The Russian disinformation threat and the EU response: the debate

2.2 Methodology

The theoretical foundation emphasising the importance of socially constructed ideas, or narratives, in international politics guides the paper's methodological choices. To reiterate, the primary focus is the performance of the Russian disinformation threat- narrative within the European Union institutions and the consequences this has on EU policy. To understand how this particular narrative is performing and how it remains unevenly accepted requires studying the policy positions of the relevant actors. Moreover, because narratives are not fixed but continuously challenged and rearticulated, understanding this dynamic requires identifying the `original narrative' and tracing the developments of institutional positions over time. For this purpose, for grasping how the European Union institutions construct objects such as the `Russian disinformation threat' and how these constructs evolve over time, ideas from content and discourse analysis are applied. To borrow from one discourse scholar, this paper treats the idea of Russian disinformation threat `not as truth', but as `one truth, held in place by language and power.' Parker, Ian (1992), p. 21

In practice, two questions structure the following discussion. The primary question is whether the Russian disinformation threat- narrative is successful in the European Union. As it is accepted as legitimate in some contexts and by some actors but not universally across the EU institutions, the secondary question asks what factors can help understand the uneven narrative performance.

For a particular narrative to be considered successful, it needs to be able to change the existing `belief systems' to an extent that the relevant actor starts seeing the subject topic `in a different light.' Freedman, Lawrence (2006), pp. 77-78 Put differently, a narrative is successful if it can convince actors, understood as either individuals or organisations, that the course of action it suggests is necessary in order to uphold their identity and legitimacy. Sandhu, Swaran (2009), p. 84 In the context of the European Union, the most important actors are the three institutional powers of the Union, namely the European Commission, the European Parliament and the European Council, i.e. the national governments of the Member States. Thus, any action from any of these bodies, or their parts, that can be perceived as being aligned with or promoting the Russian disinformation threat- narrative is interpreted as a signal of its success. Consequently, any action that seems to discredit the narrative, including the promotion of an alternative framing of events, signals the failure of the Russian disinformation threat- narrative.

In order to better assess the European Union institutions' alignment with the original narrative, attention is paid to the degree of institutionalisation of the EU's policy response. For this, the criteria developed by Swaran Sandhu are applied. According to Sandhu, the three most relevant indicators to consider in this regard are power, including `the access to the top-ranking decision-making platform', independence, i.e. `the autonomy of an organizational function' and the degree of `specialization and routinization'. Sandhu, Swaran (2009), p. 85 In the context of the European Union, these criteria are pitted against what arguably represents the EU's primary institutional response, namely, the East Stratcom Task Force established in 2015.

Following the logic of content analysis, tracing the developments in the Russian disinformation debate in the European Union is carried out by creating so called ideal types to categorise the competing narratives that can be identified in the studied material. Based on the analysis of the sources, presented in more detail in the final section of this chapter, four categories are sufficient to capture the most common discursive positions regarding `Russian disinformation threat' by the European Union institutions. These were based on the first reading of the sources and were later refined for the final edition of the paper; following the example of Mayring, Philipp (2000) and Hsieh, Hsiu-Fang; Shannon, Sarah (2005). These ideal types, together with an illustration of how they can be used to describe the situation at the end of 2017, are listed below.

The first type, the so called `original narrative', is the Russian disinformation threat- narrative. This narrative emphasises the strategic and hostile intentions behind Russia's activities. What exactly are considered as `disinformation activities' varies, but typically includes spreading false information through international broadcasting and social media.

Actors endorsing this narrative advocate for a wide set of policy responses to combat the threat to be implemented on both national and European level. These include both defensive and offensive measures taking place in both European and Russian information space. For example, limiting Russia's broadcasting activities in Europe and simultaneously expanding European broadcasting in Russia or for Russian-speaking audiences. If nothing is done, the narrative maintains, Russia is allowed to interfere in European political processes and European values, social order and, ultimately, sovereignty are at risk. Within this narrative, terms `disinformation', `misinformation', `fake news', `propaganda' and `information warfare' are used interchangeably to refer to the Russian threat. Although some of these terms such as `misinformation' do not necessarily suggest political motives on their own, in cases when they are used in connection to this narrative type, Russia's responsibility is emphasised in other ways. At the end of 2017, some factions within the European Parliament and, in principle, the European Council endorse this narrative type.

The second narrative type recognises `misinformation' and `fake news' as threats to the European Union but does not attribute these phenomena necessarily to Russia. Actors endorsing this narrative also support the idea of Union action but wish to see it targeting unattributed `fake news' rather than `disinformation' originating from a particular actor. `Misinformation' is accepted as a problem but without linking it to a greater strategic purpose.

