An ‘attractive alternative way of wielding power’? Revealing hidden gender ideologies in the portrayal of women Heads of State during the COVID-19 pandemic

ABSTRACT This paper explores the gendered discourses of the – seemingly favourable – media coverage that certain Heads of State received for their handling of the COVID-19 pandemic. Looking at media reports published in different English-speaking outlets in the US, the UK, India, Bangladesh, Singapore, New Zealand, Australia, and Ireland, and using multimodal feminist critical discourse analysis, we identify and describe strategies that on the surface appear to challenge hegemonic – and largely masculine – discourses of leadership. Upon closer scrutiny, these superficially complimentary reports rather reinforce and naturalise discriminatory gender ideologies, and, as we demonstrate, they do so to different degrees along a continuum of essentialising, contextualising, and problematising. We critically discuss the discursive and visual processes involved and show that complimenting these leaders on their performance compares them against a masculine norm to construct their leadership as ‘alternative’, exceptional, and hence marked. This gendered portrayal of political leadership in times of crisis illustrates how the discursive construction of identities, responsibilities, and relationships during COVID-19 largely hinges on power relations and political ideologies that systematically disadvantage and undermine women. The purportedly positive form in which this occurs makes it particularly difficult to challenge and subvert these discriminatory discourses and their underlying gendered ideologies.


Introduction
The political leadership displayed by women Heads of State was the topic of numerous media reports in the first 18 months of the COVID-19 pandemic that were published in English media reports.Under headlines such as 'Why More Countries Need Female Leaders' (Bear & Agner, 2020, US News); 'What Do Countries with The Best Coronavirus Responses Have in Common?Women Leaders' (Wittenberg-Cox, 2020, Forbes) and 'Why Are Women-Led Nations Doing Better with COVID-19?' (Taub, 2020, The New York Times), news media discussed and often lauded the benefits of what they referred to as 'female political leadership', often making claims that women Heads of State performed better in times of crises.Over the course of the pandemic, this topic was expanded to CEOs as well as researchers and health activists, for example 'Have female CEOs coped better with COVID than men?' (Goswami, 2020, BBC).
This widespread media attention understandably prompted academic engagement with the topic.Two published studies (Aldrich & Lotito, 2020;Windsor et al., 2020) however come to somewhat different conclusions from most media articles, as no simplistic correlation between gender and performance is identified, and the authors suggest that other factors, such as a country's norms and values (Windsor et al., 2020) or the overall number of women in legislation (Aldrich & Lotito, 2020) appeared to be more relevant.
The overt and public display of approval that women in leadership positions have received during the COVID-19 pandemic, is, however, not without precedent.Sheerin and Garavan (2022) describe a very similar phenomenon during the 2008 financial crisis, where women were lauded as the possible salvation to the problems caused and faced by the global banking industry.The authors conclude that no significant change followed and that soon after the financial crisis abated, media depictions reverted to decry the lack of women leaders, framing this lack as a women's issue and encouraging women to 'improve themselves to become leaders' (Sheerin & Garavan, 2022, p. 10), by acting more like the men around them.
The explicit linking of gender and leadership behaviour is striking in both cases and runs counter to the de-gendering of leadership that research has been calling for (e.g.Holmes, 2006;Schnurr, 2022;Schnurr & Mak, 2010).In this paper we explore this discrepancy by critically analysing the ways in which Heads of State are discursively and visually constructed in 40 English media reports published in different outlets in the US, the UK, India, Bangladesh, Singapore, New Zealand, Australia and Ireland during the COVID-19 pandemic.We do this through conducting a feminist critical discourse analysis (FCDA) (Lazar, 2007) demonstrating how, through a superficially complimentary construction of womenas in the quote from a media report reproduced in this paper's titlehegemonic masculine discourses of leadership are reinforced, and discriminatory gender ideologies are strengthened rather than challenged.
Given the critical stance that this paper takes towards the gendered construction of women in leadership positions, and the persisting but highly problematic gendered norms, expectations and assumptions around leaders and leadership that are reflected (and reinforced) in the current use of terminology in many academic and lay outlets (see also our critical discussion of the 'think leader, think male' discourse below), we have made the decision to refrain from using the marked expression 'female leaders' in this paper.Rather, in an effort to avoid such linguistic marking and hence marginalisation of women in leadership positions, we will use the terms 'leader' or 'Head of State' to refer to women holding such positions.When we occasionally talk about men in leadership positions, we make this explicit by adding the gendered descriptor 'male'.We believe that making these linguistic choices is important as they make visible (and thereby resist and challenge) some of the largely overlooked linguistic processes that contribute to the gendering of leadership.This approach is also consistent with two important principles of FCDA: research activism and reflection (Lazar, 2007), which we address by carefully considering and actively changing how our own ways of speaking impact discourses of leadership.
In what follows, we first discuss some of the relevant literature on leadership and gender, before outlining the methodology and theoretical and analytical framework used in this study.We then critically analyse and discuss a selection of discursive and visual representations of Heads of State during the COVID-19 pandemic.We end the paper with a discussion of the implications of these observations for leadership and attempts to de-gender it.
