Becoming a “Normal” and “Ordinary” Organization through Strategic Communication? Discursive Legitimation of the Swedish Armed Forces

ABSTRACT This article examines the use of discursive legitimation strategies as a mode of strategic communication in public sector organizations. The study object is the Swedish Armed Forces (SAF). As a military organization, SAF’s communication has traditionally been characterized by restrictions, regulations, discretion, and secrecy. However, changing conditions have created a new need for legitimation of SAF and its operations, both internally and externally. The aim of the study is to understand how discursive legitimation is used in internal strategic communication in a public sector context, with particular emphasis on the challenges related to changing conditions and inconsistent demands on the organization. Discursive Legitimation Analysis (DLA) is used to study 31 editorials in SAF’s staff magazine to examine how SAF’s activities and its transformed role in society are legitimated. Two main findings are identified: (1) the frequent and unexpected use of rationalization as a legitimation strategy, stressing conformity with other organizations rather than differentiation; and (2) inconsistent legitimation strategies, with conflicting values and perspectives on strategic communication operating simultaneously. These findings highlight the difficulty of maintaining coherency and consistency in practice. The study further helps advancing the general understanding of the limits of strategic communication in pursuing legitimacy.

macro-oriented focus, and the main interest has been on how strategic communication can be used to gain legitimacy in the eyes of the public (e.g., Cawkwell, 2019;Waeraas, 2019). As shown by Deverell and Wagnsson (2016), however, the changing tasks and conditions of SAF have also affected its internal communication. More specifically, the aim of the present study is thus to address the knowledge gap regarding how discursive legitimation is used in internal strategic communication in a public sector context, with particular emphasis on the challenges related to inconsistent demands.
The article makes two main theoretical contributions. First, and contrary to previous research on discursive legitimation in public sector organizations, rationalization is identified as a very salient strategy for legitimation. Furthermore, this strategy of rationalization -where the SAF is described as a modern, complete, organizational structure -highlights conformity with other organizations rather than differentiation. This finding contributes to our understanding of the relationship between reputation and legitimation, which is a key question in the research on discursive legitimation in public sector organizations (Waeraas, 2019). Secondly, the study underscores the inconsistency of legitimation strategies, with conflicting values and perspectives of strategic communication operating simultaneously. This finding adds to previous research by stressing that managers in public organizations not only need to "pay attention" to different available legitimizing voices (Aggerholm & Thomsen, 2016, p. 205), but also need to embrace and engage with such inconsistencies (Scherer et al., 2013).

Recent transformations of the Swedish armed forces and implications for communication
Sweden has been a neutral -or "non-aligned" -country for over 200 years. It is thus responsible for its own defence and is not a member of any mutual defence alliances. However, Sweden participates in several international military operations through the UN and EU and is also a partner country in NATO (Agius, 2011;Hagelin & Wallensteen, 1992; MSB [The Swedish Civil Contingencies Agency], 2016).
Since the 1970s, SAF has undergone substantial changes, with large cutbacks in units, personnel and equipment, its mission being redefined from defending against an invasion to conducting foreign operations, and the former obligatory conscription being suspended (Agrell, 2011;Petersson, 2011). Lately, however, SAF has received several budget increases and its focus has again shifted to total defence. In December 2020, budgetary allocations were further increased by approximately 45% through 2025 (Regeringen [The Government of Sweden], 2020). Obligatory conscription, which was replaced by voluntary military service in 2010, was reinstated in 2018 (Regeringskansliet [Government Offices of Sweden], 2017).
With recurring transformations and new tasks and responsibilities putting pressure on the societal support for Swedish military organizations (e.g., Berndtsson et al., 2015), there has been an urgent need to engage and communicate with society to achieve legitimacy. Strategic communication is accordingly described in official documents as "a fundamental tool in Swedish Armed Forces' leadership which, when used properly, creates momentum in the change of SAF as well as in the operative work" (Försvarsmakten [The Swedish Armed Forces], 2011, p. 3, translated from Swedish original) and studies show a transition to more market-driven communication (Deverell et al., 2015;Deverell & Wagnsson, 2016), with legitimation of international operations being achieved through military blogs  and social media playing an integral part in SAF's image-building (Stoehrel, 2013).