The idea of a hostile third-party actor disseminating `fake news' purposefully is not present. In contexts where Russia's information activities are mentioned, which happens rarely, these are portrayed as normal state behaviour and framed in neutral terms such as `public diplomacy' or `soft power'. Most typically, labels `misinformation', `fake news', `false news' or `false information' are used in conjunction with this narrative. At the end of 2017, the European Commission, some factions in the European Parliament and some Member States seem to endorse this narrative.

The third narrative type recognises neither Russian `disinformation', unattributed `misinformation' nor `fake news' as threats that require a Union level response. Russia is presented as a partner to the European Union and the talk about `Russian disinformation' is portrayed as damaging to the interests of the Union, its Member States and its citizens. Actors endorsing this narrative suggest dropping the issue from the EU agenda and do not support a Union response. At the end of 2017, no European Union institution and no Member State publicly endorse this narrative type. However, some factions in the European Parliament and, according to secondary sources, within the Commission and Member States sympathise with this narrative.

Finally, the fourth group of actors neither publicly engages in the disinformation debate nor advocates for dropping the topic from discussion. In other words, they appear indifferent to the idea. Given that the focus of the study is on the contesting narrative articulations, this fourth group will not feature in the subsequent discussion. Moreover, none of the European Union institutions belongs to this category at the end of 2017. However, it can be a useful reference point if the reader is interested in studying the performance of the Russian disinformation threat- narrative amongst the 28 Member States at large. For an example of such endeavour, see Janda, Jakub et al (2017)

In principle, this paper uses these ideal narrative types in a straightforward manner. Every instance of an EU institution or other relevant actor publicly expressing a position regarding the Russian disinformation threat- narrative, either through a statement or a policy, is perceived through the framework of the ideal narrative types.

These findings are used for identifying the actors' initial discursive positions, tracing how they change over time and spotting the contesting ideas to the original narrative. In practice, apart from very few exceptions, most of the studied texts appear to be aligned with either the first or the second narrative type. Therefore, also most of the subsequent discussion focuses on these two narratives.

What then makes the Russian disinformation threat- narrative perform the way it does? It is commonly accepted that a narrative's effect is audience- dependent, that is, a narrative has a greater chance of success when it `appeals to the values, interests and prejudices of the intended audience'. Freedman, Lawrence (2006), p. 23; Holmstrцm, Miranda (2015), p. 119 For Miskimmon et al, in addition to the `informational and emotional content' of a narrative, its degree of ambiguity, relation to action, i.e. how well it explains the events actors experience, and processes of formation, projection and reception affect its performance. Miskimmon, Alister et al (2013), p. 111 On the other hand, Swaran Sandhu lists relevance, adequacy, applicability, appropriateness, consistency and coherence as the main characteristics determining narrative success. Schmidt, Vivien (2008), p. 311 In order to manage the scope of discussion, this study follows the example by Robert Hinck et al and uses coherence and fidelity as the main explanatory factors. Coherence refers to the consistency amongst the different articulations that are made to refer to the same idea, in this case the Russian disinformation threat- narrative. Fidelity `judges whether the audience will accept the story based on the people's beliefs and experiences', see Hinck, Robert et al (2018), p. 4, who build on Fisher, Walter (1984), (1989)

Secondly, given that success in narrative promotion is affected by the presence or absence of viable alternatives and the focus of the study is on contesting narratives, this aspect is included in the subsequent discussion. Miskimmon, Alister et al (2013), p. 141 Finally, one additional factor could be interesting in the case of the European Union. Following Swaran Sandhu, public organisations tend to mimic each other especially in conditions of uncertainty. Sandhu, Swaran (2009), p. 84 In the case of European debate on Russian disinformation, this dynamic could affect policy behaviour between the European Union and NATO on the regional level. Thus, some observations regarding this are offered.

Elaborating the chosen methodology requires sparing a thought on possible alternatives. In the process of developing the research framework, comparative methods, various case study as well as process-tracing approaches were considered. While there are examples of these and other tools being employed to study Russian disinformation, the focus on the `idea' of Russian disinformation threat and not `Russian disinformation threat' per se determined the chosen approach. It is hoped that the resulting work can be used by others to further our understanding on the European debate. For example, potential avenues for future research include identifying the specific variables explaining the variation in narrative performance or the detailed internal proceedings, conflicts and compromises leading to the adoption of specific text or policy. Had these examples been chosen as the focus of this study, both of which were considered, comparative methods or process-tracing would have likely been applied.

The paper at hand could also be interpreted as a `a theory-guided case study', in the words of Jack Levy. Levy, Jack (2008), pp. 4-5 As its `case', it has the relatively successful campaign of narrative promotion by a small group of Member States towards the European Union institutions.