Leadership and gender: think leader, think male It is widely acknowledged that leadership is not a neutral concept but that it is inherently gendered (Fletcher, 2004) and characterised by a 'male bias' (Hoyt & Burnette, 2013).This male bias of leadership is reflected, for example, in the 'think leader, think male' discourse that prevails in many public (and even academic) domains, as well as in the observation that mainstream leadership discourse tends to 'celebrate [ ] masculine norms such as competition, aggression, and individualism' (Alvesson & Spicer, 2014, p. 51), while backgrounding 'more feminine qualities such as empathy, capacity for listening, relational skills and so on' (Ford, 2006, p. 96).As a consequence, leaders around the globe are traditionally portrayed as 'tall, handsome, white, alpha males (of) privilege' (Grint, 2010, p. 69), while other types of leaders and leadership are often overlooked (see also Clifton et al., 2020;Schnurr, 2022).These issues are, however, not restricted to specific geo-political contexts or culturespecific ideologies, but are of a global nature (e.g.Mullany & Schnurr, 2022).
This hegemonic masculine notion of leadership seems to prevail in spite of the socalled feminisation of leadership, which acknowledges that behaviours and attributes typically associated with womensuch as 'collaboration, sharing, and teamwork' (Fletcher, 2004, p. 653)are valid and effective ways of doing leadership.However, this has not necessarily led to a de-gendering of leadership: rather, masculine discourses of leadership continue to dominate and to be considered as the norm around the globe.This seems to be particularly true for the portrayal of leaders in the media.

Political leaders in the media
Previous research has shown that there are significant differences in media reports on men and women in leadership positions (e.g.Cameron & Shaw, 2016;Garcia-Blanco & Wahl-Jorgensen, 2012;Mavin et al., 2010;Walsh, 2015).Women are often discussed in terms of family, looks and their nurturing qualities, tying them to submissive positions in society and heavily drawing on gendered stereotypes.For example, more space tends to be allocated to discussing the physical appearance of female politicians, and which is then frequently used to make character and competence judgements (Baxter, 2018;Cameron & Shaw, 2016;Campus, 2013;Tischner et al., 2021;Walsh, 2015) thereby trivialising the politicians' achievements (Mavin et al., 2010).Women in or running for high political office are also often framed as outsiders, as those who 'clean up' politics, as extensions to male partners and senior ministers rather than as politicians in their own right (Garcia-Blanco & Wahl-Jorgensen, 2012), and as caretakers (Campus, 2013).Further to this, narratives around their success often cast them as accidental leaders, whonot fully belonging to the male clubs that surround themgot lucky in their success, or just happened to be in the right place at the right time (Campus, 2013).
Moreover, women in politics are frequently constructed as particularly empathetic and caring, for example, 'as someone who can heal the country and bring peace and reconciliation' (Campus, 2013, p. 45).In line with these framings, media reports also often describe these politicians as emotional, gentle, and attractive, and tie them to issues that are stereotypically associated with womanhood, such as health, education, and welfare.Male politicians, by contrast, are often portrayed as intelligent, strong, aggressive, and linked to topics stereotypically associated with manhood, such as finance, taxes, and defence (e.g.Aaldering & Van der Pas, 2020; Kittilson & Fridkin, 2008).Such gendered reporting inadvertently positions men as better leaders because the attributes and characteristics assigned to them correspond to the hegemonic masculine notions of leadership discussed above.As a consequence, 'gendered media constructions highlight how political leadership remains a non-place for women' (Mavin et al., 2010, p. 564).
In an analysis of media coverage of the 2015 General Election in the UK, Cameron and Shaw (2016) noticed that while most commentators hailed the increasing number of women in senior political roles as 'a positive development ', '[b]eneath this overt enthusiasm, though, the rhetoric of the press often covertly reinscribed stereotypical, sexist and on occasion misogynistic propositions about women' (p.107).These findings are supported by Baxter (2018), who, in a comparison of different newspapers in the UK, found that although more conservative tabloid papers tend to draw on gendered stereotypes relatively explicitly in their reporting of women in power, even those newspapers with an openly liberal and feminist stance reproduce the hegemonic male leadership discourses described above that marginalise and exclude women.
Such gendered and stereotyped media reports are also remarkably similar across a wide variety of socio-cultural contexts and media outlets.Campus (2013), for example, observed great commonalities in the way politicians around the world are discussed by the media.Similarly, in an analysis of media coverage in France, Italy, Spain and the UK of the first majority female-led government in Spain, Garcia-Blanco and Wahl-Jorgensen (2012) observed that 'while certain discourses celebrate women ministers for their (symbolic) emancipatory value, others judge them by their physical appearance or their performance as wives, mothers, and mothers-to-be' (p.422) (see also Bengoechea, 2011).