Strategic communication
In its original sense, the word "strategy" refers to the use of military power to achieve political goals (Lyth, 2017). It is thus no wonder that strategic communication has always played an important role in the military context (M. Eriksson, 2011). Today, it continues to be an important defensive tool for states and other international actors to counter propaganda, hybrid warfare, fake news, and efforts to manipulate election results (Michelsen & Colley, 2019). Paul (2011), however, argues that there is a difference between the military (governmental) conceptualization of strategic communication and what he refers to as "the industry." The latter, he claims, uses the term in a much too broad and abstract way, in which anything and everything could be considered strategic communication. Paul instead defines strategic communication as "coordinated actions, messages, images, and other forms of signaling or engagement intended to inform, influence, or persuade selected audiences in support of national objectives" (p. 17). The key term here is "national," which suggests that strategic communication cannot be performed in support of the parochial goals of military (or any other governmental) organizations alone, but must be part of the grand national interests. Our understanding of strategic communication, however, is more in line with that of Hallahan et al. (2007). Strategic communication should hence be regarded as a transboundary concept and practice which focuses on organizations' intentional and purposeful communication to reach their goals, through various strategies of communication and meaning-making. Consequently, we assume that individual public sector organizations could also pursue their own organizational goals and not only those of the overriding state. Notably, SAF seems to share this more general definition of strategic communication, as can be seen, for instance, in its view of strategic communication as a tool to facilitate organizational change (Försvarsmakten [The Swedish Armed Forces], 2011). That is not to say, however, that the national objective of Swedish sovereignty is not also prominent in our data.
Although we have a slightly different understanding than Paul (2011), his argument raises the important question of where to draw the line between strategic communication and corporate communication. Van Riel and Fombrun (2007) define the aim of corporate communication as "creating favorable starting points with stakeholders" (p. 25), and effectively aligning the communication internally and externally. Christensen and Cornelissen (2011) add to that by clarifying that the "vision of contemporary corporate communication . . .. is to manage all communications under one banner" (p. 386). This concept thus seems to be even more broadly defined than strategic communication. Although elements of what some would consider corporate communication certainly can be found in the data of this article, such as communicating the uniqueness of SAF (cf. Christensen & Cornelissen, 2011), reputation management with the aim to stand out from the crowd is also considered an important feature of strategic communication (Waeraas, 2019). Furthermore, consistency and alignment, which are considered key features of corporate communication (e.g., Christensen & Cornelissen, 2011;Van Riel & Fombrun, 2007), are equally important components of strategic communication (e.g., Dimitriu, 2012;Pamment, 2015;Volk & Zerfass, 2018). The operative word for distinguishing between the two should thus be "strategy", we believe. According to Zerfass et al. (2018), one should "consider an issue as strategic when it becomes substantial or significant for an organization's or other entity's development, growth, identity, or survival" (p. 493). The changing conditions for SAF have presented challenges in all four of these areas, with communication being an important tool to tackle them. In distinguishing between the two forms of communication, it is also helpful to look at the word itself. "Strategy" has an all-encompassing, long-term character and concerns political goals such as maintaining or changing power relations (Falkheimer & Heide, 2014). While one could of course argue that there are also strategies behind corporate communication (Van Riel & Fombrun, 2007), with its focus on the formation of the organization as an entity (Jackson, 1987), this latter aspect is not necessarily included. Furthermore, legitimacy, which will be further developed in the following section, is considered not only an important purpose of strategic communication, but also a prerequisite for it (Aggerholm & Thomsen, 2016;Lischka, 2019).

Legitimation and strategic communication
In reviewing the literature on organizational legitimacy and discursive legitimation, our main argument is that, with a few significant exceptions (Aggerholm & Thomsen, 2016;Waeraas, 2019), the existing research tends to neglect public sector organizations. From these works, we know that discursive legitimation takes place on the microlevel of public sector organizations, and that different "voices" (i.e., strategies) are integrated within the legitimation process (Aggerholm & Thomsen, 2016). We also know that while public sector organizations are dependent on legitimacy, it can be a tricky matter to maintain a favourable reputation without jeopardizing legitimacy (Waeraas, 2019). However, as Waeraas points out, there are few examples in the literature of how public sector organizations manage this balancing act in practice. Furthermore, the clash between public sector organizations' need to be recognizable and understandable and their striving for differentiation "should be further examined in empirical research" (Ibid., p. 54). Building on this particular research strand, we thus argue that challenges related to inconsistent demands involved in the legitimation of such organizations need more attention. By studying these contradictory forces -and their divergent legitimation strategies -we can gain new understanding of how conflict of goals are handled in strategic communication.
Legitimacy deals with an organization's societal license to operate over time (e.g., Scherer & Palazzo, 2011), and it implies "a perceived congruence between the goals, actions, and values of an organization and those of the larger social system of which an organization is part" (Waeraas, 2019, p. 45). Legitimacy has been described as a "confusing construct" in management theory (Suddaby et al., 2017), and it has been seen as highly contested and elusive (Hurrelmann et al., 2007). In a thorough review of how legitimacy has been used in organization and management studies, Suddaby et al. (2017) distinguish between legitimacy as a property of organizations, as an interactive process, and as a socio-cognitive perception. The present study focuses on legitimacy as a process, stressing how legitimacy is a product of interactions and communication. Hence, the emphasis is on processes used to construct and maintain legitimacy (Ibid.). Furthermore, legitimacy is considered a socially constructed sense of appropriateness, in the sense that an organization's actions are considered "desirable, proper or appropriate within some socially constructed system of norms, values, beliefs, and definitions" (Suchman, 1995, p. 574). Therefore, legitimation can be seen as an inherent part of the stability and change of institutions (Vaara & Monin, 2010), as well as being linked with terms such as authority, power, and ideology.
Within the legitimacy-as-process perspective, Suddaby et al. (2017) identify a significant group of researchers interested in legitimation understood as a process of persuasion and influence grounded in language. This research strand is important for the present study, as meaning-making through language is seen as the basis of legitimation (e.g., Maguire & Hardy, 2009;Vaara & Tienari, 2011). This stream of research tends to focus on discursive legitimation (e.g., Glozer et al., 2019;Luyckx & Janssens, 2016;Vaara et al., 2006). Several studies within this tradition have documented how communication, rhetoric, and communication management (in various forms) are pivotal for legitimation (Elsbach & Sutton, 1992;Suddaby & Greenwood, 2005;Vaara & Monin, 2010). For instance, it is well documented how leaders or other organizational spokespersons establish legitimacy with a range of different strategies and tactics. Hardy and Phillips (1999) show how discursively shaped legitimacy is intimately related to stakeholder relations and identity constructions. Glozer et al. (2019) discuss discursive legitimation in a context of new media and interactivity, arguing that the interrelated and overlapping nature of legitimation processes should guide further research. Luyckx and Janssens (2016) show that different legitimation strategies co-occur, and that the strategies work together in shaping legitimacy.
In accordance with Vaara and Monin (2010), we argue that it is important to pay attention to "discursive sensemaking processes through which legitimacy is established" (p. 5). Consequently, legitimacy is of the utmost importance for public sector organizations, as by their very nature these organizations are accountable to society. They are subject to political decisions and debates, and they need to show agreement with values in society (such as equality, fairness, democracy, anti-corruption, and transparency). This also holds true for military organizations, as these exist for the purpose of serving public interests. In the case of SAF, with its radically changed tasks and conditions, the search for legitimacy is therefore growing. Yet, few studies address issues related to the legitimacy of public sector organizations, and "even fewer investigate the relationship between communication and legitimacy within a public sector context" (Waeraas, 2019, p. 54). The lack of public sector organizations in the study of legitimation-as-process -particularly regarding the role of discourse and textual strategies -has also been identified by Luyckx and Janssens (2016), who state that "for the most part, this approach has been used in studies on controversial business activities [. . .] and institutional change" (p. 1596).
To develop our understanding of discursive legitimation in public sector organizations, two theoretical puzzles are important: (a) the relationship between reputation and legitimation; and (b) the challenge of inconsistent and conflicting external demands. Both Aggerholm and Thomsen (2016) and Waeraas (2019) tap into these puzzles. It has been noted that the introduction of new communication practices has the potential to harm public sector organizations' legitimacy, and that the focus on strategic communication has impelled them to strive to improve their reputations. As Waeraas (2019) puts it, new and more active communication policies "could induce public organizations to focus on communication in a way that ultimately is more designed to benefit themselves than their constituents" (p. 45). Strategic communication does not necessarily aim at reaching general societal objectives, but may instead seek to frame and present an organization as "standing out." In other words, the organization's legitimacy can potentially be at stake if too much attention is given to selfpresentation, and too little to respecting socio-political justifications. However, and as noted by Suddaby et al. (2017), organizations suffer from an ongoing need to be both different/unique and, at the same time, similar to other organizations. This has been termed the "uniqueness paradox" by Martin et al. (1983), who focus on the pressure that organizations face to be different but in the same way. Given the discussion about strategic communication in military organizations, and whether the parochial goals of governmental organizations tie in with grand national interests (cf. Paul, 2011), this becomes particularly salient.
The second puzzle has to do with the dynamic, heterogenic, and often inconsistent demands and expectations facing public sector organizations (cf. Waeraas & Byrkjeflot, 2012). As noted by Aggerholm and Thomsen (2016), public managers have become "negotiators" who mediate between heterogeneous texts and between different discourses and political interests. This role forces them to navigate between different legitimation strategies. One way of understanding expectations of external legitimacy is that they function as a conformity constraint on organizational behaviour. Thus, the quest for legitimacy forces organizations to act and communicate in accordance with external expectations and norms. However, because expectations are socially constructed, discursive legitimation also involves manipulation of institutional expectations (Lischka, 2019). Lischka (2019) argues that organizations might prefer manipulation to conformation when a significant change of conditions or a transformational process is underway within the organization. Hence, strategic communication might be seen as a means for signalling both conformity and manipulation of expectations. Scherer et al. (2013) claim that there are three main responses to legitimation problems: organizations can either adapt to external expectations, they can try to manipulate the perceptions of stakeholders, or finally, they might engage in discourse with those who question their legitimacy. Moreover, Scherer et al. (2013) put forward a "paradox approach", where organizations can employ different strategies simultaneously, "even where these are in conflict with each other" (p. 274). They argue that organizations cannot choose between different strategies; on the contrary, they need to learn to accommodate conflicts between different strategies for dealing with expectations and demands. From this perspective, paradoxes become a "critical theoretical lens", for illuminating and understanding internal and external tensions and contradictions. Glozer et al. (2019) show that more fluid and temporal understandings of legitimation have been concerned less with binary distinctions (e.g., legitimate/illegitimate), and more with degrees of worthiness or appropriateness. Furthermore, this ties in with Christensen et al.'s (2020) view on communicational hypocrisy. Here, hypocrisy is seen as a normal and inevitable practice, "through which organizations of all sorts seek to handle conflicting demands" (p. 328). The legitimacy of public sector organizations must therefore be understood in a context of heterogeneity, paradoxes, and hypocrisy.