However, the paper's primary objective of studying the dynamics of narrative contestation surrounding the Russian disinformation threat- narrative rather than the causal variables explaining the uneven narrative performance could lead other case study scholars to disagree. Since the space constrains prohibit a further debate on whether case study methodology must focus on causality or result in `constructing and testing generalizable theories', a judgment on this question is left for the reader. For example, see Levy, Jack (2008), p. 4 and Gerring John (2004)

A final note on methodology is inspired by the epistemological and ontological implications of studying socially constructed narratives. During the research process, the author has made countless choices of inclusion and exclusion of events as well as their presentation in the study. These choices, to at least some effect, stem from the cognitive shortcuts and other socially determined characteristics of the researcher rather than a careful research design. While measures have been taken to expose the resulting shortcomings, for example, by subjecting the final draft for reading by people from different backgrounds, presenting the European debate on Russian disinformation threat through the prism of strategic narrative contestation ultimately represents the researcher's attempt to capture and categorise reality. Therefore, the author recognises that different discursive constructions can be equally useful for studying the issue and wishes that the merits of each analytical approach are assessed against the premises they are built on.

2.3 Sources

The first time the Russian disinformation threat- narrative is publicly articulated on the European level with an accompanying demand for Union response is in January 2015. Thus, this forms the starting date for the scope of analysis. In terms of setting a final date, however, there is no obvious choice. Seminal texts are located unevenly across the three and a half years since January 2015 and, at the time of submission in May 2018, new developments occur on weekly basis. Given the lack of an obvious cut-off date based on the subject topic, the Gregorian calendar will do. Thus, the scope of the discussion is set from January 2015 to the end of 2017 including texts from the first three full years of the Russian disinformation threat- debate in the European Union.

In terms of selection of relevant texts, the choice is made easier by the institutional set-up of the European Union. In other words, the available European Council, European Commission and European Parliament decisions regarding disinformation and fake news are used to assess the evolution of these institutions' articulations regarding the Russian disinformation threat and their endorsements of the different narrative types. In terms of the European Council and its variations, this means the relevant Council conclusions, budgetary decisions as well as some other policy documents adopted by the European or Foreign Affairs Councils. In total, eight such texts are included. With regards to the European Commission, this includes various Commission Communications and Action Plans as well as relevant letters, website content and speeches connected to the debate. In total, 12 such texts are included. Finally, with regards to the European Parliament, this means relevant Parliament resolutions, which in one case includes the draft text and the tabled amendments. In total, three European Parliament resolutions are included.

Alongside the three main institutions of the Union, texts from additional four other types of actors are included. On the European level, these include the European Economic and Social Committee, from which there is one text, two think tanks affiliated with the European Union, from which there are four texts, and the political parties in the European Parliament, from which there are seven texts. While these materials are not as authoritative as the texts from the three main institutions mentioned first, they are illustrative of the Russian disinformation threat- narrative's legitimacy and can help understand a policy change on the institutional level. Finally, several EU Member States contribute to the disinformation debate by both individual and collective action throughout the period from 2015 to 2017. In total, 19 such cases are included in the study.

Chronologically, the total of 53 texts used in the study are split as follows: eight texts are from 2015, 11 from 2016 and 34 from 2017. From these numbers alone, some observations are possible. First, it appears the debate intensifies and becomes more mainstreamed across the different EU institutions towards 2017. This is visible in how often the topic is brought up as well as in the increasing number of actors engaging in the discussion. In terms of the participation of the three main Union institutions, it appears their behaviour differs from one another. While the European Council is the first one to discuss the Russian disinformation threat and appears to debate it at least once every year, the European Parliament and the European Commission become more engaged over time. By the end of 2017, the European Commission's engagement with the second narrative type, namely focusing on unattributed `fake news', exceeds the combined activity of both the European Council and the European Parliament. Finally, it is evident that the group of Member States engaged in the debate on Russian disinformation threat grows larger over time. However, this group still does not represent a majority amongst the Union members at the end of 2017.

To conclude the discussion on sources, limitations must be recognised. In general, much more material was studied than is included in the final paper and not all publicly available texts that could be relevant for the discussion are covered. The choice regarding what to exclude is guided by the advice from Jennifer Milliken, who suggests that adding texts to the analysis is justifiable only if it challenges the ideal types that the researcher has created for categorising the different discursive positions. Milliken, Jennifer (1999), p. 234 Given the focus of the study, an attempt has been made to capture all relevant texts from the three main EU institutions. In fact, with regards to the European Commission, the Parliament and the Council, nothing is intentionally omitted. In contrast, the chosen texts from think tanks and Member States represent a small sample of what is available. The two think tanks referenced in the discussion are chosen due to their affiliation with the European Union institutions and the assumption that, consequently, they affect EU decision-making proportionally more. In terms of Member States, given the focus of the study on EU level debate and policy response, the most relevant texts are those that either introduce the original narrative or its variations on the European agenda or, in other ways, steer the subsequent discussion. While interesting pieces are unavoidably left out, it is important to recognise that not all 28 Member States publicly engage in the debate on Russian disinformation threat in 2015-2017. Janda, Jakub et al (2017)