However, to the best of our knowledge, none of the articles discussing the discursive constructions of politicians in the media have systematically investigated the visual representations of politicians.This paper addresses this gap by critically discussing both the discursive and visual construction of Heads of State during the COVID-19 pandemic by English-writing media.In terms of the visual depiction of political leaders in the media, large scale empirical investigations suggest that news reporting generally favours singleperson photos, even if the text engages with several people (Kwak & An, 2016).Womenin general, not just in leadership positionsare often presented as happy, calm, smiling, and submissive, while men tend to be depicted as sad, excited, and dominant (e.g.Kwak & An, 2016;Rodgers et al., 2007).With this, women's bodies become objects of consumption existing to be looked at: '[i]n contrast to white male leaders' bodies, which enjoy the privilege of not being "seen", "women's identities, gender and bodies are routinely tied together and attributed meanings antithetical to leadership" (Sinclair, 2013, p. 242)' (Bell & Sinclair, 2016, p. 325).This emphasis on women's bodies further contributes to women being constructed as exceptions and outsiders on the politicalor leadershipstage.This paper takes these observations into consideration when critically investigating the construction of Heads of State during the COVID-19 pandemic as outperforming their male colleagues.

Methodology
We collected 40 media reports published in English between March 2020 and May 2021 that discuss 'female leadership' during the COVID-19 pandemic (see the overview of primary data sources).We conducted an additional search in September 2021 that did not result in any newer articles, indicating that interest in this topic had somewhat abated.
The articles were found using two separate procedures.Most were identified through a Google search using different keyword combinations of female leaders, COVID, news, leadership, women, pandemic.Results were filtered manually to select only media reports, which tended not to occur beyond the first seven pages of results, after which hits tended to be unrelated.We are aware this method has its flaws and had initially followed a more standard procedure by identifying large news outlets and doing a keyword search in their electronic archives.This procedure resulted in several hits per outlet, but the scope of the data was considerably restricted in comparison with the Google search procedure.This enabled a geographically wider spread of results and more diversity in terms of outlet type in our sample.
Overall, we collected articles from 34 different outlets located in the US and UK primarily but also in India, Bangladesh, Singapore, New Zealand, Australia, and Ireland.Outlets include everyday broadsheet newspapers such as The Guardian, Indian Express, and The New York Times, tabloids such as The Daily Star and Metro, and online outlets such as Inc.Magazine and TED Ideas.The sample also includes specialist outlets, including business and management outlets such as Forbes, Fortune, and Management Today, medical outlets such as Medical News Today, and political outlets such as Modern Diplomacy.
Since our sample does not include many smaller or local news outletsdue to the fact that Google search engines do not promote them in the same way as larger, more global outletsand due to our focus on English-only reports, we are making no claim to have captured all articles on this topic or that the reports that we include are representative.However, what the data sample captures is a fairly widespread and surprisingly homogenous discourse around women, leadership, and COVID that circulated in the English-reading-world in a wide variety of geographical locations.Moreover, the often global distribution, as well as the reputation and size of the outlets and opinion pieces we have collected, enabled us to identify some of the important trends and patterns through which issues of leadership and gender tended to be discussed in our sample during the COVID-19 pandemic.

Data analysis
We used Feminist Critical Discourse Analysis (FCDA) to investigate the discursive construction of Heads of State in these media reports.Lazar (2007) explains that FCDA aims to uncover the 'complex, subtle, and sometimes not so subtle, ways in which frequently taken-for-granted gendered assumptions and hegemonic power relations are discursively produced, sustained, negotiated, and challenged in different contexts and communities ' (p. 142).This seemed particularly relevant to the superficially positive and complimentary discourse that can be found in many of the media reports we collected.The aim of our subsequent analysis is thus to identify, make visible, and ultimately challenge the gender ideologies, hegemonic power relations, and taken-for-granted assumptions that were mobilised and underpin the news reporting about Heads of State during the COVID-19 pandemic.We do this by exploring both the text of the media reports and the images accompanying the headlines.The value of multimodal analyses within FCDA is stressed by Lazar (2007) who argues that this contributes to more holistic critiques of the discursive constructions of gender inequality.
Our analyses were guided by a series of critical questions, including: . How (and with whom) are Heads of State portrayed in the texts and images?
. How and to what extent do the texts and images reinforce or challenge hegemonic masculine discourses of leadership? .Do the texts and images implicitly or explicitly orient to and invoke overarching gender ideologies that endorse or resist the gender order?
For the discursive analysis, we first analysed how Heads of State were constructed through: (a) the attribution of specific traits and characteristics; (b) the degrees to which their gender was foregrounded, and (c) the ways in which they were positioned in relation to others.The trends and patterns we identified were remarkably similar across the different texts and led us to devise a continuum along which these texts categorise, position, and ultimately construct leaders.We discuss this continuum in more detail below and provide a detailed analysis of the discursive strategies along the continuum.
For the visual analysis we focused on the headline image of each article. 1 The images were stored and coded separately according to type: portrait, composite, stock image, or artwork.We followed similar procedures as in the textual analysis and focused on: (a) the specific traits and characteristics the images constructed for the Head of Statefor example, caring and empathic; (b) the degree to which their gender was foregrounded, and (c) the ways in which they were positioned in relation to others.In addition, we drew on Machin's (2004) distinction between news images as either bearing witness to a specific historical event or as symbolic representation.Machin (2004) argues that news photography historically favoured bearing witness but has increasingly shifted towards symbolic representation.We think this is particularly significant in a feminist approach to news photography as it renders specificwomen-doing-specific-things to a generic and symbolic category.To analyse this use of symbolic images versus images that 'bear witness', we have focused on modality, that is, the degree to which an image appears natural or editedas heavily edited images lean more towards symbolic representation, and to making generic rather than specific claims (Machin, 2004).As with the textual analysis, we identified clear patterns in the images used that cut across outlets and geographies and helped in seemingly innocuous ways to construct the image of Heads of State as different by comparison with a male norm.