Material
To reiterate, the research questions that have guided this article are: (RQ1) How does SAF discursively legitimate its organization, activities, and practices internally? (RQ2) How can public sector organizations signal legitimacy by means of strategic communication in a context of heterogenic and dynamic expectations? As previously stated, the Swedish Armed Forces is an interesting case, as this organization has typically been characterized by strict, hierarchical internal communication with little concern for gaining internal popularity. The recurring transformations in SAF's tasks and conditions, for instance, regarding recruitment, have, however, necessitated a change in their external and internal communication in the direction of more open and market-driven communication. This shift could arguably be seen as simply a sign of the times and a consequence of new principles for managing public sector organizations. However, given the basic purpose of a military organization, the shift from secrecy to openness seems quite counterintuitive, even unnatural. Furthermore, for a non-aligned country like Sweden, with its long history of peace, the challenge of communicating the need for rearmament puts legitimation at the very centre of attention.
In the present study, editorials from the staff magazine Försvarets Forum (The Armed Forces' Forum) have been analysed by means of discourse analysis. Försvarets Forum is produced by the Office of Communications and Public Affairs and is issued six times per year. According to SAF, the magazine should describe the everyday life of employees as well as provide information about current events. The magazine is also available to the public, as every issue since 2011 is published on the SAF website. In the editorials, the head of SAF speaks directly to the organization, and these texts are prominent examples of how the very idea of the organization is formulated, justified, and passed on to organizational insiders. The editorials provide arguments for the existence and transformation of the organization and create a sense of belonging and identity for the insiders. Furthermore, these texts are important mediators between the macro-level context of SAF and the nitty-gritty world of organizational concerns. The editorials are thus a relevant source of information about how SAF's strategic communication is put into play internally. Although there are many other sources of internal communication in SAF, to our knowledge there are no other channels in which the Supreme Commander directly addresses all employees on a regular basis. For this study, 31 issues have been analysed, starting with issue 6, 2014, and ending with issue 6, 2019, which comprise all editorials in the chosen period. The starting point of the analysed material marks the launch of a new format for Försvarets Forum, which is why this period was selected for the analysis.
The editorials are relatively short texts, printed on a single page, and in total the text corpus consists of 10,750 words. In the editorials, the Supreme Commander addresses employees by presenting and discussing emerging issues related to SAF. As mentioned, communication in military organizations has historically been highly centralized and hierarchical, but these editorials to a large extent transcend such conventions. Instead, they are written in a personal tone, and tend to have a "human voice" (cf. Kelleher & Miller, 2006). Typically, the editorials cover 3-4 topics. Almost all the editorials tap into the economic and political situation of SAF, which is the most prevalent topic in the texts. This is related to the political-economic contexts surrounding SAF which frame the organizational insiders' environment. Hence, the editorial functions as a "filter", where top-level decisions within the organization and external societal conditions are internally communicated through the organization. Other main topics covered in the editorials concern SAF's roles and activities, personnel related topics, and matters of risk, threat, and preparedness.
The study is limited by its focus on the editorials in Försvarets Forum, which do not represent the totality of SAF's strategic communication. SAF also hosts other channels such as websites, social media, interviews, and documents that could have been chosen instead. Different kinds of material from the same magazine could also have been chosen; an analysis of all types of articles in Försvarets Forum would have opened up for a multimodal approach that also investigates the interaction between text and image. By relying on textual data alone, there is a risk of missing out on the many nuances and variations of communication that a richer data set might have provided and that might have pushed the analysis further. However, the choice of data was largely based on the source of the communication; it was considered crucial that it comes from the top management of the organization to illustrate how organizational strategies are practised communicatively. This seems to be particularly relevant in a hierarchical organization such as the military. The limitations of the data aside, there is still reason to believe that the insights from studying SAF's editorials are transferable to other parts of the world due to the similar structures, conditions, and tasks of military organizations globally (cf. Moskos, 1977).