The final disclaimer concerns the general availability and quality of data. Given the political sensitivity of the subject topic, it is possible that the debate on Russian disinformation threat conducted behind closed doors differs from how it is presented to the European publics. For example, some European states rarely side with Russia in public against their fellow Union members, but nevertheless could object when possible punitive measures are discussed in a more private setting. On the other hand, the very prospect of gaining publicity could sometimes be politically useful and this could lead to over-zealous articulations that do not fully represent the held beliefs of the actor or actor coalition in question. In fact, both instances appear present in the material.

Given the European political system where the executive branch is subjected to public scrutiny and regular elections, however, policy and budgetary decisions are rarely taken without any public articulation and public opinion itself can be considered as a valuable resource. This seems to be the case also in the disinformation- debate: both the states promoting the Russian disinformation threat- narrative and the three main Union institutions engage in public narrative contestation. Given the resulting amount of available material, the public texts from EU institutions are used as primary evidence while secondary sources, for example citations from unattributed EU officials in newspaper articles, are used to complement the analysis when appropriate.

With these caveats in mind, the next chapter applies the abovementioned theoretical grounding, methodology and sources to trace the introduction, performance and contestation of the Russian disinformation threat- narrative in the European Union institutions in 2015-2017. The discussion is structured chronologically with a section dedicated for each year. Each section begins with an overview of the relevant texts and proceeds chronologically through the developments during that year. Finally, each section closes with an analysis of these developments from the perspective of strategic narrative contestation and narrative performance.

CHAPTER 3. DISCUSSION AND ANALYSIS

3.1 Developments in 2015

The Russian disinformation threat- narrative is first introduced in the European Union foreign and security policy discourse by the means of a public letter by a small group of Member States in January 2015. Other key texts from 2015 include Foreign Affairs and European Council conclusions from spring 2015, and European Parliament resolution on EU-Russia relations and EU Action Plan on Strategic Communications from summer 2015. Additionally, the April European Agenda on Security, May Information Report from the European Economic and Social Committee and November briefing from the European Parliamentary Research Service (EPRS) are used to trace the efforts to legitimise the Russian disinformation threat- narrative on the European Union policy agenda in 2015.

On 8 January 2015, Denmark, Estonia, Lithuania and United Kingdom introduce the Russian disinformation threat- narrative in the European Union foreign and security policy discourse. In a public letter addressed to the High Representative of the Union for Foreign and Security Policy and three other European Commissioners, the four countries accuse Russia of conducting a `disinformation and propaganda campaign' against the Union. Denmark et al (2015), please note that all citations in this paragraph are from the same text. Furthermore, they argue, Russia's activities `discredit EU narratives' and `erode support for legitimate Governments'. In response, the four states demand a `concerted and comprehensive effort' from the European Union institutions and national governments in four distinct areas. In terms of `awareness', they ask for regular high-level discussions on Russian disinformation threat at the EU level as well as EU-wide public awareness campaigns. In terms of `assertiveness', they ask the EU to do more to `tell truth, facts and deconstruct propaganda proactively'. This is also the area where the group foresees the establishment of a dedicated Strategic Communications task force. In terms of `alternatives', the four countries ask for the EU to invest in new `credible and competitive information alternatives' and in supporting the `existing credible Russian-language media outlets'. Finally, in terms of `accountability', they ask for the EU to support the work of media regulators to `draw attention to cases which manipulate, deceive, incite hatred or propagate war'.

In the following 29 January Foreign Affairs Council meeting, the foreign ministers of the 28 Member States debate the issue for the first time and consequently task the European Commission to `further improve [EU's] strategic communication' and `to explore options for the establishment of a dedicated communication team to lead these actions… including correcting misinformation.' EU Foreign Affairs Council (2015), all citations in this paragraph are from the same text. Evidently, the 29 January meeting is not an immediate success for the Russian disinformation threat- narrative. Although the Member State representatives recognise the possibility of institutionalising EU's response through a dedicated team, they do not commit to it nor mention `disinformation' or Russia as a threat specifically. Furthermore, although the conclusions do not elaborate what is meant by `misinformation', in its common usage, the term does not imply hostile political intentions behind the spread of false information.

