just abandon this or on the contrary become really engaged in the topic of Crimea, so they basically become volunteers [J4].

The relative success of this new kind of reporter, working almost without any protection, is due to their connection with local activists and civic journalists who provide first-hand information, arrange interviews, explain the context, and act as fixers and guides. In their survey of a hundred journalists working in conflict zones, Høiby and Ottosen (2019) show how risks faced by journalists and reduced access to conflict zones result in more journalistic coverage based on second-hand observations. In the case of Crimea, “second-hand” coverage has been actively used by the internally displaced Crimean media outlets now operating from Kyiv. An informal network of activists, informants, and even Russian independent journalists has been developing to help maintain reporting from Crimea to the Ukrainian mainland. Displaced journalists actively monitor Facebook pages of their old colleagues, friends or local activists groups in search of news. They monitor the websites of Crimean pro-Russian authorities, and attempt to verify this information using either their old networks (colleagues, friends, family) or new local activist groups such as “Crimean Solidarity,” an NGO created to provide first-hand information on cases of Crimean Tatar political prisoners.

Crimean Solidarity members organize video streams from court hearings, hold protest actions, gatherings and rallies, and provide informational, legal and financial support to the Crimean Tatar community, the most persecuted ethnic group on the peninsula. However, streamers are constantly threatened. A Crimean Human Rights Group report ¹⁶ mentions at least three administrative cases against streamers in just the second half of 2017. In the context of full-scale invasion of Ukraine, persecutions of the members of Crimean Solidarity have intensified, including arrest of prominent journalist Vilen Temerianov, accused of terrorism.

The informational annexation has created what we call an “asymmetric” scenario, as displaced journalists, reporters and local activists have different risk perceptions of digital and operational threats. The next part of this paper analyzes these different models and their effects on the production and diffusion of media content.

Crisis-driven security culture: risk perceptions and digital self-defense practices of Crimean journalists

Our interviews took place during the period between December 2017 and May 2018, almost 4 years after the beginning of the Russian -war against Ukraine. This temporal distance influences how interviewed journalists describe their attitude to professional digital and operational risks. When they recall their work and life during those first years following the eruption of the conflict, they tend to critically evaluate the changes in their usages of ICT and the evolution of self-defense workflows over time. The analysis of these accounts helps detect an interesting scenario: first, the conflict acts as a bifurcation point and provokes a wave of digital security-oriented training for journalists and NGOs across Ukraine. However, after a certain period of time, we observe a decrease of vigilance on the one hand and changes in the formats of digital security training for journalists on the other. Instead of serving as an “instant” remedy, the training is now oriented to build long-term organizational capacities in Crimea.

For the majority of our respondents the Maidan revolution of 2014 was an important bifurcation point, or a dislocation moment, “a moment when the subject’s mode of being is experienced as disrupted” (Glynos and Howarth, 2007: 110). This moment boosted their interest in digital and operational security. Others became concerned with digital security during the first months of the annexation of Crimea. While all interviewed journalists and activists have attended digital security trainings, what has impacted them more was their personal experience of a device seizure, phishing or other threats, or the experience of a colleague or friend:

“People do not understand the threats before something really bad happens. They don’t understand that even the most insignificant information can be interesting for someone” [J4].

During the first weeks of Crimean annexation journalists and activists shared a certain “state of shock” due to the general crisis in the country, and digital self-defense was not on the agenda. Ukrainian journalists did not have prior experience working in high-risk situations, and media organizations were not ready to provide support and lacked solid organizational security policies:

“A Ukrainian journalist before 2014 was not used to working in uncomfortable situations […] You go to a press-conference, you go to the parliament, if sometimes you do some fieldwork somewhere, you feel like you are ready for a Pulitzer prize. And then we have Crimea, and the war, and we are like… what’s going on? We are like… ok, we will go film the “men in green. ¹⁷ Then “men in green” come and hit your head because you film them. […] Everything was boiling and moving fast… In 2014 journalists were kidnapped, kept in the basement, held as hostages by some marginal militia-type gangs… No one was really understanding what was happening” [J6].

According to our interviews, the security culture among journalists has developed mainly following the “learning by doing” scenario. Proper security trainings were popularized only after a certain time:

In the beginning it was all experience-based. Because no one, and it’s very important, no one was ready for digital security in Ukraine, no one knew how to cover armed conflicts. I would say that everything I and many of my colleagues now know came from experience, our own or that of our colleagues, who happened to be in certain risky situations. And after international organizations came to Ukraine to make digital security trainings [J1].