Discursive legitimation analysis
The analysis of SAF's communication in Försvarets Forum is based on the so-called discursive legitimation approach (DLA), inspired by previous work by Van Leeuwen and Wodak (1999), Vaara et al. (2006), Aggerholm and Thomsen (2016), and Mampaey et al. (2020). The analysis is conducted in two steps. First, we seek to identify systematic patterns, recurring words, and frequent topics. This was thus an inductive process, and we embarked upon it with an open mindset. We did not create any coding schemes or categories in advance, but let the data guide the definitions that were later used in the coding. Krzyżanowski (2010) argues that defining topics inductively is an integral part and important first step of any discourse analysis. This takes place "via several thorough readings -and then ordering them [topics] into lists of key themes and sub-themes" (p. 81). All editorials were read several times by both researchers and coded according to their topics. Furthermore, word counts were conducted (using Voyant tools) to identify frequent terms and concepts. This first step thus painted an overarching picture of the contents of the communication and gave us deeper knowledge about patterns in the material.
Secondly, we focus on how legitimacy is constructed through different discursive strategies. Discursive legitimation analysis (DLA) is based on critical discourse analysis (Van Leeuwen & Wodak, 1999). Among the vast panoply of discourse analytical approaches within social sciences, DLA in particular may be able to shed new light on organizations' efforts to make sense of and generate legitimacy for their existence and activities (cf. Vaara et al., 2006). A central idea within DLA is how discourses play a part in social constructions of social order and the distribution of power. Hence, organizational communication is not taken at face value (Vaara & Monin, 2010) but is analysed to flesh out underlying meanings. Vaara et al. (2006) argue that "looking at these kinds of micro-level discursive elements helps us to understand the complexities, ambiguities, and contradictions in legitimation, which easily pass unnoticed with more traditional approaches" (p. 790). The analysis of discursive legitimation of SAF highlights the role of sense-making, the political nature of legitimation (cf. Vaara & Monin, 2010), and specific discursive strategies for legitimation in relation to organizational insiders.
Hence, in the editorials we searched for passages dealing with constructing SAF as an organization, SAF's relationships to external constituents, and SAF's values and conditions, in addition to expressions seeking to gain trust. These passages and expressions were coded in accordance with five basic legitimation strategies, described in Vaara et al. (2006), which later have been said to constitute a specific "grammar of legitimation" (Aggerholm & Thomsen, 2016). Our framework thus consists of five widely used categories of legitimation in DLA studies: normalization, rationalization, authorization, moralization, and narrativization (Table 1). Although this grammar of legitimation is based on pre-given categories, the framework certainly provides leeway for exploring data in sensitive ways. The grammar of legitimation is a flexible framework, where patterns and topics emerging from the empirical data can be analysed and presented in a structured way. The choice of relying on established and pre-given categories instead of creating new concepts was based on two considerations. Firstly, these categories have been developed over time, and there is a common understanding of how to interpret them (e.g., Aggerholm & Thomsen, 2016;Vaara et al., 2006;Van Leeuwen, 2007;Van Leeuwen & Wodak, 1999). In the analysis presented in the next section, we further use commonly accepted and widely used tools from critical discourse analysis (e.g., Fairclough et al., 2011;Machin & Mayr, 2012;Van Leeuwen, 2008) to illustrate how these five legitimation strategies are discursively constructed in our material. With this approach, it should be fairly easy to test the reliability of our study. Secondly, using the same categories as previous researchers opens up for possible comparisons with previous (and future) studies of different organizations and types of material.
The dialectical tensions embedded in legitimation processes tend to be underscored in more recent studies (cf. Mampaey et al., 2020). The idea is that different forms of legitimation foreground particular aspects while obscuring others. For instance, authorization grants authority to some persons or institutions, while others are marginalized or disregarded. Our approach is not normative in the sense that we propose or prescribe strategies that should be implemented to shape the identities of organizational insiders. On the contrary, and in line with Mampaey et al.'s (2020) arguments, the aim is to demonstrate, investigate, and explain complexities, challenges, and dilemmas.

Results
Table 2 summarizes important topics in the editorials related to the five forms of discursive legitimation of SAF. Each of these legitimation strategies is then explained in depth and exemplified with quotations to illustrate our argument. All quotations have been translated from Swedish to English and are referred to by issue number and year (e.g., 5, 2017).

