Conceptualizing Journalists’ Safety around the Globe

Abstract Killings, as the most extreme form of violence against journalists, receive considerable attention, but journalists experience a variety of threats from surveillance to gendered cyber targeting and hate speech, or even the intentional deprivation of their financial basis. This article provides a comprehensive, interdisciplinary framework of journalists’ safety, summarized in a conceptual model. The aim is to advance the study of journalists’ safety and improve safety practices, journalism education, advocacy, and policy making - vital as press freedom and fundamental human rights face multifaceted challenges, compromising journalists’ ability to serve their societies. Journalists’ occupational safety comprises personal (physical, psychological) and infrastructural (digital, financial) dimensions. Safety can be objective and subjective by operating on material and perceptional levels. It is moderated by individual (micro), organizational/institutional (meso), and systemic (macro) risk factors, rooted in power dynamics defining boundaries for journalists’ work, which, if crossed, result in threats and create work-related stress. Stress requires coping, ideally resulting in resilience and resistance, and manifested in journalists’ continued role performance with autonomy. Compromised safety has personal and social consequences as threats might affect role performance and even lead to an exit from the profession, thus also affecting journalism’s wider function as a key institution.

covering topics that raise the ire of governments and drug lords, and reporting from conflict zones or, more recently, from places especially hit by  Journalists' occupational safety, however, is jeopardized in many more ways these days -even in countries where the workforce is considered to be physically comparatively safe. For one, journalists experience a variety of psychological threats. Particular areas of concern include aggressions such as hate speech (Obermaier, Hofbauer, and Reinemann 2018) and sexual harassment (Ferrier and Garud-Patkar 2018;Idås, Orgeret, and Backholm 2020;Unaegbu 2017), often in digital environments (Adams 2018;Chen et al. 2020), losing a colleague to violence, and experiencing other multiple forms of job-related threats and aggressions on journalists' mental and emotional wellbeing (Garc� es-Prettel, Arroyave-Cabrera, and Baltar-Moreno 2020; Hughes and M� arquez-Ram� ırez 2018;Osofsky, Holloway, and Pickett 2005;Parker 2015). The financial security of journalistic labor has also become an issue of increasing concern as dysfunctional media markets and political entanglements, advertisement cuts and, in conjuncture, the lack of strategies to monetize digital content jeopardize the continuous, professional performance of journalistic work all over the world (e.g., Hayes and Silke 2019;Matthews and Onyemaobi 2020).
Journalists' safety not only affects individual journalists and the profession but, by threatening freedom of expression, also democracies and societies as a whole (Høiby and Ottosen 2019). A growing body of research is dealing with these issues, in large part prompted by the UN Plan of Action on the Safety of Journalists and the Issue of Impunity (2012), but it still remains the case that the majority of studies focus on one context or a limited number of contexts/case studies and a specific aspect of safety (e.g., Garc� es-Prettel, Arroyave-Cabrera, and Baltar-Moreno 2020; Hasan and Wadud 2020; Hughes and M� arquez-Ram� ırez 2018;Kim 2010;Nyabuga 2016). Comparative studies are rare and mostly focus on specific regions (Asal et al. 2018;Bjørnskov and Freytag 2016;Chen et al. 2020;Marcesse 2017). Despite large policy interest (Torsner 2017), this topic remains under-or unsystematically researched and there is, to date, no common definition of journalists' safety upon which a conceptual framework can rest that relates the roots of the lack of safety to individual and societal consequences. A conceptual framework would capture, in a holistic way, the main factors posing threats to journalists' occupational safety, the coping strategies for dealing with both the threats and the resulting stress and, depending on the support structures and resources available to help journalists cope, journalists' differing responses and the final outcomes of these responses. Given the multifaceted issue at hand, this framework must be interdisciplinary, cover the totality of these complexities in a truly global way, and guide future investigations.