So far, all respondents have attended a digital security training at least once. However, apart from exiled journalists working in international media with Western funding that have the capacity to help regional chapters maintain their infrastructure and keep up-to-date with informational security, no other interviewed journalist mentioned any holistic information security policy within their organization. As a result, their digital security knowledge is based on rather fragmented bits of information from trainings, advice from “tech-savvy” friends or online guides:

We have only a few journalists who fully understand what social responsibility of journalism means. So journalists are actually working by themselves, they have to think about their security. Editorials are rarely protecting journalists in this sense. Journalists are by themselves, they have no money, no time, no resources. That’s why it’s chaos [...] Overall, the situation is bad [J5].

Bigger media organizations with international funding have a more holistic centralized security policy that implies a “bureaucratization of security,” as J5 puts it. This means that a certain set of software (not necessarily the most secure – for instance, Outlook and Dropbox) is recommended for employees. Our interviewees mention a ‘conflict’ between these centralized organizational policies and their own DIY security practices. For instance, J5, an exiled Crimean journalist working in Kyiv, has separate email addresses that are not related to their work email provider. However, due to the security protocol of their media organization, they cannot configure an Outlook account for this pseudonymized email account.

Other “side effects” of this kind of organizational security policy are, for example, usage of specific software solutions for conference calls decided by foreign sponsors, for example, Skype, which is not a particularly secure tool. Employees may disagree with the choice of these applications, and often have their personal laptops as alternatives, which leads to multiplication of vulnerable points and uncontrolled copies of data, such as interview recordings. In general, we have observed that freelance journalists have more advanced digital security knowledge compared to the average level of organizational security.

In the absence of a holistic organizational policy, a figure of “digital security champion” appears as a new intermediary that helps propagate digital security culture across communities of journalists and activists. It is usually a member of staff, or a person in an informal network of journalists or activists, who is personally concerned by digital security and has a more advanced knowledge of privacy-enhancing technologies. These “champions” would take a lead on disseminating best practices in their network of colleagues.

For Crimean Tatar communities, the IT-champions are often recruited from school teachers, doctors or other members of local “intelligentsia” who already have a certain reputation in the community. As trainers underline, “Crimean Tatar communities are quite hermetic, it is hard to inspire trust in these communities. However, some civil society organizations such as [anonymized] have credit among Crimean Tatar. So it becomes possible to influence activists of “Crimean solidarity”, and through them, influence larger Crimean Tatar communities”. [T3]

Predominance of freelance or civic journalism creates new kinds of threat models. These journalists work in asymmetric scenarios where their own self-defense strategies do not always correspond with strategies of their sources or colleagues. Digital security trainers ¹⁸ say it is harder for them to provide appropriate audit and risk assessment to freelancers on a case by case level. In contrast, NGO activists in general have more developed and holistic organizational security policy, more capacity and resources for long-term digital and physical security trainings, including development of tailored roadmaps and protocols for those working in Crimea and those staying on the mainland:

When someone goes somewhere, we have a contact person in the organization who has all passwords, numbers of family and close friends, whom to contact in case of emergency, copies of IDs and so on. And we keep connections all the time. If folks are going to Crimea, we are doing it all the time, a few times a day, to see if everything is fine. [She] is writing to me every time – I have left the flat, I meet with X or Y there and there [A2].

In this asymmetric scenario, people present on Crimean territory become the weakest part of the communication network. The biggest threat is not online, but physical. Digital security trainers working with Crimean activists and journalists insist that “Crimean government does not have a capacity to survey using advanced technologies. They are using most stupid offline ways” [T2]. Physical pressure, device seizure or even kidnapping become the most frequent threats to local and traveling journalists:

Occupied territories are a grey zone for the media, it’s hard to work there, journalists often have problems and can be arrested. For example a journalist from [Ukrainian media] was kidnapped and kept as a hostage while in Crimea. There was totally no rule of law in Crimea at that time, everything was destroyed in the office of the journal [J3]

One of the interviewed journalists was kidnapped twice by the FSB and was subject to interrogations after trying to cover court hearings against a group of Crimean Tatar activists. Reporters who still travel to Crimea complain of permanent offline surveillance. They are subject to frequent ID checks, verification procedures; they are refused entrance to court hearings, official meetings of administration and other events.

“It was not a direct pressure, but psychological pressure. It was a constant observation, constant feeling that these people are present around you, and they control you. When you are experienced, you can easily disclose these people. You go to a cafe, you see this person, and he is just sitting there drinking juice for three hours, eating, looking around, observing. And also constant checks, ID checks, coming to your hotel room to “verify additional information”, and so on” [J1].