Normalization: Uncertainty and impending threat
Van Leeuwen and Wodak (1999) highlight a form of "conformity legitimation" related to authorization, which rests on the principle of "everybody does it." Hence, this conformity legitimation refers to the normalization (or sometimes "naturalization") of something. Vaara et al. (2006) expand on this discussion, defining normalization as "the primary type of legitimation, as it seeks to render something legitimate by exemplarity" (p. 798). This also implies that other actions or phenomena are turned into something "abnormal." Here, normalization has to do with establishing a worldview where there are "no other options"the editorials present military expenditures and rearmament as only natural, given the circumstances. SAF is thus legitimated with reference to trends such as uncertainty, troubled times, and unstable surroundings. This is done partly through repeated formulation of a threat in "our immediate surroundings," which conveys a sense of urgency. These wordings are found, for example, in the legitimation of the militarization of Gotland and in descriptions of the situation in Ukraine. The latter is further described in the quotation below as "a mere two-hour flight from Stockholm." The world around us does not pause. I got a reminder of the tensions in our own immediate surroundings when I recently visited Ukraine and could see with my own eyes how the ongoing war is affecting the country's armed forces and civilian population. It was a visit to another reality, a mere two-hour flight from Stockholm. (5,2017)  Establishing legitimacy by moral arguments or specific value systems that provide the moral basis for legitimation.
References to higher/divine purposes, final causes, value systems, and emotional aspects Narrativization Legitimation by telling stories or constructing narrative structures to indicate how the issue in question relates to the past or the future.

Construction of compelling plots or visions of the future
Though not explicitly stated, this reference to geographical proximity strongly implies that tensions could quickly spread to Sweden's borders, thus legitimating defensive rearmament. Using the Supreme Commander as a reliable eyewitness elevates the truth claim in this statement, which is further dramatized with the graphic metaphor of "another reality." This example is also interesting, as it is one of the few times that "war" is mentioned in the editorials, which adds to the sense of an acute threat.
Of course, a discourse of threat highlighting the need for rearmament is what could only be expected in this type of material. SAF is partial in this discussion, and for them not to encourage a strengthening of the military's defensive capacity would be quite unnatural. There is furthermore little doubt that the military activity in and around the Baltic Sea has increased in the past decade, and that this activity centres around Russia (e.g., Bengtsson, 2015;Kragh, 2018). This article does not question these premises per se, nor does it question the need for a robust defence system for a sovereign country -non-aligned or not. Nevertheless, the discursive strategies around this matter are interesting, as they imply that the described gravity of the threat to Sweden is politically noncontroversial and commonly accepted, which has not been the case during this time (e.g., Regeringskansliet [Government Offices of Sweden], 2019, appendix 3). Opposing views are not only backgrounded, but almost ridiculed. For instance, it is stated that "no serious participant in the discussion" (2, 2015) would question the need to strengthen Sweden's operational capacity and thus the need to increase the defence budget, a presupposition which suggests that this is a common-sense truth, and that anyone who thinks otherwise is unreliable (cf. Machin & Mayr, 2012).
The militarization of Gotland is legitimated using the same normalization strategy: On a sunny Monday in May it was once again time. Gotland's regiment is reestablished. It is great and it is right. You only have to look at a map of the Baltic Sea to realize Gotland's significance for security policy. (3, 2018) Again, we observe a no-other-way approach, where the decision is over-simplified and potentially opposing views are suppressed. Here, high modality (Machin & Mayr, 2012;Mautner, 2008) and a presupposition of common understanding are used to legitimate the reestablishment of military bases on Gotland: "It is great and it is right." One need look no further than at "a map of the Baltic Sea." The same phenomenon can be seen in the normalization of NATO cooperation. Even though Sweden is officially non-aligned (but indeed a partner country), exercises and cooperation with NATO are mentioned in passing. There is no discussion of the controversy around this kind of cooperation, nor any mention of the political divisions between the parties about this question (Ydén et al., 2019).
Despite this normalization of uncertainty, Russia is notably absent in the editorials. Several explanations of this are possible. One reason could be a fear of "crying wolf" combined with a wish to avoid the sometimes-comical effect of portraying Russia as the perpetual Bad Guy. Another could of course be of diplomatic nature. Meanwhile, the listed reasons for the need for rearmament ("cyber security," "the situation around the Baltic Sea," "the Arctic region," "events in Ukraine") most often clearly allude to Russia as the threat, rendering its suppression in the editorials quite superfluous.