This article aims at providing such a conceptual framework by combining innovatively perspectives from the sociology of journalism, political theory, psychology, media economy, risk analysis, and wider safety research literature. We start by proposing a comprehensive definition of journalists' safety built upon an understanding of the range of threats to safety since safety is mostly considered from the perspective of its absence. We continue by carving out what we diagnose as causes of threatswhich we see rooted in power dynamics that play out between journalists and the political elite or specific social groups making claims to power but also in journalists' responses and actions when they find themselves in unsafe situations and conditions. Power is the fundamental dynamic that undergirds journalists' safety. Power manifests in factors that put journalists' safety at risk, and these factors spawn threats, which in turn lead to stress, coping, self-preservation efforts, and actions. We propose a conceptual model of journalists' safety as a methodological tool, thus offering potential directions for future research (Figure 1). We conclude by discussing the implications of our contribution to journalism education, advocacy, and policy making.

Conceptualizing and Defining Journalists' Safety
Safety research defines "workplace safety as an attribute of work systems reflecting the (low) likelihood of physical harm-whether immediate or delayed-to persons, property, or the environment during the performance of work" (Beus, McCord, and Zohar 2016, 353). While this conceptualisation provides a good starting point, our definition must detach itself from a concrete workplace and focus instead on work-related tasks because for journalists, workplaces include their offices, reporting sites, and their homes (especially for freelancers). Consequently, we propose to use the term occupational safety because journalists' safety is linked to their occupational activities as journalists. In performing these tasks, the well-being of a journalist as a person as well as their autonomy as an institutional actor are interconnected and thus are integral to a journalist's safety. 2 In the literature, the issue of journalists' safety is typically approached from the perspective of its absence. Research usually focuses on situations in which the likelihood of harm is high. As Workneh (2022) explained, "in its most basic interpretation, particularly in authoritarian and semi-authoritarian contexts, safety denotes the absence of harm generated from the journalist's reporting activities". Existing research into the risks journalists face that can result in potential harm for them and the institution of journalism shows that not only journalists' physical but also their psychological safety is at stake (Table 1). Scholars such as Frazier et al. (2017), Edmondson (1999), and Kahn (1990) define psychological workplace safety as the belief that the workplace allows for risk-taking without the likelihood of severe harm within an organizational context, for example, in a team. In journalism research, a definition must also look beyond the organizational and institutional context. Journalists face scrutiny from a wider range of actors and the reactions to their work are often public. Hence, for journalists, psychological safety includes the possibility of risk-taking without the likelihood of harm from outside actors and institutions, too.
Further, the functioning of journalism is highly dependent on two types of infrastructure: financial and technological/digital. If these are torpedoed or the digital space rendered as threatening, journalistic performance is jeopardized. Thus, journalists must have what we term digital safety: the digitization of their work must not put the workforce at risk. They also need financial safety in terms of continuation of their employment and occupation, currently threatened by precarity in the form of low or reduced wages and job loss, resulting in a shrinking workforce and de-professionalization tendencies observed globally. Therefore, we conceptualize journalists' safety as the extent to which journalists can perform their work-related tasks without facing threats to their physical, psychological, digital, and financial integrity and well-being. Given that these threats are experienced in the course or as a result of performing their professional duties, all four dimensions -physical and psychological, digital, and financial -are part of journalists' occupational safety. This definition captures both objective manifestations of threats in the real world (material level) and journalists' subjective awareness thereof (perceptional level). We use the term threat instead of harm because threats can be both a stated intention by antagonists that harmful actions will be taken against an individual and the anticipation by the threatened individual of imminent harm. Thus, threat refers to anything with the potential to cause harm, while harm is the actual effect(s), which we include under dimensions of safety. Threats are harmful in their own right because they compromise journalists' safety and their ability to perform their duties and serve their respective societies.