Discursive construction of Heads of State
The patterns we identified in our discourse analysis were strikingly similar across the different texts, which is in part due to a high degree of intertextuality across our overall corpus.That is, there were a number of studies and reports published in the time period of our data collection (e.g.Garikipati & Kambhampati, 2021;Windsor et al., 2020;Zenger & Folkman, 2020) that were picked up and discussed by different media outlets, as well as interviews or statements by public figures that were then drawn upon in various articles.The discursive strategies deployed in the texts to construct leaders as 'successful' (or not) can be placed on a continuum, ranging from explicitly essentialising women (and men) in leadership positions at one end to (seemingly) problematising the notions of 'female' and 'male' leadership at the other.In between those two poles, various degrees of contextualisation of Heads of State in the COVID-19 pandemic can be identified, as is illustrated in Figure 1.
The boundaries between these strategies are not clear-cut, and even within individual articles, different strategies are foregrounded at different points.We could identify these three strategies in all articles to varying degrees, regardless of the type of media outlet.Bar a few exceptions, the discursive strategies through which the Heads of State are constructed in these texts contribute to perpetuating and solidifying hegemonic masculine notions of leadership and the association of leadership with a masculine norm.In line with FCDA, we focus on identifying how such ideological assumptions 'entrench asymmetrical relations of power' (Lazar, 2007, p. 372) reinforcing the gender order (Connell, 1987), and discursively construct women leaders as deviant from a male norm, and therefore exceptional.We discuss the dominant discourses identified in our dataset by illustrating each aspect of the continuum in turn.
Essentialising 'women leaders' and 'female leadership' Across the articles, an overarching discourse of gender difference is endorsed through repeated reference to 'women leaders.'This discourse draws heavily on the so-called 'difference' approach in language and gender research (e.g.Tannen, 1990), according to which men and women are fundamentally differentlargely due to differences in their socialisation.In most cases, this difference is simply projected onto leadership styles.Such binary and uncritical assignments of leadership styles, however, further contribute to essentialising men and women as homogenous groups, often achieved through the ascription of specific behaviours and traits stereotypically associated with masculinity and femininity to these Heads of State (such as being strong-minded and caring, respectively).Even attributes listed that are not necessarily linked to femininity or masculinity (such as being knowledgeable and truthful) contribute to an essentialisation of leaders through homogenising women as better leaders.
Although such characteristics and traits are often used to compliment the women about their 'different' but 'effective' leadership performance, such a construction both implicitly and explicitly compares the women's performance to a male norm.As a consequence, stereotypes of what are often considered to be intrinsic female characteristics that make for better leaders are perpetuated and reinforced, rather than challenged and resisted, endorsing hegemonic masculine notions of leadership.
Example 1 is an extract from a much-cited article in Forbes that epitomises the dichotomous, homogenising approach to the portrayal of leaders at this end of the continuum.In this extract, Wittenberg-Cox (2020) elaborates on a list of proposed characteristics of 'female leadership'.To illustrate 'love', or showing empathy and care, she cites Erna Solberg (Norway) holding a children-only press conference.
Example 1. Essentialising women leaders (extract from Wittenberg-Cox, 2020) In lines 1-2, the action of addressing children directly in a press conference is complimented as original 2 , foreshadowing further 'humane innovations' that 'female leadership' would 'unleash' (line 2).At the same time, however, this idea is also framed as obvious (line 1) and 'simple' (line 2).Linking the consideration of children's needs directly to women, the text posits that for them, it is a 'simple' and obvious choice to take such aspects into account.While this is complimented as innovative, the discursive construction of Heads of State as caring perpetuates discourses of 'female caregivers' and 'woman as mother' (Sunderland, 2004) and thus endorses a discourse of women as responsible for childcare in line with the gender order.According to Connell (1987), the gender order refers to 'a historically constructed pattern of power relations between men and women and definitions of masculinity and femininity' (p.168).It dictates, for example, what roles are typically assigned to men and women, and what behaviours are considered to be feminine and masculine in a particular group, society, or even on a global scale.For example, in many Western societies, displays of power and authority are closely associated with masculinity, and as a consequence, (senior) politicians are traditionally expected to be men.The author constructs Solberg as close to the audience even in video-streaming and as 'hold[ing]' the viewer in a 'heart-felt and loving embrace' (line 3), thus foregrounding a women's body and physical touch in the reporting (Bell & Sinclair, 2016).Positioning this leadership as stemming from 'an alternate universe,' the author invokes a gender difference approach (Tannen, 1990), echoing the popular myth that 'Men are from Mars and Women are from Venus' (Gray, 1992).Whereas academic inquiry has moved away from uncritically adopting such an approach (e.g.Mullany & Yoong, 2017) this excerpt is a prime example of the prevalence of such stereotypes in public discourse.In lines 4-5, the author continues to invoke the difference paradigm by comparing leadership styles from an 'alternate universe' with traditional ways of doing leadership displayed by 'strongmen' (line 5) engaging in 'authoritarianism' (line 6), giving examples of selected male Heads of State (lines [7][8]. Examples on this end of the continuum largely rely on problematic comparisons between men and women as homogenised groups, without challenging underlyingand largely essentialistconceptualisations of gender.Rather, such gender ideologies are appropriated to support an overall positive framing of stereotypes, positioning the qualities women allegedly possess as crucial to solving crises, for example as stereotypically female qualities that 'men may want to learn from women' (Wittenberg-Cox, 2020).In other words, the discursive construction of these Heads of State as deviant from (male) hegemonic notions of leadership upholds systems of power in which the available subject positions for these Heads of State are firmly in line with the gender order.More generally, the Heads of State in our corpus are collectively described as: . Communicating well .Being knowledgeable and truthful, explaining scientific causalities .Evidenced-based decision making (trust in science) .Being decisive and confident (quick decision-making) .Showing emotions, love, and empathy .Being caring and kind, interested in other people's wellbeing .Focusing on community and people rather than economy, reaching out to vulnerable groups Many of the leaders we mention above are linked to several of these different qualities or characteristics, and in doing so, the texts contextualise these leaders and their leadership styles to varying degrees.We elaborate on this in the next section.