Rationalization: SAF as a "normal" organization
Rationalization is a form of legitimation achieved by referring to the utility of specific actions based on accepted knowledge claims in a given context. It refers either to the utility of the social practice or some part of it (instrumental rationalization), or to the facts of life (theoretical rationalization) (Van Leeuwen & Wodak, 1999).
The rationalization has two dimensions in the data. One is largely related to reputation and has to do with constructing SAF as a "normal", "modern", and even "progressive" organization with EHS activities, up-to-date personnel policies, and well-implemented policies for equality and diversity. As illustrated in the quotation below, the reason for stressing issues related to diversity and equality is ultimately presented as rational rather than ideological or moral: it will strengthen the organization and increase Sweden's military defence capability. The truth of this claim is strengthened by the personal authority of the Supreme Commander, who has chosen to be "personally involved in this matter".
Women, peace and security are a prioritized matter for both our own total defence and for Swedish security policy at large. I have been personally involved in this matter. The reason is simple: Gender equal armed forces are stronger armed forces. It has to do with operational effect. (5,2018) This is opposed to, for instance, the militarization of Gotland, which was simply described as "right"; it thus seems that gender equality must be legitimated by more productivity-oriented arguments. It is beneficial because it gives SAF better results -"operational effect" -not because of the other potential advantages of a mixed work force. This statement could also, however, be understood in relation to the other dimension, which to a greater extent taps into rationalization based on national objectives. From this viewpoint, it is perfectly rational for SAF to make whatever adjustments are necessary to maintain Swedish sovereignty. It is thus rational to maintain good psychosocial health, as in the example below, or to be "modernizing [SAF] regulations" -i.e., to oversee the structure of the organization and renegotiate previous agreements -as this will "deliver the greatest possible effect" (2, 2016) for Sweden at large.
In this issue you can read, among other things, about psychosocial health within the Swedish Armed Forces, something I take most seriously. We must take care of ourselves and each other, and make sure we create a good working environment for everyone if we are to manage Sweden's armed defence. (5,2017) These two dimensions are often entangled with each other. For instance, the construction of SAF as a listening organization is evident, conveyed through lexical choices that connote mutual exchange (see e.g., Drew & Sorjonen, 2011). The Supreme Commander of SAF "has conversations" with subordinates, "visits units to learn about their work" and calls for "open" and "direct dialogue" (examples from 1 & 4, 2016) with units and employees. This paints a picture far from the stereotypical image of the hierarchical military organization, which seems quite counterintuitive in an organization based on strict chains of command and clear orders. Again, however, the reason for this approach is expressed in rational, productivity-oriented terms. Just as with gender equality, it does not seem to have a value of its own but is rather described as "crucial . . .. if we are to be able to deliver what is ordered" (1, 2016), which could be understood both from an organizational and a national perspective.
Other times, it is solely the reputational dimension that is in focus. This can be seen in the frequent mentioning of internal surveys as a way to legitimate SAF as a modern, rational organization. Perhaps not surprisingly, the reports on these surveys are highly positive. For instance, employees' "high trust in the SAF leadership and great enthusiasm in their work" is "confirmed" (6, 2017). Problematic issues such as harassment and victimizing behaviour are addressed to some extent, for instance, in connection to the MeToo movement of 2017. However, as illustrated in the quotation below, aggregations such as "many," "many, like me" and "the great majority" in connection to positive job satisfaction suggest that this is a limited problem, without any supporting statistics to confirm these claims (cf. Machin & Mayr, 2012;Van Leeuwen, 2008).
It is obvious from this year's employee survey that many, like me, share the pride of working in the Swedish Armed Forces. The trust in our organization has increased sharply since we carried out FM Vind [internal employee satisfaction survey] three years ago. The great majority feel that their nearest leaders are really good and that the teams work well. (6,2016) Here, the declarative commitment (Mautner, 2008) to statements "it is obvious" and "has increased sharply" leaves little room for differing opinions, although supporting arguments or statistics are lacking. The take-home message in these instances is that most employees -often conveyed by an inclusive "we" (Fairclough et al., 2011) -"feel well and are thriving at work" (6, 2018), and are "proud to work in SAF" (6, 2016; 6, 2019). Meanwhile, the 1,768 women who signed the MeToo appeal for SAF in 2017 suggest something else ( DN Debatt, 2017).
The form of rationality evident in the editorials relates to claims about SAF as a complete and rational organization (cf. Brunsson & Sahlin-Andersson, 2000). SAF is presented as an organization like any other modern organization, with recruitment policies, well-established personnel policies, and concern for psychosocial health. Just like any other organization, SAF thus must "work hard" (4, 2016) to increase recruitment. Although external campaigns often emphasize SAF's "unique offer," i.e., action and armed combat (Deverell & Wagnsson, 2016), the editorials instead highlight more general recruitment terms such as "career," "attractive opportunity" and "high-quality education" (3, 2018), reflecting the recruitment communication of any other corporation (e.g., Young & Foot, 2005).
Hence, SAF is legitimating itself by adapting to general organizational models, which differ from the traditionally rigorous and hierarchical military organization in many respects (e.g., Hasselbladh & Ydén, 2020). Within this context, a standardized, caring, and well-regulated personnel policy with a corporate-style organizational culture can potentially create a sense of a stable and promising future, which makes this form of legitimation highly rational for the insiders as they are enlisted or employed in this organization. As for dialectical tensions, this form of rationality at the same time relegates other aspects to the background. Military organizations are substantially and essentially different than most other organizations. The backgrounding of this fact is an interesting strategy, and seemingly contradicts the strategies of SAF's marketing communication (cf. Deverell & Wagnsson, 2016) as well as the primary task of the organization.

Authorization: Individualized responsibility, impersonality, and conformity
Authorization is creating legitimacy by referring to the authority of tradition, custom, law and/or persons. Van Leeuwen and Wodak (1999) argue that this form of legitimation is achieved with reference to implicit questions like "why is it so?" and "why must it be so?" The answer "is essentially 'because I say so' or 'because so-and-so says so'" (Van Leeuwen & Wodak, 1999, p. 104).
In the editorials, authorization serves as a particular form of constructing hierarchical leadership in combination with individualized responsibilities. This seems somewhat paradoxical, as the following two examples illustrate: At home, I began the autumn term with several meetings with commanding officers all around Sweden. I listened to what they had to say and can confirm that we are on the right track, but that the pressure is still high. Our staff has a heavy workload. Many bear witness to how skill shortages and vacancies are straining the organization. Solving the situation is both a responsibility of commanders and a matter of resources, and during the process, we must take care of ourselves. In the end, it is a question of the individual employee's dedication and safety. (5, 2018)