Conceptualizing and Defining Threats to Journalists' Safety
We define threats to journalists' safety as the actions and conditions that increase the risk of physical, psychological, digital, and financial harm to journalists as human beings and as institutional actors (Brambila and Hughes 2019). While safety research operates with the term likelihood, we propose the use of the term risk in the context of journalists' safety. Both convey a similar meaning but the term risk bears more relevant, normative implications for what is "safe". Drawing from the interdisciplinary field of risk analysis, risk can be defined "by the probability and severity of adverse effects" (Aven 2011, 515). Accordingly, risk is a combination of likelihood and the extent of harm that may follow as a consequence. Applied to journalism research, (high-)risk situations evolve if journalists face (existential) threats to themselves as individuals and institutional actors and to the viability and sustainability of journalism as an institution making a meaningful and vital contribution to social life. Here again, our definition aims to encompass both the objective manifestations of risk and threats, and journalists' subjective perceptions of them (see also . Existing research (Table 1) revealed a diverse set of threats but our focus is on threats exerted intentionally. 2 To start with, physical threats may lead to bodily harm. They include murder, beating, torture, sexual assault, and other work-related attacks, forced relocation, arrest, detention, imprisonment, abduction and disappearance as well as acute biological and weather-related threats, such as risks to health in the COVID-19 pandemic and from covering disasters. Psychological threats target emotional wellbeing. They include intimidation, coercion, extortion, sexual harassment, doxing (dissemination of personal information), demeaning and hateful speech, public ridicule and discrediting, citizen vigilantism, workplace bullying, and potentially traumatizing work assignments. Digital threats include various hacking and surveillance attacks as well as limiting or blocking access to information, sources, and audiences, thus pointing to the urgent need to study the "multifaceted chilling effects of online attacks" (Waisbord 2020(Waisbord , 1038. A chilling effect has also been detected in the study of financial threats leading to precarious work conditions (Hayes and Silke 2019). Precarity challenges the operational basis of journalism as an institution. On the individual level, precarity manifests in unemployment, the loss of income or position, professional standing, and reputation -or, in less well-equipped journalistic cultures, it characterizes work conditions that had never been economically stable (e.g., Matthews and Onyemaobi 2020). Freelance work, in particular, deprives a journalist of the support of a news organization that can implement safety precautions and training for reporting (Brambila 2018), or provide life insurance and access to good healthcare (Steiner and Chadha 2022). Freelance work may also increase the incentives to take risks because pay is determined by the number of stories produced as well as "scoops" that may require entering into dangerous scenarios.
These threats can originate from actors and groups within the state, but also from non-state and/or foreign actors (e.g., hacking or disinformation, see Table 1). While not of primary focus, all four types of threats can also be directed at family and loved ones and cause risk of harm to their integrity and wellbeing. Moreover, while there are clear distinctions between the different dimensions and threats to safety, there are also significant overlaps. Digital, physical, and financial safety are clearly interlinked with psychological safety (e.g., online hate speech, physical attacks, and job instability could have psychological consequences). In spite of these overlaps, we classify into dimensions as an analytic exercise that will help us understand interactions between them and allow for tailored support systems.
All these threats by their nature compromise journalists' occupational safety and potentially jeopardize the continuation of the performance of journalistic tasks, as well as journalists' autonomy, performance and, fundamentally, journalists' ability to fulfill their social functions. Threats are also a gross violation of individual human rights, and of transparency and accountability principles. They undermine people's right to know that in turn reduces the level of trust in media and participation in public debate so essential for democracy (Hasan and Wadud 2020;Marino 2013). Thus, the safety of reporters is equally a problem for journalism, media freedom and freedom of expression, democracy, and society as a whole (Høiby and Ottosen 2019).

The Root of Safety Issues: Power Dynamics
By building on political and sociological theory, we argue that power dynamics are at the root of the described threats, therefore, undergirding journalists' safety. While different ideas about power exist, we focus on conceptualizations of "power as domination, largely characterized as power over" (Haugaard 2012, 33). According to Weber (1979), power can be understood as the probability of being able to carry out one's own will (or agenda) despite resistance. More differentiated, power can be conceptualized as encompassing three dimensions (Lukes 1974(Lukes /2005Heywood 2015): it may, first, "involve the ability to influence the making of decisions; second, be reflected in the capacity to shape the political agenda and thus prevent decisions being made; and third, take the form of controlling people's thoughts by the manipulation of their perceptions and preferences" (Heywood 2015, 110). The last form of power is also linked to ideological/symbolic or discursive power (115) as the "power to construct reality" (Bourdieu 1979, 79; see also Hayes and Silke 2019).