Contextualising women in leadership positions
While this essentialising of women and men occurs across all texts, other discursive strategies contextualise 'women leaders' and associated notions of 'female leadership' to varying degrees.In such cases, a more or less explicit link is established between leadership, gender, and broader societal contexts: for example, where authors consider if, and why, countries with women in leadership positions have fared better during the pandemic.
Often this includes generalising and stereotyping claims that are problematic but also acknowledge contextual factors (in addition to gender) that impact on leadership performance, as illustrated in Example 2, published in US News on International Women's Day.
Example 2. Contextualising women in leadership positions (extract from Bear & Agner, 2020) The extract illustrates well how contextualising leaders only seemingly breaks with dominant discourses that endorse the gender order.In lines 1-3, the authors list various issues claiming that these will benefit from including women in decisionmaking processes.The ideological assumption that the inclusion of women will automatically advance issues of gender (in)equality is linked to various other problematic claims, for example that all women can provide expert guidance on these issues and that addressing them is a women's issue due to their 'importance to women's resilience' and 'ability to recover from this crisis' (line 4).This places the emphasis of solving such problems on individual women at the expense of acknowledging structural inequalities (Rottenberg, 2014).In lines 5-6, this discourse is further endorsed through the argument that the 'people in charge' must 'share the perspectives of those they serve' (lines 5-6), for example that there is a need for more women in charge to tackle gender issues.This makes an implicit assumption that women share a perspective on gender issues that can simply be represented by including 'a woman at the head of the table' (line 1), while ignoring the diversity of women and the heterogeneity of opinions, for example along party lines.
The authors go on to endorse the 'incredible female leadership' (line 7) a homogenised group of women has shown, 'working hard' to 'get us through this crisis' (line 8).This acknowledges the work of the leaders, but the use of 'incredible' also marks women as deviant from the (male) leadership norm, reinforcing hegemonic masculine discourses of leadership.
Even though important issues are at stake here that disproportionately affect women, the ways in which they are discursively associated with women in leadership positions in this and other examples largely orient to gender ideologies that endorse the gender order rather than challenge it.The underlying assumption that women in leadership positions advance gender equality (see also Batha, 2021) not only locates the responsibility for addressing inequalities at an individual rather than structural level, but also holds in place masculine hegemonic notions of leadership that are not linked to (advancing) issues of gender inequality in any way.
These contextualisation strategies further solidify hegemonic notions of masculine leadership through other associated claims, such as the claim that including women in decision-making processes leads to better outcomes (e.g.Sandoiu, 2020), and the claim that women are Heads of State in more egalitarian societies, which display more trust in government and hence, government measures (e.g.Henley & Roy, 2020).The underlying assumption that women comprise a homogenous category of people remains however unchallenged.With this, akin to the first category of the continuum, women (and men) not only continue to be essentialised but are constructed in line with binary and dichotomous gender ideologies that uphold the gender order.
(Seemingly) problematising 'female' and 'male' leadership At the right-hand end of the continuum, we have identified discursive processes that problematise the very notion of 'female leadership'.Examples of such problematising include consideration of the fact that a 'disproportionate focus on women' (Wenham & Herten-Crabb, 2021) distracts from other aspects of successful leadership in the pandemic, as well as an acknowledgement that notions of effective leadership may be changing and may include skills and attributes that are not traditionally associated with masculine notions of leadership.However, just as we have observed in the previous two sections, this problematisation does not necessarily mean that the underlying ideologies and gender stereotypes described above are being challenged.Rather, often even these more critical or nuanced discussions of leadership also rely and draw on hegemonic masculine discourses around leadership and gender ideologies that propagate rather than challenge and resist the 'think leader, think male' mantra discussed above.
Example 3 from The Guardian illustrates this end of the continuum.At the same time, we find aspects from the other categories present, further illustrating how the authors of media reports shift their stance, sometimes even within the same paragraph.The article focuses on listing (female) leaders and their handling of the pandemic and concludes as follows.