The topic is debated again in March, this time by the 28 heads of states in the European Council. In contrast to the earlier outcome, the 20 March Council conclusions explicitly recognise the `need to challenge Russia's ongoing disinformation campaigns'. European Council (2015), unless otherwise indicated, all citations in this paragraph are from this text. This is a significant step towards the legitimisation of the Russian disinformation threat- narrative since the wording recognises both the intentionality behind the phenomenon and names the country responsible for it. However, why disinformation is a threat, what this term encompasses and what the appropriate Union response should be are not elaborated in this text. This task is delegated to the European Commission with Council mandating the Commission to `prepare by June an action plan on strategic communication'. While the March European Council meeting represents a success for the Russian disinformation threat- narrative, its relative novelty and low degree of institutionalisation at this stage is apparent from how it is not included in other closely related texts. For example, when the Commission publishes its communication for the European Agenda on Security on 28 April, it covers such threats as `terrorism, organised crime and cybercrime', but does not mention misinformation, disinformation or any other related concept. European Commission (2015a)

Despite the first narrative type being inconsistently present on the top-level discourse concerning European security in April 2015, the work for its promotion continues in other institutional contexts. On 19 May, the European Economic and Social Committee adopts an Information Memo with the title `How media is used to influence social and political processes in the EU and Eastern neighbouring countries'. European Economic and Social Committee (2015) The text is prepared by a team of nine members out of which three are Lithuanians, including the rapporteur. The memo identifies `heavily funded disinformation' and `Russian state-sponsored propaganda' as threats to the European Union and its Member States. European Economic and Social Committee (2015), p. 2 While this is not an authoritative policy document, the language in the Information Memo suggests coordinated action across the different European institutions amongst the actors promoting the Russian disinformation threat- narrative.

In contrast to the European Economic and Social Committee, a more heavy-weight institutional actor enters the Russian disinformation debate in summer 2015. In its resolution on `EU-Russia relations' adopted on 10 June, the European Parliament calls for `strengthened analytical and monitoring capabilities of Russian propaganda' and for `adequate funding for countering Russian propaganda and misinformation'. European Parliament (2015a), p. 7 The passing of the resolution with these references is important because despite its focus on broader EU-Russia relations, out of the resolution's 23 substantive paragraphs four discuss the threat caused by Russian `disinformation', `misinformation' and `propaganda'. In the plenary debate preceding the final vote, references to the Russian threat are made by the Commission representative as well as British, Dutch, Estonian, Swedish, Romanian and Lithuanian Members of the European Parliament (MEP). Also this document is prepared by a Lithuanian rapporteur. For the first time on record, one British and one Hungarian MEP openly challenge the use of such language with reference to Russia and, although unsuccessfully, attempt to remove it from the final text. The plenary debate is available online, see European Parliament (2015b)

Although the European Parliament resolutions on foreign policy are not legally binding, given the timing of this resolution just before the publication of the European Commission's Action Plan on Strategic Communication on 22 June, it is possible that it affects the Commission document. European Commission (2015b) In either case, the Action Plan transforms what was a policy discourse until this point into a policy practice. Drawing its legitimacy from the 20 March Council conclusions, in other words, `the need to challenge Russia's ongoing disinformation campaigns', it proposes a number of ways to do this. European Council (2015) To combat the threat, the Action Plan emphasises the need for a `positive narrative' regarding the European Union that `resonates with the target audience' and for increasing awareness of `disinformation activities amongst the general public'. European Commission (2015b), pp. 1, 3 For this, the action plan commits to setting up a dedicated East Stratcom Team `focused on proactive communication of EU policies and activities'. European Commission (2015b), p. 1 This initiative is in line with the January 8 letter, 29 January Foreign Affairs and 20 March European Council conclusions.

When reading the EU Action Plan from the perspective of the January 8 letter, the proposed initiatives match all the four areas of response demanded by the British, Danish, Estonian and Lithuanian governments. However, most of these are presented as continuation of EU's existing work. Even when this is not the case, the new initiatives are expressed vaguely without committing the Union to any specific actions or time-line in their implementation. The one exception is the establishment of the East Stratcom Team (later called Task Force) under the European External Action Service (EEAS). Given that many of the tasks listed in the 8 January letter, from raising awareness on the disinformation threat to debunking false information and supporting EU institutions in strategic communications, appear to be assigned to the newly found Task Force, its resources appear modest at best. Despite the broad mandate, the Task Force initially receives no funding from the EU budget relying instead on annually agreed voluntary contributions from interested Member States. European Parliamentary Research Service (2015b)