These power dimensions can refer to an individual or collective actors' power over journalists, or conversely to the journalists' power over these actors. In accordance with the three dimensions of power, while journalistic influence (1) does not comprise any kind of legislative authority, it can under certain circumstances (2) impact the political agenda and (3) shape individuals' perceptions, attitudes, and public opinion (e.g., Levendusky and Malhotra 2016;Lohner and Banjac 2017;Van Dalen and Van Aelst 2014) and, thus, (4) construct reality. Journalists can have this influence because they have privileged access to two scarce resources: (a) information and (b) public attention. Independent of the political and media system under investigation, these resources are granted either on behalf of the state or in the interest of the general public. Journalists can thereby potentially threaten the interests of specific actors and groups. Triggers might be fear of being harmed physically or deprived materially due to journalistic coverage, but also concern that journalists' reporting may publicly challenge the "integrity or validity of the ingroup's meaning system" (Stephan, Ybarra, and Morrison 2009, 43). This can create a feeling of being "dishonored, disrespected, or dehumanized" by members of the journalistic corps (44). Accordingly, journalists are targeted not only because they might uncover illegitimate exertion of power, but also because of their (feared) influence over public opinion and, in the case of female and minority journalists, because they embody cultural challenges to systems of oppression that privilege certain groups or identities. To counter this influence, actors and groups apply means of power in a fight over dominance. Such direct or indirect attempts to limit the symbolic power and influence of journalists affect journalistic autonomy (L€ ofgren Nilsson and € Ornebring 2016), i.e., their ability to self-govern, to make decisions in (moral) independence from internal (e.g., management, economic) and external (e.g., political, legislative) factors (e.g., Sjøvaag 2013;Reich and Hanitzsch 2013).
Use of power can create resistance (Weber 1979), which those in the position of power then have to cope with (see also, Barbalet 1985). Antagonists can resist journalistic power that creates the perceived and unwanted media effects by using (1) decision-making power to limit journalistic freedoms; (2) power to not make/allow a decision, such as (not) passing legislation for trespasses against journalists, or preventing the application of such laws, e.g., impunity; and (3) symbolic power to endorse ideologies antagonistic to journalism, e.g., establishing a delegitimizing "fake news" narrative (Lukes 1974(Lukes /2005; see also Carlson, Robinson, and Lewis 2021). Connected to this, antagonistic forces can also exercise instrumental power (Popitz 2017), i.e., the power to 'control', conduct and safeguard conformity through "the ability to give and take … to be able to make arrangements concerning punishments and rewards that appear credible to those concerned" (12-13). Examples include economic pressure such as governments' withdrawal of advertisements, media's dependence on clientelistic quid-pro-quos with government, and other influence exerted through private sector advertisers' interests (e.g., Guerrero and M� arquez-Ram� ırez 2014;Hanusch, Banjac, and Maares 2020;Wright, Scott, and Bunce 2020). Societal pressure for journalists' protection may be lacking where populist rhetoric or political instrumentalization of news has eroded trust; thus, governments, corporations, cartels, even some members of the public may join in harassment when the press system is perceived as acting against public or group interests (Gonz� alez Macias and Reyna Garc� ıa 2019).
The most intimidating power is the power of action that constitutes the "capacity to inflict harm on others" (Popitz 2017, 11). This form of power, exerted as psychological or physical violence (and captured by our psychological and physical dimensions of safety), is based on inequalities stemming from "inborn endowments, muscular strength … and, above all, from unequal control over contrived devices that enhance the efficiency of the harming action-weapons and the organizational arrangements for combat" (ibid., 11). Such violence is exerted by deviant, disinhibited individuals through misogynous or racist comments, including rape and death threats, but also strategically by members of deviant organizations (e.g., radical right-wing protesters) or by antagonists who constitute the 'power elite' in a country (e.g., Krøvel 2017;Preuß, Tetzlaff, and Zick 2017). When all these types of power coalesce in government, they can accumulate in their effect (Imbusch 2016).