Example 3: (Seemingly) problematising 'female' and 'male' leadership (extract from Henley & Roy, 2020) In the first paragraph of the extract, quoting experts as acknowledging that women are 'disproportionately represented to a rather startling degree' among countries managing the crisis well' (lines 2-3) works to endorse the claim that they have been more successful in handling the pandemic as well as to set up the counter-claim for caution against 'dividing men and women Heads of State and government into homogenous categories' (lines [3][4]. The authors therefore challenge the discourse of gender difference through problematising the binary categories 'men' and 'women' that form the basis of this argument.In lines 6-10, a sociologist is quoted who further contextualises this issue using the discursive strategies we identified in the previous section.She argues that the election of women leaders is linked to a 'political culture' (lines 7-8) that does not make 'stark distinctions between women and men' (line 9).
In line 9, however, the authors continue to endorse, rather than challenge, a discourse of gender difference.They do this by quoting an argument from the same sociologist that it is 'harder for men to escape 'the way they are expected to behave' as leaders' (lines 9-10).Seemingly problematising hegemonic notions of male leadership, the use of homogenising categories continues through the article, invoking gender ideologies that endorse, rather than resist the gender order.
In lines 10-13 of the extract, both stereotypically male and stereotypically female qualities are listed that are important to leadership.This seemingly contributes to a changing discourse of what a competent leader is (e.g.strong, decisive, as well as 'capable of displaying feeling' [line 13]), but the underlying assumptions rely on homogenised categories alongside the premise that women have inherent qualities (i.e.displaying feelings) that can change notions of good leadership.Thus, in propagating the notion that male leadership is the norm from which women deviate, and in making use of some of the stereotypes we identified above, these discursive strategies actually contribute to a gendering of leadership.In this way, instead of de-gendering leadership through focusing on what makes competent leaders and effective leadership, the discursive strategies we identified in the sections of the articles located on this end of the continuum contribute to reinforcing hegemonic masculine discourses of leadership.
Throughout the articles we find gender continually used as an interpretative category (Lazar, 2007) that gives meaning to the behaviours we observe with very few other explanatory variables being considered.In the next section we turn to the analysis of the images that accompany these texts about leadership during the COVID-19 pandemic, and we explore how successful politicians are constructed through visual means.

Visual construction of Heads of State
Heads of State are visually represented in three distinct ways in the articles: through photo portraits and composites, through stock photos not depicting famous but generic women, and through artwork.Only three articles featured no images, and while several articles featured more than one image, we are focusing in our analysis on the main headlining image, as it functions as the main attention grabber alongside the headline.

Photographic images: single portraits
The majority of articles ( 28) are accompanied by a photo of Heads of State, half of which are single portraits of a specific leader who symbolically stands for women in leadership positions more generally.There are clear preferences as to the chosen 'poster girl', with Jacinda Ardern (New Zealand) 9 times, Angela Merkel (Germany) 4 times and Tsai Ing-Wen (Taiwan) and Katrín Jakobsdóttir (Iceland) 1 time each, being the only Heads of State with individual portraits out of the 28 women who were in power in 2020/2021.The articles themselves then discuss more than just the one leader depicted, which supports the textual framing that women are a homogenous group and that therefore the picture of one Head of State can metonymically stand for all Heads of State.
While it is not unusual to use a portrait photo of a Head of State, this further functions to construct a country's leader as the sole one in charge, instead of framing them as surrounded, supported by, and collaborating with a large team of ministers, aides, and advisors.This supports a leadership discourse focusing solely on a singular leader, misleadingly attributing the success of a whole country to a single person (or rather their gender), a claim we have also found in the textual analysis.The only exception to this solitary depiction in our visual analysis occurs in two shots of Ardern, making these pictures stand out in our image corpus: In Example 4 3 , Ardern is seen with three male advisors; however, she is walking ahead of them with a resolute facial expression.The visual thus constructs her as in charge and ahead of the menand with this, still solitary.Atypically, Example 5 4 depicts her as interacting happily with children and appears to provide visual evidence that connects women with children and with caring and nurturing roles.The image has evidently been chosen for its symbolic value as the lack of masks makes it clear that it predates COVID and thus does not 'bear witness' (Machin, 2004) to any specific event discussed in the text.The article itself makes no explicit claims about Arden at all, nor does it connect Heads of State or their success to their engagement with children.Instead it focuses solely on death numbers in different countries in a sober and neutral tone.The image then becomes central in achieving the ideological work that we encounter across our corpus: it constructs Ardern as caring, friendly and engaged with children and its positioning next to a text about performance and death numbers works to allege that these attributes are somehow relevant, without this claim having to be made explicitly in the text.
Leaders are also smiling in about half of the images, which constructs women in leadership positions as (disproportionately) more friendly, joyful, and relational than the typical male politicianin that around three-quarters of news photos feature neutral facial expressions (Kwak & An, 2016).Similar to the text of the articles, these images construct women as subtly different from their male counterparts.This becomes even more evident in the second type of news photo used.