While there are no other significant developments in the Russian disinformation threat- debate on the European level during 2015, several Member States promote the narrative through various means. A good example of national policy measures aligned with the original narrative is the move from states such as Latvia, Lithuania and Poland to limit the broadcasting rights of Russian media operators through legislative interventions. NATO Parliamentary Assembly (2015), p. 10 Another initiative is to increase investment in Russian language media. In 2015, this approach is adopted by Dutch, Polish and German governments. Ibid., pp. 10-11 Notably, although the threat of `Russian disinformation' is already endorsed on the highest level in the European Union, i.e. in the European Council conclusions, also other terms are used. For example, in September, Poland hosts an international conference dedicated to countering `Russian propaganda'. EU Observer (2015) Later in October, following the publication of the MH17 report by the Dutch Safety Board implicating the use of a Russian anti-aircraft weapon not possessed by Ukrainian military, a finding Russia immediately disputes, prompts the president of Finland to state that an ongoing `information war' threatens the West. For the Dutch report, see Dutch Safety Board (2015); for the Finnish president reaction, see Niinistц, Sauli (2015)

Given the diverse articulations regarding the Russian disinformation threat, the final text from 2015 is the November European Parliamentary Research Service briefing titled as `Understanding Propaganda and Disinformation'. European Parliamentary Research Service (2015a) Although not an authoritative or political document, the briefing is interesting for its reflexivity regarding the terminology being used in the EU parlance. At first, the briefing notes the primacy of the term `disinformation' in the European Union, given that both European Council and European External Action Service (EEAS) use it to describe Russia's hostile information operations. Ibid., p. 6 This observation is supported by both the March Council conclusions, the June EU Action Plan as well as the East Stratcom Task Force's two weekly newsletters exposing and correcting Russian disinformation. These are named 'Disinformation Review' and `Disinformation Digest', see East StratCom Task Force (2015) However, the briefing recognises that the European Parliament is inconsistent in its discourse and uses terms 'propaganda', `misinformation' and `disinformation' interchangeably when referring to the same issue. This phenomenon is visible in the European Parliament resolutions on EU-Russia Relations and European Neighbourhood Policy, see European Parliament (2015a) and (2015c), respectively In addition to its reflexivity, the November briefing is notable for being the first EU publication attempting to define the commonly used terms and establishing a hierarchy amongst them. Given that it presents the term `disinformation' as the most authoritative, it further legitimises the Russian disinformation threat- narrative. However, it does this only partially given that, as a briefing note focusing on past developments, it does not call for further action.

Analysis of 2015. In overall, the texts from 2015 show that the Russian disinformation threat- narrative is successfully introduced in the EU foreign and security policy discourse on the highest level through a coordinated action by a small group of Member States. This is evident from the actions taken by the EU bodies that can be interpreted as being aligned with or promoting the Russian disinformation threat- narrative. These include the March European Council conclusions, which explicitly endorse the narrative, and the June Action Plan for Strategic Communications stemming directly from this endorsement. While the original narrative is not consistently present across all relevant texts, lacking completely in the European Agenda on Security, and there remains variation in its articulation, there seems to be no clear alternative narrative. Rather, the different terms such as `disinformation', `misinformation' and `propaganda' are not presented as mutually exclusive but seem to co-exist in same texts. Thus, in 2015, it appears there are no visible actions by EU bodies, or their parts, that openly discredit the narrative, including the promotion of an alternative framing of events.

With regards to the institutionalisation of the EU's response, Russian disinformation threat- narrative transforms from a policy discourse into policy practice in mid-2015 with the EU Action Plan on Strategic Communications and the subsequent establishment of the dedicated East Stratcom Task Force under EEAS. However underfunded and understaffed in comparison with its stated objectives, the very establishment of the nine-men team represents an increase in the institutionalisation of the Union response. Given that the Task Force did not exist in 2014, it obviously enjoys an increase in its power, independency and specialization and routinization of operations in 2015. However, the access the Task Force's staff enjoys to the top-ranking EU officials is difficult to assess from public documents. In terms of independence, while there are no signs suggesting its work is being directed by other actors, the Task Force does not have a single euro of its `own' funding and instead relies on annually negotiated national contributions, seconded staff and local volunteer networks. From the perspective of financial independence, these conditions provide an uncertain ground for planning long-term operations. In terms of specialization and routinization, the fact that the Task Force commits to publishing weekly reports from the very beginning suggests at least an intent to routinize the Task Force's work early on. It must be acknowledged, however, that these public reports likely represent only a portion of the Task Force's work and assessing the degree of specialization and routinization in other areas is difficult due to lack of public evidence.

In terms of understanding the performance of the Russian disinformation threat- narrative in 2015, several observations are possible. In terms of its coherence and fidelity, it can be concluded that the narrative is not entirely consistent and that different articulations co-exist on the European level. With regards to this variation, one particular aspect needs elaborating, namely, the use of the term `propaganda'.