Such power-based antagonistic political, ideological, legislative, economic, and public, i.e., societal, institutional/organizational, and group/individual pressures are the risk factors that create the threats classified in Table 1. They define the room for action within which journalists are expected to perform their work. If and when journalists maneuver to maximize performance and (intentionally or unintentionally) transgress this space beyond a threshold acceptable to antagonists, they become subject to threats and thereby expose themselves to risk. Given that safety is related to the absence, prevention and mitigation of risk that could lead to harm, we turn to a discussion of risk factors, which are rooted in the overriding power dynamics and are present at societal, institutional, organizational, and occasionally individual levels.

Risk Factors
Based on our definition of threats above, risk factors are able to influence whether, and thereby how likely, threats are to manifest in material or perceptional form (moderating factors). In statistical terms, risk factors determine to what extent threats (as independent variables) impact journalists' safety (as dependent variable). We differentiate between individual (micro level), organizational or institutional (meso level), and societal (macro level) risk factors, and start with the latter, which comprise political, economic, and cultural risk factors.
Political risk factors determine journalists' room to maneuver. In societies where the political system cannot guarantee the rule of law, crime, corruption, violence, and human rights abuses are frequently sources of threat (Hughes and M� arquez-Ram� ırez 2018). Authoritarian controls reducing press freedom can also create threats that lead to serious harm to journalists, even in electoral democracies (Hamada 2022). This is especially true in hybrid democratic-authoritarian political systems where some journalists subscribe to norms that seek to restrain the power of the state, but they do so without the protection of the full rule of law (Hughes and Vorobyeva 2021). Closely related to political risk factors, we identify cultural risk factors which may be present in societies where racism, sexism, and homophobia prevail or at least find soil to thrive. Lastly, also closely related to political risk factors are economic risk factors, which include the condition of the journalistic labour market (e.g., O'Donnell and Zion 2019), and-from a political economy viewpoint-concentration of ownership, media owners, and collusion with the state (Garc� es-Prettel, Arroyave-Cabrera, and Baltar-Moreno 2020). All macro level risk factors are interlinked. Thus, in a context such as Central and Eastern Europe, there have been recent democratic backslides, which include the problematic interlinking of political and economic interests and a worrying trend towards "media capture", wherein "media outlets have been brought under direct or indirect government control", including financially through the "weaponization" of government advertising revenue "to harm critical media and financially boost friendly outlets" (Selva 2020, 7-15).
Next, risk factors operate not only on the macro level but also on the meso level in the form of institutional and organizational risk factors. The rise of the platformisation of news and algorithm-mediated messages through social media as well as the blurred boundaries of journalistic workers with content providers pose threats in various ways, including digital hate and mob censorship (Waisbord 2020). Institutional and organizational risk factors may even interact with other risk factors, described above, because in more dangerous work environments they usually also manifest themselves at an institutional or organizational level. For instance, in societies where violence and suffering become frequent topics of coverage, safety risks for the journalists who cover them elevate (Backholm and Bj€ orkqvist 2012;Feinstein, Audet, and Waknine 2014;Reidl Mart� ınez 2012, 2014;Himmelstein and Faithorn 2002;Høiby 2020;Newman, Simpson, and Handschuh 2003;Slavtcheva-Petkova 2018;Tumber 2006;Weidmann and Papsdorf 2010;Weidmann, Fehm, and Fydrich 2008). Consequently, certain newsbeats (e.g., crime, corruption, and politics) make these risks more salient in political systems that are authoritarian or mix democratic and authoritarian traits (Bromley and Slavtcheva-Petkova 2018;Garc� es Prettel and Arroyave Cabrera 2017;Hughes and Vorobyeva 2021). Linked to journalistic routines, task-based risk factors can include witnessing violence and suffering or working in dangerous conditions while covering natural disasters, conflict or violence (Backholm and Bj€ orkqvist 2012;Feinstein, Audet, and Waknine 2014;Reidl Mart� ınez 2012, 2014;Himmelstein and Faithorn 2002;Høiby 2020;Newman, Simpson, and Handschuh 2003;Slavtcheva-Petkova 2018;Tumber 2006;Weidmann and Papsdorf 2010;Weidmann, Fehm, and Fydrich 2008). Even within the organization, risk factors exist in the form of bullying bosses or colleagues, and sexual harassment.