Photographic images: composites
The second type of news photo departs from the solitary depiction of politicians through grouped portraits that have been edited to show several Heads of State together.In these images attention has been paid to assembling women to indicate they somehow belong together and share certain attributes (see also Cameron & Shaw, 2016).
Example 6-8 illustrate are examples of these grouped photos.While executed with varying degrees of finesse, they all visually construct cohesion between the depicted Heads of State, not just because the leaders are grouped but through various tools including gaze, colour, and composition.In Example 6 5 , five different Heads of State are grouped in front of a globe.Their bodies are rendered in monochrome, and five disparate -but tonally complementary -colours have been applied to their jackets.In line with our observation that Ardern appears to be the poster girl for leadership, size marks relative importance within the group: she is much larger than the others, with Merkel somewhat smaller but substantially larger than Tsai Ing Wen (Taiwan), Mette Frederiksen (Denmark) and Sana Marin (Finland).Other differences lie in the juxtaposition of (confident and/or happy) facial expressions and gestures, Ardern smiling benevolently upon the group, Merkel in serious explaining mode, Ing Wen with clenched fist and so on.It is tempting to see these as a representation of (diverse) leadership styles, but despite distinctions in size and style, what makes them successful is the same: they are all women.Furthermore, the colour-coordination and physical proximity in the composite portrait constructs them as having relationships with each other, and arguably invokes the look of a girl band (Deane, 2021).Example 7 6 , three headshots in a row, also constructs such relations and similarities through the use of edited proximity and colour-coordination, with Merkel and Jacobs in blue, framing Ing-Wen wearing black.The five photos that comprise Example 8 7 have been selected so that each woman's gaze has the same focal point, depicting them as if looking in the same direction, potentially sharing a vision and forming a united, homogenous group (Cameron & Shaw, 2016).These types of edited images provide further visual evidence for the claims made in the text: that women in leadership positions are fundamentally different from men; that they are more relational and also constitute a homogenous group sharing specific characteristics.
All these group portraitsby virtue of being heavily editedhave a lower modality than the solo portrait shots.However, some among these composite images have a particularly low modality.This lower modality is one way of moving images from the specific that is, editorial images that bear witness to specific eventsto the symbolic and iconic (Machin, 2004).This functions to make a claim about women in leadership positions more generally, portraying the Heads of State not as individuals but as metonymic representationsnot necessarily of their country, but of 'female leadership' more generally.
The gif in Example 9 (Figure 2) below is among the least naturalistic grouped portraits and was published on the online news/article version of TED Ideas.Photographs of Ardern, Solberg, Sheikh Hassina (Bangladesh), and Ing-Wen have been digitally transformed with cartoon photo effects.The group pose depicts the four leaders in close proximity and alignment, their gaze directed right-and-centre of the image, which helps to construct them as future-oriented and by implication even as 'the future'.
The image thus constructs these four different Heads of State as similar, aligned, and sharing a future vision.The gif is also animated so that their hair and veil appear to move in the wind, drawing attention to their physical bodies, while also positioning them as special, as icons of their time, and with that, very much as exceptionsnot ordinary women.
The similarity alleged in these images takes no account of country differences, geographical distance, party lines, and ideological political positions.Instead, gender is constructed as the only salient category tying these Heads of State together, thereby further constructing gender as a meaningful interpretative category and continuing the homogenising discourse found in the text of the articles.
The remaining two categories, stock images and (non-portrait) artwork, push this low modality and symbolic representation of 'female leadership' to an even greater extreme.Symbolic representation through stock images Unlike leader portraits, stock images appear to bear witness to a specific phenomenon although they actually constitute symbolic representationsthrough the use of generic spaces, attributes, and people (Machin, 2004).While this makes the photos highly versatile (and thus commercial), it also limits what they communicate to a few stereotyped messages.
In Examples 9-10 (Figures 3 and 4), we see generic workplacesrather than specific or symbolic political settings.They seem to be used in articles that move away from solely discussing political leadership to discussing leadership more broadly and in other domains, thus emphasising generic women over specific women.At the same time, there is little connection to the articles themselves, for example, Example 10 (Figure 4) is immediately preceded by the line 'There needs to be more nuance in discussions around gender and leadership' and is subtitled 'Woman leading a meeting'.This, and all stock photos in our sample, lack a clear link to political leadership, and instead generically testify to the fact that women participate in the workplace and sometimes sit at the top of the table.Example 9 (Figure 3), preceded by the headline 'The COVID crisis shows why we need more female leadership' rather depicts 'female followership' in that it shows an audience of several women.As with most of these stock images, a direct link to the text seems to be missing and the image seems to communicate little beyond the fact that women exist in workplaces.
One very noticeable feature about these images is that they construct a women-only narrative.This further supports the problematic essentialising tendencies and the exaggerated focus on women as a homogenous group identified in the discourse analysis aboveand at worst might feed into a 'women replace men at the top of the hierarchy' discourse that is antithetical to feminist thinking (Lazar, 2007).

Symbolic representation through artwork
The artwork accompanying some of the articles, by contrast, does include representations of men.The men are typically portrayed negatively compared to women who are depicted as making smart choices.