Physical force and tortures are often used by the Russian Federal Security Office (FSB) to get access to passwords. This undermines trust in the usage of encrypted communication tools (“they will get my password anyway”). In response, Crimean activists and journalists avoid storing sensitive information locally, prefer cloud storage or seek offline face-to-face communications with sources when possible. The practices of “ephemeral” identities, multiple devices, compartmentalization and pseudonymization are preferred. Our survey shows that journalists do not always trust devices provided by their employers, and prefer procuring their own device, often a cheaper phone and laptop for missions in Crimea (such as an old Windows computer, a Nokia 3310, or an older Android phone). The usage of downgraded devices creates additional vulnerabilities. Because of the risks related to device seizure, reporters often prefer to bring as few devices as possible.

Risk perceptions are different between field reporters and exiled journalists working in the news rooms on the Ukrainian mainland. Exiled Crimean journalists perceive themselves as being physically in a “safe space” and thus often tend to fall back from more secure to more convenient means of communication:

I have stopped encrypting my files and devices, because this software was on the old laptop which I was using for my travels. And after, when I moved to work to Germany, I stopped using it, I did not need it, and never took it away from home [...] I do not need to contact people who are located there [J1].

They estimate that the Russian government will not intercept the traffic between them and their sources in Crimea (even though the SORM system of lawful interception of traffic and phone communications has to be installed on all networks operating under Russian regulation). Moreover, Crimean exiled journalists need to communicate with family and friends who still live on the peninsula, and the choice of tools is often driven by families. In the time of the study, they still used Vk.com and other insecure social media platforms, unencrypted phone calls. Crimean journalists’ attention to digital security decreased over time as they moved from more secure to more convenient tools:

“I now share news on Facebook messenger. I do not use any kinds of, I don’t know, secure tunnels, or Signal, or these kinds of tools. But initially in 2014-2015 I forced myself to study and use all this, even in 2016 still I was like, conspiracy, informational security...” [J6].

Interviewed journalists say that they feel “alone” facing risks, and claim that their personal knowledge of digital security does not improve communication with colleagues, friends or sources. Field reporters share the need to “take a break”, and underline that constant stress related to working in high-risk areas lowers vigilance and attention to digital security hygiene.

Conclusion

This paper introduces the term of informational annexation to analyze the transition of Crimean media and the Internet under Russian control. The multi-layered analysis of this process gives a framework to grasp the infrastructural changes that led to legal and technical isolation of Crimean Internet, and second, to explain the rise of the civic and freelance journalism in this particular context, and describe new digital security practices of these journalist groups. We argue that this methodological and theoretical frame can be used to analyze similar processes going on in the context of the full-scale invasion of Ukraine in 2022, as well as in other war-torn territories or disputed areas. The individual journalistic “coping strategies” and new tactics for information distribution should be analyzed in a broader context of “governance by infrastructures” as they are shaped by materialities of communication infrastructures. Journalism on occupied territories appears therefore as an experimental practice “in-the-making” mediated and affected by new tools of circumvention and self-defense.

The network of civic informants and freelance reporters continues operating despite lack of support from Ukrainian government or from established media organizations. It ensures coverage of main political events, such as persecution of Crimean Tatar activists, thanks to a decentralized and informal structure involving freelance reporters, human rights activists, friendly Russian independent media who offer their coverage and legal help, and, of course, local activists and civic journalists from Crimean Tatar community and other active communities pre-established in Crimea.

This specific configuration in which Crimean media operates creates new forms of digital and operational risks. Throughout the duration of Russian-Ukrainian conflict risk perceptions and the consequent self-defense practices of journalists have been considerably changing. The resulting security culture is described by several respondents as “crisis-driven” or “experience-based”, which means that practices of digital self-defense are not based on institutionalized support or systematic trainings, but on personal or second-hand experiences of threats and a patchwork of know-hows coming from different sources, including new intermediaries such as “IT-champions” or other informal sources.

With the full-scale invasion of Ukraine by the Russian Federation in February 2022, any official Ukrainian media activity in Crimea became impossible. Other Ukrainian territories in the meantime are experiencing similar processes of informational annexation, involving the actors that first appeared in the context of Crimean crisis (for instance, the Miranda-Media provider, responsible for rerouting traffic via Russia between May and November 2022 in Kherson and its neighboring areas). While this study did not include the most recent events, Crimean case can be seen as a sort of a “laboratory” of information control.

Since February 2022, in the context of an open phase of the war, the possibility of freelance and civic journalism on occupied areas, as explored in this article, is questioned. The threat models and digital security practices have also drastically intensified. And while the war in Ukraine is being told by thousands of civilian voices currently under Russian occupation, they remain anonymous and mediated by social media. Further research is needed in order to understand how informational annexation is unfolding in Ukrainian territories since February 2022.