On the micro level, we also evidence individual risk factors such as journalists' personal and professional backgrounds, their occupational and political orientations, as well as personal role perceptions. Personality traits can be relevant as some might simply be willing to put themselves at risk, e.g., by reporting from specific situations or places hit by natural disasters or violent conflict (e.g., Dworznik-Hoak 2020;Feinstein 2006). Further, journalistic role perceptions can play a role, too. Journalists understanding themselves as watchdogs or critical change agents are more likely to pose a threat to business and political elites, and to influence public opinions or agendas, than journalists seeing themselves as opportunist facilitators or populist disseminators (Hanitzsch 2011, 485).
Interaction between the three forms of risk factors should also be taken into consideration. Examples abound -from the impact of intersectionality, for instance, being a female journalist in a "male" coded domain, and being female or a minority journalist and engaging online with audiences (e.g., Adams 2018; Chen et al. 2020) to challenging prevailing cultural norms through reporting on certain topics (e.g., immigration politics). Precarious working conditions have also been linked to the femininity of and in the workforce (Hayes and Silke 2019; € Ornebring and M€ oller 2018). All countries harbour such factors. In their positive form, e.g., constitutional guarantees of press freedom and labour security, these factors could be drivers of journalists' safety; however, more often they are experienced as barriers to safety. For instance, even in countries where press freedom is enshrined in the constitution, the lack of independence, will, and capacity of judicial systems to enforce the rule of law, protect human rights, and respect press freedom become barriers to journalists' safety (Waisbord 2002). They reflect the conflict between journalists' attempts to act with autonomy and the aggressive means criminals, political factions, hate groups, and powerful officials use to try to silence them.

The Consequence of Safety Issues: Work-Related Stress
Antagonists' threats emanate from a resistance to journalistic power (perceived or real), requiring journalists to react in self-defense. Socio-psychologically, the presence of threats creates work-related stress, particularly when resources to react (i.e., to cope) are insufficient. The psychologist Lazarus (1990) conceptualized stress as a process occurring "when demands tax or exceed the person's resources" (3), thus indicating a transactional model between an individual and their environment, and the idea of individual differences and variance in response to stress. According to Lazarus (1990), "[t]ransaction implies that stress is neither in the environmental input nor in the person, but reflects the conjunction of a person with certain motives and beliefs (personal agendas, as it were) with an environment whose characteristics pose harm, threats, or challenges depending on these person characteristics" (3). Based on this, cognitive process models from psychology continue to suggest that journalists experience work-related stress when they perceive an imbalance between the potential harms of work-related risk and their resources, i.e., their ability to avoid or ameliorate these harms (Robinson 2018). Unattended work-related stress may harm journalists' wellbeing and professional performance. Post-traumatic stress disorder (PTSD) in journalists covering war, disaster, violence or suffering is frequently documented (Feinstein, Audet, and Waknine 2014;Reidl Mart� ınez 2012, 2014;Dworznik-Hoak 2020;Seely 2019). Harms from job stress may be similarly physiological as well as psycho-emotional, but less severe than PTSD. The range of harms from stress include physical symptoms, such as fatigue, numbness, exhaustion, headaches, insomnia, stomach ailments, and dermatological disorders as well as psychological and emotional symptoms such as depression, despondency, anxiety, sadness, shame, fear, and pessimism . As Robinson (2018) pointed out, the term stress is very contentious and "the ways in which the term stress [emphasis in original] is used in research is almost as subjective as an individual's expression of a stress."

Coping, Autonomy and Role Performance
The ability to cope successfully is critical for journalists to continue their work and stay in the profession and is related to power dynamics, i.e., the power vested in the institution of journalism in a country, the power exercised by the organization/its editors, and the professional and personal power of the journalists. According to Lazarus (1990), once a person has appraised a transaction as stressful, they bring coping strategies into play. Coping strategies develop in two steps: an initial emotional appraisal of possible harm, followed by an appraisal of the resources available for coping (Folkman and Lazarus 1980;Lazarus and Folkman 1984).