***
There is still a lack of insight within the organization about why women are needed to develop our military capability. Therefore, it is my responsibility and that of all commanding officers to make this clear and to take action to improve the situation. But this work must also always begin with you. I expect everyone to do their part so that our daily work is defined by respect, capability, and moral courage. (3,2017) In these two quotations, issues such as personnel shortages and harassment or other forms of victimizing behaviour are first made a leadership matter through active wordings such as "I listened", "it is my responsibility and that of all commanding officers . . . to take action", and "I expect". Explicit comments on how to address these issues are absent however, which might divert attention from the part leadership must play in solving the problem (cf. Machin & Mayr, 2012). Instead, declarative statements and statements of high modality such as "In the end, it is a question of the individual employee's dedication" and "this work must also always begin with you" shift the focus to the individual level. The order from the Supreme Commander is however vague and abstract -how does one operationalize "dedication," "capability," and "moral courage?" Ultimately, nothing gets said about how to fix these problems.
Further, the authorization in these editorials is rooted in impersonal decision-structures ("the government bill," "the study," "the analysis," etc.). This is particularly apparent when the editorials discuss military expansion: In the beginning of September we received some welcome news, giving the Swedish Armed Forces the means to continue growing. In 2020, a defence policy government bill will be presented based on The Swedish Defence Commission's report. There is now broad support for the financing. This shows that there is consensus around further strengthening our military capability. (5,2019) These impersonal wordings communicate that the search for -and legitimacy of -increased defence spending does not only come from SAF itself, but also from the more "impartial" authority of policy, decisions, reports, etc. This quotation also illustrates how legitimation of military expenditures is also achieved by authority of conformity (cf. Van Leeuwen & Wodak, 1999) in wordings such as "the political majority," "the political intention," "our elected representatives," "parliament," "consensus", "political consensus," and "broad [political] support." Not only does this strategy firm up SAF's arguments -as Mackie (1987) puts it, "the more people that agree with a position, the more valid the position must be" (p. 42) -it also makes it difficult to question, as challenging the political majority might seem fruitless.
The agreement on military investments across political party lines is thus consequently foregrounded in these editorials through lexical choices connoting consensus. This consensus does not however necessarily reflect the different party lines, which demonstrate a more diverse approach to defence matters (e.g., Regeringskansliet [Government Offices of Sweden], 2019, appendices 2, 3, and 4). Just as opposing views on the impending threat towards Sweden are backgrounded, opposing views on military funding are put aside.

Moralization: Comradeship and diversity
Moralization is legitimation by referring to specific value systems that provide the moral basis for legitimation (Van Leeuwen & Wodak, 1999). In the studied editorials, moralization as a form of discursive legitimation taps into well-known values of military organization (comradeship, responsibility, virtue, strength, and so on) in combination with contemporary ideals of (gender) equality and diversity. The "cog in the wheel" metaphor is for instance, used to signal the moral value of equal participation and contribution (e.g., 3, 2016).
Pride, responsibility, importance, and comradeship is consequently conveyed through an inclusive "we", the repetition of which can function to increase group identification (cf. Machin & Mayr, 2012). The repeated use of "we" also depicts an organization where equality -not hierarchy -is valued, which basically goes against the very idea of the military organization. Furthermore, its meaning is fluctuating throughout the editorials; sometimes "we" refers to all members of SAF, sometimes just to the leadership, and sometimes to Sweden at large. When matters of rearmament are addressed the meaning of "we" suddenly becomes very fuzzy -is it Sweden as a nation or SAF as an organization who needs to "gear up and act" (1, 2018)?
The emphasis on equality and diversity becomes ambiguous when it comes to issues of harassment or other victimizing behaviour. The editorials tend to tone down problems for female workers by depicting their challenges as subjective "experiences" and "narratives" (6, 2017), and by describing harassment as related to feelings instead of facts, as illustrated in the quotation below. Here, the hedging technique of using "feel" instead of a factive wording such as "have been" makes the statement less certain (cf. Machin & Mayr, 2012). These lexical choices give the impression that while these issues might exist in SAF, they seem to be more a matter of personal than structural problems.
It is unacceptable that so many of our colleagues -six per cent on average, and as many as eleven per cent of the women -feel badly treated. (6,2016) Another interesting aspect of SAF's moralization strategy is the lowered modality and abstraction used in describing casualties and deaths connected to military service. This is in line with the rationalization of SAF as a modern, "normal", organization, where death is of course not an issue. It is thus not described directly but presented as "the worst that can happen" immediately followed by assurances of "constant efforts to improve safety" (e.g., 2, 2019). Hence, the risk, threat, and danger associated with military organizations -the flip side of the "unique offer" -is toned down. In the context of international operations, casualties and deaths are legitimated by referring to the higher moral value of contributing to global peacekeeping, "a sharp reminder . . .. of why we are there": In the beginning of May, the UN camp in Mali came under rocket fire. A Swedish soldier was injured, a Liberian soldier lost his/her life, and many more were wounded. This is a sharp reminder of the difficult operational environments that many of our colleagues are working in -but also of why we are there. (3,2017) The functionalization of the victims as soldiers and comrades is also notable here. Any individualizing markers such as name, age, rank, or unit are excluded, thus distancing them and the event further from the reader (cf. Machin & Mayr, 2012).

Narrativization: international cooperation and total defence
Narrativization (or "Mythopoesis", in Van Leeuwen & Wodak's, 1999 terminology) refers to legitimation that is accomplished by telling stories or constructing narrative structures to show how an issue relates to the past and/or future. The narrative plot about SAF has two threads in these editorials. The most prominent thread relates to the internationalization of conflicts and security, and the narrative is that continuous international cooperation through joint exercises and missions is imperative for building global stability. This is illustrated in the following quotation: The Swedish Armed Forces are very much in-demand internationally. Our contributions to different peace operations, not least now in Mali, are very important. This is something that was clearly confirmed to me during my May visit to the UN headquarters in New York for a meeting with more than 100 military chiefs from near and far.
Our international participation also contributes to supporting Sweden's interest in a stable global environment, one where our crucial trade flows can be upheld. In addition, it has a direct effect on increasing our competence level -what we learn in Mali makes us sharper at home as well. (2,2015) The declarative commitment to this kind of statement is high; there is no question that SAF's contribution is "very much in-demand" and "very important." This is also "clearly confirmed" by "more than 100 military chiefs from near and far" -again relying on the authority of conformity to strengthen the claim's truth.
In this narrative too, war is notably absent, at least as far as SAF is concerned. War is something that the bad guys do, while SAF's operations "contribute to peace and security in vulnerable places" (4, 2019). SAF are thus not increasing their readiness for action, but simply "maintaining security" (4, 2019); they are not procuring new armaments, but are ready to "gear up and act if needed" (1, 2018). As the quotation above shows, a very rational legitimation of international involvement is presented, similar to that offered for gender equality earlier: this sort of activity will build a stronger Swedish defence. How this involvement corresponds to Sweden's official policy of non-alignment is not addressed.
The other narrative dimension relates to SAF's broadened national mission in times of crisis, and the undertaking of a task that encompasses more than just traditional warfare (cf. J. Eriksson, 2001). This broader mission is exemplified by recollections of SAF's contributions during the 2018 forest fires and justified by the need to be prepared for "new arenas for warfare" (4, 2016). An integral part of this narrative is the "urgent" return to total defence, for which it is "high time" (1, 2018). Again, these lexical choices convey a sense that the Swedish rearmament is a pressing matter.