This term, used by both European Parliament and the European Social and Economic Committee as well as some Member States, is a powerful textual marker and clearly resonates more with some audiences than others. While for some audiences it emphasises the malicious nature of Russia's activities even more than `disinformation', for others it is an alien word in the context of peace time EU-Russia relations. It appears these two groups are geographically concentrated, with eastern European states predominantly in the former. However, a thorough exploration of national differences is beyond the scope of this paper. Crucially, in terms of the European Union response, the uneven acceptance of the term `propaganda' by EU Member States is visible in how in institutional contexts requiring unanimity, such as the European Council, the term is absent. At the same time, the more flexible decision-making procedures of the European Parliament and the European Economic and Social Committee appear to allow for its wider usage.

In terms of potential mimicking behaviour between EU and NATO, there are signs that could indicate such a phenomenon taking place. First, it needs to be recognised that in NATO the debate on Russian `disinformation and propaganda' started already in 2014, sometimes in connection with the debate on `hybrid war', and sometimes independently from that. For example, see Vershbow, Alexander (2014) and Breedlove, Philip (2014) An important text in this regard is the report from the Sub-Committee on Democratic Governance of the Committee of Civil Dimension of Security of NATO Parliamentary Assembly from October 2015 titled as `The Battle for the Hearts and Minds: Countering Propaganda Attacks Against the Euro-Atlantic Community' and prepared by a Polish rapporteur. NATO Parliamentary Assembly (2015), p. 12 This text bears many similarities with the resolution adopted by the European Parliament 13 months later in November 2016.

Second, the NATO-run web portal titled as `NATO-Russia Relations: the Facts' resembles in many ways the EUvsDisinfo- website set up by the EU East Stratcom Task Force in autumn 2015, and precedes it by several months. Ibid., for the website, see NATO (2018) Finally, in summer 2014, Latvia, Estonia, Germany, Italy, Lithuania, Poland, and the United Kingdom initiate the establishment of the NATO Strategic Communications Centre of Excellence in Riga to `help the Alliance enhance its strategic communication capabilities'. NATO StratCom Centre of Excellence (2018) In terms of its operations and the participating countries, this centre resembles the European Centre of Excellence for Countering Hybrid Threats, a joint EU-NATO initiative founded in Helsinki almost three years later in autumn 2017. The European Centre of Excellence for Countering Hybrid Threats (2018)

It should be noted that although the push for Union response against Russian disinformation threat comes from certain national governments, it appears the governmental efforts are aligned with the efforts from these states' individual representatives in the European Parliament and the European Social and Economic Committee. As this discussion has shown, nationals from countries such as Latvia, Estonia, Poland, United Kingdom and Lithuania seem to be particularly active in this regard. In fact, Lithuania successfully introduce a variation of the narrative also on the agenda of the United Nations during its chairmanship of the UN Security Council in May 2015. For the UN Security Council resolution 2222, see United Nations (2015)

While Russia is not mentioned by name, the timing and the threat articulation as well as the initiating country of the resolution suggest coordination between national representatives for the same political purpose across different institutional contexts. In Miskimmon et al's words, the narrative promotion by these actors appears indeed `strategic'.

In conclusion, while 2015 sees the Russian disinformation threat- narrative being introduced in the EU security and foreign policy discourse, the subsequent EU response is nevertheless modest when contrasted with the 8 January letter's ambitions. Put differently, although the idea of the threat is endorsed in principle by the European Council, its severity and the appropriate Union response remain contested. Furthermore, despite the European Council's usage of the term `disinformation' in its communications, other actors, namely the European Parliament and some Member States, continue to use different terminology sometimes emphasising the hostile intentions behind Russia's activities and sometimes not.

3.2 Developments in 2016

The key texts from 2016 include the April EU Joint Framework on Countering Hybrid Threats, June EU Global Strategy, October European Council conclusions and November European Parliament resolution on `Anti-EU Propaganda by Third Parties'. While all four are important for tracing the further legitimisation of the Russian disinformation threat- narrative in the European Union in 2016, these texts expose existing and introduce more diversity in the debate. Finally, illustrating the growing interest by a growing number of stakeholders on the disinformation debate in comparison to the year before, relevant texts from one European political party, one EU think tank and several Member States are included in the discussion for 2016.