Example 11 8 is a pen-and-ink style cartoon illustrating an article called 'Women leaders and alpha males up against COVID-19'.It makes explicit reference to different forms of traditional (male) leadership via the costumes that the blue-masked men weardressing up and behaving as caveman, Napoleon, tank driver, and syringe-dart player, literally fighting comic COVID-19 molecules while the pink-masked women are surrounded in a huddle shielding children.This constitutes a blatant invocation of the gendered discourse that ties women and men to traditional roles and domains through their actions and use of stereotypic colours.While the article and the image criticise men's behaviour in the 'battle' against COVID-19, they nevertheless portray this fight as divided strictly in line with the gender order (Connell, 1987).
Example 12 9 is the most abstract artwork in our corpus.It consists of a design based on one female silhouette and one male silhouette, blocked in scarlet and teal respectively (bolder colours than pink and blue, but still suggestive of them).In these colour-coded silhouettes, the male head is fissured by a vertical crack through the skull, while the female head is intact.This image thus powerfully supports the main message of the article, namely that women are better-functioning (i.e.not broken) leaders during the pandemic.Both artworks clearly draw on stereotypical gendered discourses, and despite the positive notes in the respective articles, focus more on male failures than female successes.
Overall, the images construct Heads of State in similar ways as the texts: as different to male leaders, as a homogenous group, and as having specific traits such as being relational, nurturing, caring, and ahead of, or more successful than, men.Low modality portraits also appear to glamourise Heads of State, and overall images have been selected to be flattering, lending evidence to the salience of female bodies in public spaces and to the need to appear attractive to be able to claim competence and leadership skill.

Discussion and conclusion
It was the aim of this paper to identify and describe some of the discursive and visual strategies through which the media construct women in leadership positions around the world as handling the COVID-19 pandemic more competently than their male counterparts.In our critical analysis of texts and images in media reports published in 40 English-speaking outlets, we have illustrated that although some of these texts and images appear to challenge hegemonicand largely masculinediscourses of leadership, the ways in which Heads of State are constructed draws heavily on ideologies that systematically undermine and disadvantage women.In most cases, in spite of a positive and often complimentary portrayal, the women's performance is still measured and compared (even if often favourably) against a male norm, a discursive move that has long been criticised by feminist discourse analysts (Lazar, 2007).Through this positioning, the hegemonic 'think leader, think male' discourse is oriented to and strengthened.
Moreover, many of the attributes and characteristics that the texts and images assign and project onto these successful women reinforce rather than challenge traditional gender stereotypes (such as, for example, women being caring and collaborative) and thus speak to rather than challenge the gender order.We would argue that it is precisely these superficially positive and complimentary ways in which gender stereotypes and underlying gender ideologies are drawn upon that make it particularly difficult to challenge and resist them.This is further problematised by the issue of the so-called feminisation of leadership, according to which behaviours stereotypically associated with femininitysuch as 'collaboration, sharing, and teamwork' (Fletcher, 2004, p. 653)are increasingly acknowledged and celebrated as valuable for leadership in some industries (Appelbaum et al., 2003 ;Eagly & Carli, 2003;Fletcher, 2004).However, celebrations of this alleged 'female advantage' (Eagly & Carli, 2003, p. 807), similar to the discursive and visual portrayal of Heads of State in our data, do not necessarily benefit women in leadership positions.On the contrary, studies have shown that these behaviours are perceived as particularly laudable when displayed by men (Appelbaum et al., 2003;Brescoll, 2016).
The 'feminisation of leadership' discourse also over-simplifies and derails important discussionsfor example around attempts to de-gender leadership (e.g.Clifton et al., 2020) to stereotypical assumptions about how men and women allegedly do leadership.This focus on men versus women who allegedly bring certain characteristics and behaviours to the leadership table is counter-productive as it over-simplifies complex relationships and dynamics between underlying gender ideologies and the perception, portrayal, and actual performance of leaders.As a consequence, recent calls for a degendering of leadership (e.g.Clifton et al., 2020) are undermined by the ways in which these opinion pieces treat men and women leaders as if they were homogenous groups, and as if identifying as man or woman were in any way an indication of their leadership performance.
What is needed instead, we believe, is a media coverage that talks about leadership and gives credit to those who participate in leadership processes without making this about men versus women.Such an approach would also address one of the potential dangers of a de-gendering of leadership, namely, to withhold credit and recognition from women in leadership positions who have the potential to function as important role models.So, we would argue that instead of comparing how men and women do leadership, we need media coverage that identifies, brings to the surface, and problematises underlying stereotypical assumptions about leadership and gender when celebrating successful leadership.Such a coverage would challenge hegemonic masculine discourses as the benchmark against which leadership performances are being compared (see also Schnurr, 2022), and would instead focus on what constitutes effective ways of doing leadership in a crisis and explore which strategies work in what context.Following Garcia-Blanco and Wahl-Jorgensen (2012), we would argue that such a more egalitarian media coverage 'can only emerge if and when the representation of gender is rendered politically irrelevant' (p.437).We hope that the arguments and evidence presented here will be a first step towards this goal.Continued.
and media contexts.The critical nature of 'people-centred' questions in her research on graduate job interviews at the University of Auckland prompted her interest in how teamwork and leadership are interactionally achieved, as well as in the co-construction of gender and professional identity.