Coping strategies may focus on fixing a problem (problem-oriented coping) or calming negative emotions, e.g., by reframing the situation (emotion-oriented coping), depending upon whether the person affected feels able to prevent or mitigate a potential harm (Alok et al. 2014;Alonso-Tapia et al. 2016). Coping strategies are not rooted solely in the psychological characteristics of the individual. Rather, coping also has a strong social component: people learn coping strategies from social support groups, such as friends, family, and co-workers. In collectivist cultures, norms, traditions and beliefs are also influential (Mesquita, Feldman Barrett, and Smith 2010). A body of work in multiple journalistic cultures and contexts has suggested that, when under duress, journalists form professional communities and display peer support, shared values, and teamwork (Dworznik-Hoak 2020; Gonz� alez Mac� ıas 2021; Novak and Davidson 2013;Obermaier, Hofbauer, and Reinemann 2018;Osofsky, Holloway, and Pickett 2005). They turn to trusted communities of colleagues and evoke public service beliefs that anchor professional identities to support adaptive coping. In places where uneven rule of law leaves journalists without protection from police or the courts, collaborative reporting and forming alliances with civil society organizations to demand government protection seem to strengthen journalists' safety (Cueva Chac� on and Saldaña 2021; de Le� on Vazquez et al. 2018; Gonz� alez Mac� ıas 2021; Gonz� alez de Bustamante and Relly 2021; . Studies documenting coping strategies thus suggest that professional solidarity, shared values, and teamwork elevate all dimensions of safety, by supporting physical safety through training for prevention and collaboration, and strengthening mental fortitude through camaraderie. Coping strategies can be adaptive and maladaptive (Buchanan and Keats 2011). Adaptive coping strategies are typically problem-oriented and include active planning and seeking social support. Maladaptive coping is usually associated with avoidance (such as denial, suppression of emotions and thoughts, and social withdrawal) as well as alcohol and narcotics abuse. The coping approach adopted, therefore, can result in either resilience, compliance or avoidance with the ultimate outcomes being either resistance, submission or exit (see Figure 1). A problem-focused coping strategy that journalists resort to is to make changes to their reporting practices and overall role performance . Several studies explore autonomy-reducing changes in journalists' work performance as a response to stress-creating threats (L€ ofgren Nilsson and € Ornebring 2016;Hasan and Wadud 2020;Hayes and Silke 2019;Hughes and M� arquez-Ram� ırez 2017;Adams 2018;Chen et al. 2020;Larsen, Fadnes, and Krøvel 2021). Such changes include self-censorship and curtailing reporting activities, examples of "compliance" responses that mitigate risk (e.g., Post and Kepplinger 2019) and whose outcome is submission. Another compliance response of sorts is minimizing risk, e.g., by limiting interactions with audiences or hiding one's gender online. While these strategies may be functional and healthy for the person (adaptive), they can be considered dysfunctional for journalism's performance (maladaptive). This two-faceted nature of journalists' coping makes it a particularly relevant research subject. An extreme outcome of maladaptive coping is exiting the profession (see, e.g., € Ornebring and M€ oller 2018 or Stahel and Schoen 2020 on exit). Journalists' individual coping responses have also been found to include tactics to safely continue to work with autonomy. Successful coping processes create resilience, defined as a sustained process of "adapting well in the face of adversity, trauma, tragedy, threats or significant sources of stress" or put simply, 'bouncing back' (American Psychological Association 2012). Resilience is, therefore, a sustainable and effective way of dealing with risk. Based on literature, Bergstr€ om, Van Winsen, and Henriqson (2015) conceptualize resilience as the intra-and interpersonal ability to cope with adversity or to regain a previous state following a disturbance or abrupt shock, suggesting also that resilience can be practiced on both meso/organizational and macro/societal levels (see also Aven 2011). When journalists achieve resilience, the outcome is resistance.
Ultimately, journalists' autonomy and performance can be affected to a higher or lower degree by the safety threats they experience and the resulting stress as well as their ability to cope with threats and reach resilience. Hence, it is important to study closely the interrelation between risk factors, safety, work-related stress, coping and resilience, and journalistic autonomy and performance.