Concluding discussion
Two theoretical puzzles pinpointed by previous research (e.g., Aggerholm & Thomsen, 2016;Suddaby et al., 2017;Waeraas, 2019) have contributed to our understanding of the discursive legitimation of SAF: (a) the relationship between reputation and legitimation, and (b) the challenge of inconsistent and conflicting external demands. Our findings can be summarized in two main points: (1) the frequent and unexpected use of rationalization as a legitimation strategy, highlighting conformity with other organizations instead of differentiation; and (2) inconsistent legitimation strategies, with conflicting values and perspectives of strategic communication operating simultaneously. While these findings partly agree with the previous literature, there are also more surprising elements which could help advance the general understanding of discursive legitimation in public sector organizations beyond the specific case.

Rationalization: conformity instead of differentiation
In line with Aggerholm and Thomsen's (2016) findings, we find interconnected legitimation strategies -"voices" -in the internal communication of SAF. Some are only what could be expected, such as normalizing and moralizing military rearmament and work. However, where Aggerholm and Thomsen identify authorization, mostly with reference to laws, as the most prominent strategy in their study, rationalization is just as frequently used in our material. This relates to the legitimation of SAF as a "complete organization" (cf. Brunsson & Sahlin-Andersson, 2000), which is done using concepts commonly found in organizational studies and human resource management. The rationalization strategy discursively legitimates SAF as a normal, modern, and even progressive organization. This is done by referring to EHS policies, issues of equality, and diversity policies in the organization. This type of rationalization connects well with the narrativization strategy, in the sense that "war" is omitted from the descriptions. Through rationalization, SAF is presented as just like "any other" organization, where "war" has no place. This is also evident from the absence of veterans as a category in the editorials. Organizational insiders are instead addressed as employees with modern-day careers who face organizational problems that are best solved through an expansive personnel policy focusing on psychosocial environments, in combination with well-planned recruitment policies and campaigns. Hence, the organization is constructed as "desirable, proper and appropriate" (cf. Suchman, 1995). This also relates to the moralization strategy, whereby contemporary ideals of gender equality and diversity are highlighted in a discourse gravitating around a rhetorically constructed and inclusive "we." The repeated use of "we" also depicts an organization where equality and "gemeinschaft" -not hierarchy -are valued. As such, this resonates well with findings from other studies within the sphere of security, where promoting everyday conceptions of "ordinariness" has become prevalent (cf. Rasmussen, 2020).
These legitimation strategies highlight the "uniqueness paradox" (Martin et al., 1983) in a very specific way. We know from the literature that organizations need to be both similar to and different from other organizations (Suddaby et al., 2017). Intriguingly, it is thus SAF's similarity with other organizations that is highlighted rather than its uniqueness, which could have been accomplished by focusing on the distinct features of military organizations. Waeraas (2019) argues that "if a public sector organization responds to the pressure of building a favorable reputation by differentiating from others, it may ultimately jeopardize cognitive judgments of its legitimacy" (p. 52), which could explain SAF's strategy. However, SAF is not just any organization; it is an important instrument for the Swedish state's security, and it has the power to protect and destroy lives and material (Olsson, 2013). Acknowledging this difference might in this case be more "appropriate" than the opposite. There is reason to believe that this challenge in discursive legitimation is particularly salient for public sector organizations due to the changing conditions and demands of recent years. Public sector organizations have to a large extent been rebuilt as more complete structures, and are no longer sharply distinct from their private counterparts (Waeraas & Byrkjeflot, 2012). At the same time, they have duties that require inconsistent value-sets, and they are a priori subject to political decisions. These are considerations that should be addressed by public communicators.

Inconsistent legitimation strategies: Conflicting values and perspectives
The rationalization of SAF also incorporates a "national" dimension, where Swedish sovereignty is in focus. In its latter form, the rationalization of SAF is thus closer to Paul's (2011) understanding of (military) strategic communication, while the former is more connected to strategic communication as conceptualized by Hallahan et al. (2007), with prominent features of reputation management (Waeraas, 2019). This shows how inconsistent strategies are put into play simultaneously, and how the view of strategic communication seems to shift throughout the texts: Should SAF be promoted as an organization, or should the national objectives steer the communication?
Furthermore, the inclusive "we", which is an important part of the moral legitimation of SAF, stands in contrast to the legitimation strategy of authorization, which highlights impersonal decision structures in a way that is more in line with Aggerholm and Thomsen's (2016) findings. The focus on diversity and equality is also contrasted with a persistent underscoring of individual responsibility. By foregrounding individual responsibility in tackling diversity and harassment issues, the structural dimensions of such issues are relegated to the shadows. This, in turn, discursively counteracts the strategy of becoming a modern and progressive organization. These examples illustrate that inconsistencies exist on the micro-level of discursive legitimation. Furthermore, and as noted by Christensen et al. (2020), it might build up to the level of communicational hypocrisy. Christensen et al. (2020) view hypocrisy as "a normal and inevitable -if not acceptable -practice through which