The late 2015 and early 2016 witness two developments with possible long-term repercussions to the disinformation debate in Europe. First, the Cologne attacks in December and the so-called Liza affair in January, both in Germany, draw in a major European state that had previously shown little public interest in the debate or a coordinated EU response. For details of these events and their effect, see The Economist (2016) The second major development is the Dutch consultative referendum on Ukrainian Association Agreement in April 2016. The result of the referendum attracts substantial speculation regarding the role of Russia's information operations and could explain the increased interest of the Dutch government on the topic in 2016. Applebaum, Anne (2016)

In light of these events, in its Communication on 6 April titled as `Joint Framework on Countering Hybrid Threats a European Union Response', the European Commission includes the threat of `massive disinformation campaigns'. European Commission (2016a), p. 2 In fact, the intentionality and political use of disinformation are strongly emphasised in the text by qualifying `disinformation' as `systemic' and emphasising its objective to `control the political narrative' and `destabilise European societies'. Ibid, p. 4

In response, the document calls for investing more in `strategic communications' and rebuffing disinformation. Ibid., pp. 4, 17 Moreover, the Joint Framework is notable for it explicitly recognises that many hybrid threats are cross-national and should therefore be tackled `by using EU policies and instruments.' Ibid., p. 2 Thus, although Russia is not mentioned by name, the emphasis on `systemic' and politically motivated `disinformation' and the call for a Union level response can be interpreted as legitimising the Russian disinformation threat- narrative.

Summer 2016 brings about several developments in the disinformation debate. The first one is the 19 May publication titled as `In-depth Analysis: EU Strategic Communications with a View to Counteracting Propaganda' commissioned by the European Parliament's Committee on Foreign Affairs and produced by European Union Institute for Security. Legrand, Jйrфme; Zampa, Ifigeneia (2016) Rather than labelling Russia's activities as `disinformation', the text frames them as part of a `strategic communications campaign' and a `soft power offensive'. Ibid., pp. 29, 13 While this publication represents a step away from the original narrative, the June EU Global Strategy prepared by the European Commission and adopted by the Member States specifically recognises `disinformation' as a threat and commits the Union to `enhancing its strategic communications'. European Commission (2016c), p. 23 However, similarly to the earlier Joint Framework, also in this document the threat remains unattributed and Russia is not mentioned.

In summer 2016, also the European Parliament and its different political factions start showing an increasing interest in engaging in the disinformation debate. The ALDE group, Alliance of Liberals and Democrats for Europe, is the first to act by organising a conference titled as `Kremlin lies: EU's response to disinformation' on 30 June.

The concluding statement calls `Russia's disinformation campaign' a `homeland security issue inside the EU' and urges the European Commission to scale up the Union's response. Alliance of Liberals and Democrats in Europe (2016) The three MEPs quoted in the statement come from Belgium, the Netherlands and Lithuania. Despite the fact that this is a standalone statement from one political group in the European Parliament without any direct impact on Union's activities, it is a clear endorsement of the Russian disinformation threat- narrative.

Moreover, the timing of the event could be connected to a particular resolution process that began in the European Parliament Foreign Affairs Committee in May and culminates in November.

The 23 November European Parliament Resolution titled as `EU Strategic Communication to Counteract Anti-EU Propaganda by Third Parties' is a seminal text for the Russian disinformation threat debate in the European Union. European Parliament (2016a) As the name suggests, it draws attention to hostile information operations conducted against the European Union by third parties, which it explicitly names two: Russia and the Islamic State (Daesh). With regards to the threat from Russia, it names the Russian `information, disinformation and misinformation campaigns', `modern hybrid warfare' and `hostile Kremlin propaganda' threatening the European Union. Ibid., pp. 4, 5, 10 Russia's English-language channels RT (Russia Today) and Sputnik are mentioned specifically as `instruments' that `challenge democratic values [and] divide Europe'. Ibid., p. 5 While the precise terminology used to describe the `disinformation threat' varies from one paragraph to another, Russia's `alternative narrative' that `erodes the European narrative' is presented as the core threat. Ibid., pp. 6-7; in fact, there was a suggestion to add definitions to the proposal, but it failed the vote.

The draft resolution put forward by the Committee of Foreign Affairs and prepared by a Polish rapporteur receives a total of 358 amendments out of which 244 deal with Russia. For all amendments, see European Parliament (2016b)

Out of these 244 amendments, almost one hundred accept the Russian disinformation threat- narrative in principle but elaborate how it is a threat to the Union and another 70 or so suggest what should be done about it. Of the remaining amendments, almost 50 contest partly or fully the Russian disinformation threat- narrative. It appears the aim of these amendments is to make the text less Russia-focused and more universally applicable, but it is difficult to identify a coherent alternative framing of events in them. In contrast, some 20 amendments clearly attempt to rearticulate the threat by shifting the text's focus to the misdeeds of the United States or the European Union in the information domain and by presenting Russia as a valuable partner. Of the 358 tabled amendments, 72 are about ISIS, 24 technical, 18 linguistic and the remaining 244 about Russia. Out of the 244 amendments, 94 accept the narrative but define it further while 73 suggest what should be done. From the remaining amendments, 48 challenge the Russian disinformation threat- narrative, but do not propose a coherent alternative interpretation while 17 put forward such a narrative. Finally, 12 amendments do not fit into these categories.