Conclusion
The documentation of threats to the journalistic profession other than violence is infrequent, the exposition of journalists' coping mechanisms to deal with work-related stress is somewhat incomplete Iesue et al. 2021;Monteiro, Marques Pinto, and Roberto 2016;Monteiro and Marques, 2017;Relly and Gonz� alez de Bustamante 2017), and the explication of risk factors that beget threats remains fragmented. Safety research does not often distinguish between causes and symptoms, state and non-state actors, and structural and individual factors; and it is rarely sensitive to issues of gender, race, ethnicity, and sexual orientation and to intersectionality in general. Herein lies the significance of advancing both theory and empirical knowledge on journalists' safety.
Based on this diagnosis of a vivid, yet scattered research field, our aim is to offer a comprehensive and holistic conceptual framework for the study of journalists' safety. Drawing on a vast array of literature across multiple disciplines, our framework explicates the roots, risks, consequences, and requirements to advance the safety of journalists, culminating in an integrative model that conceptualizes pathways for studying journalists' safety ( Figure 1). The various aspects of journalists' safety outlined in this model are ensconced within the dynamics of power.
Occupational safety comprises physical, psychological, digital, and financial threats. These threats are moderated by societal, institutional or organizational, and individual risk factors (macro, meso, and micro levels), influencing how safe journalists are/feel (material and perceptional levels). The threats, in turn, may lead to work-related stress, creating a need for coping. If successful, coping builds resilience and allows the journalists to prevail in a threatening situation, even if their perceived safety is continuously compromised. Depending upon whether coping strategies protect or reduce editorial autonomy, coping may allow journalists to continue performing their roles and serving their societies.
The literature highlighted a key essential for coping and building resilience: the social resources journalists gain from their occupational networks of solidarity, or their social capital as defined in Brambila (2018). If systematic support is scarce, which is often the case, and because journalists at risk often share positions and occupational identities strengthened through adversity, the most important source of fortitude seems to be bonds of solidarity between colleagues in similar situations and sharing similar values. Other societal actors can also provide support. Journalism education and advocacy groups can support journalists in building peer-networks beyond workplaces and also include the issues outlined in this article in journalism curricula. In addition, policy makers can have an important voice in extending the space in which journalists can maneuver despite the fact that countries may differ in their willingness to support journalism, financially and otherwise. Journalists' perception of how safe they feel, how they assess their resources for coping, and how much support they receive from education, advocacy, and policy-making groups is, for most countries, still an open empirical question to be investigated in future studies.
Based on our conceptual model, we assume that the more the power dynamics shift in the direction of journalistic voices, the more risk factors they face, the less safe their occupational activity, the more stress they will experience, and the more resources they will require for coping. Bearing in mind that our definition of safety comprises four aspects of occupational safety (see Table 1), this implies a rather novel insight that journalists in liberal democracies, who place higher value on freedom of speech and economic liberalism, may be confronted with a less secure occupational situation -with an important exception: In regimes in which the power dynamics privilege the ruling elite, the occupational safety of critical change agents, and of interventionist and watchdog journalists will fall short of the safety levels of their colleagues in liberal democracies. The most desirable outcome of the process described in Figure 1 seems to be resistance. However, we have to bear in mind that the underlaying perspective is a profoundly Western one; it implies that ruling powers and journalists do not or should not collude. In cultures that highly value social harmony, compliance would be an equal or even a preferred strategy, as such a way of coping is beneficial to both journalists and society. This point illustrates how important it is to acknowledge the perspective through which one classifies a coping strategy as either adaptive or maladaptive, and we strongly encourage future studies to focus on how strategies can interact. Another important implication of Figure 1 concerns policy-making. When systems are in transition, decision-makers should be cognizant of the impact governance has on journalists' safety, too. Since resources available for coping represent a pivotal point in our conceptual framework, decision-makers need to make sure such resources are provided. Notes 1. See, for e.g., UNESCO, https://en.unesco.org/themes/safety-journalists/observatory?page=50; the Committee to Protect Journalists, https://cpj.org, or Reporters Without Borders, https:// rsf.org/en.

2.
Journalists' safety can also be endangered as a result of routine organizational processes (Bergstr€ om 2015, 131) not specific to the journalistic profession, such as being in a car accident while executing work-related tasks. Such general work-related risks are not part of this paper's scope.