Understanding sovereign leadership as a response to terrorism: A post-foundational analysis

Introduction

This study seeks to better understand how notions of sovereign power as a response to terrorism are built and bolstered through use of the signifier ‘leadership’ and how such articulations of leadership foreclose possibilities for critical engagement and dissent. Recent years have witnessed a growth of scholarly interest in the sovereign power of public leaders, particularly in how such power is manifested in their domestic and foreign responses to terrorist attacks. Such work has developed understanding of sovereign power as inhering in ‘emergency’ (Honig, 2011) or ‘exceptional’ (Agamben, 1998, 2005) acts, where the sovereign judgements of leaders are pre-eminent and establish an ambiguous ‘zone of indistinction’ (Agamben, 1998: 19) between law and the authority of leaders, helping us to locate the underlying justification for acts such as torture, rendition, war and surveillance.

Some research exists connecting the rhetoric of leaders to claims and exercises of sovereign power (e.g. Bligh et al., 2004; Esch, 2010; Widmaier, 2007). These can be situated within the broader area of critical leadership studies (CLS) (Alvesson and Spicer, 2012; Collinson, 2011), where analysis can make visible workings of power achieved via a ‘particular authorized language’ (Alvesson and Spicer, 2012: 374) of leadership. Within critical and empirical studies of leader rhetoric, however, the tendency is usually to assume that the language uttered by leaders can be claimed as valid terrain for analyses of leadership. While acknowledging the validity and value of these studies, particularly in making more visible the ways in which leader status is constituted through language, we seek to accomplish something different, through a post-foundational (Cederström and Spicer, 2014; Marchart, 2007) and ‘categorical’ (Kelly, 2008) analysis of the word ‘leadership’ and how it is put to use in relation to terrorism.

Based on the above conceptual positioning, our guiding research question is as follows: ‘How is the signifier “leadership” used to foreclose meaning concerning notions of sovereign power in relation to terrorism?’ Adopting a post-foundational analysis, we operate on the assumption that the meaning formed around the word ‘leadership’ is accomplished through ongoing and contingent rhetorical acts. Such an analysis approaches ‘leadership’ as a significant signifier, which draws out the political, affective and ethical connections made by the speaker (Laclau, 2015). Methodologically, we address the research question through a purposive sample, the speeches, writings and press conferences of former UK Prime Minister (PM) Tony Blair, a key figure in the US/UK-led ‘war on terror’. Blair was a noteworthy leader: a proponent for military action in response to the 9/11 terror attacks in the United States and later for the invasion of Iraq and something of an outlier in these terms among leaders of governments in Europe. Recognised for his rhetorical abilities (Fairclough, 2000), Blair’s numerous public statements and speeches provide a rich bank of data for better understanding how the case for sovereign power as a response to terrorism is accomplished in language.

Building from our data, we theorise ‘sovereign leadership’ as the deployment of the signifier leadership in ways that foreclose language so as to normalise the discourse and acts of sovereign power. Henceforth, rather than the complicated formulation of ‘the signifier leadership in relation to the discourse of sovereign power’, we shall refer simply to ‘sovereign leadership’ as a means of expressing the imbrication of leadership and sovereign power, except where the analysis and discussion benefits from their conceptual separation. Our key finding is that sovereign leadership offers a (misleadingly) straightforward solution through contingent rhetorical connections and positioning, foreclosing language so that alternative responses to the complex problem of terrorism are shut down. Within the operation of sovereign leadership, we posit three foreclosing ‘moments’ (Laclau and Mouffe, 1985: 105), ‘differential positions’ (Laclau and Mouffe, 1985) that are drawn upon to bolster the urgency, affective salience and justness of this rhetorical response to terrorism: emergency, positivity and vulnerability.

Our contribution lies primarily within the area of CLS, in providing an analysis of the use of the word leadership within rhetoric and in a situation of perceived terrorist threat, unpacking some of the ways in which leadership seems central to the justification of sovereign power. We view our study as contributing to a demystification of language (Śliwa et al., 2012) by laying bare the articulatory systems established to convey a sense of sovereign leadership, a tendency we later posit as holding broader significance for future studies of leadership. The value of our contribution lies in the identification of a sovereign language of leadership and a deeper understanding of its operation, which in its articulation seeks to foreclose possibilities for alternative perspectives. Leadership is a ubiquitous word in organisational and social parlance – so much so that, in everyday use, it may be difficult for consumers of the word to discern how it is being deployed – and offering a means of better understanding its use therefore holds a value beyond the analysis of what people in leader positions say. A secondary contribution lies in the area of post-foundational theory, offering more empirical depth to previously conceptualised explorations of the power of public leaders in times of perceived crisis. While we do not claim generalisability from our findings, we do note that as leadership has been acknowledged as an increasingly influential signifier in the areas of organisation studies, political studies and public administration, our article may act as a useful resource for a broad range of studies that seek to explore the ways in which leadership is deployed rhetorically.

Before continuing, as this is a study relating to language use, it is important to clarify the post-foundational terminology employed. By discourse, we adopt the definition from Laclau and Mouffe (1985: 105) that a discourse is a ‘structured totality resulting from [an] articulatory practice’ and implies a degree of ‘fixation’ in meaning (Jørgensen and Phillips, 2002: 26), although, as will be explored later, a key tenet of post-foundationalism is that meaning is always fluid and contingent. In this study, sovereign power is approached as a discourse. Leadership is used in this study as a signifier of a particular ‘nodal’ and ‘empty’ status, meaning that it acts as a kind of anchoring and connective word that threads together the moments that assemble a discourse of sovereign power. The word ‘moment’ is used to designate ‘differential positions, insofar as they appear articulated within a discourse’ (Laclau and Mouffe, 1985: 105). By moments in this study, we refer to notions of emergency, positivity and vulnerability that are drawn upon to assemble sovereign leadership. The term ‘rhetoric’ is used in this study to signal a system of language with a persuasive force, which ‘functions to persuade and motivate’ (Finlayson, 2003: 67).

CLS and ‘categorical’ studies of leadership

We situate this study within the area of CLS perspectives that assume that ‘leaders and leadership dynamics … exercise significant power and influence over contemporary organizational and societal processes’ (Collinson, 2011: 181) but that ‘critique the power relations and identity constructions through which leadership dynamics are often reproduced, frequently resisted and occasionally transformed’ (Collinson, 2011). Our approach is critical, in as much as we seek to examine ‘the patterns of power and domination associated with leadership, and relate it to broader ideological and institutional conditions’ (Collinson, 2011: 373) and ‘critique forms of power and dominance that relate to what leaders/managers do and how they do it’ (Fairhurst and Grant, 2010: 186).

Our study also follows the discursive (Fairhurst, 2008) and interpretive (Alvesson and Spicer, 2012) traditions within CLS, which consider ‘leadership as socially constructed through actors beginning to “see” a set of activities as leadership’ (Alvesson and Spicer, 2012: 372). Language use is therefore of central importance to the study, exploring how leaders ‘exercise power by “managing meaning”, and defining situations in ways that suit their purposes’ (Collinson, 2011: 185). This is the terrain of meaning framing (Fairhurst, 2005; Ganz, 2010), where the underlying assumption is that leaders seek to shape frames of interpretation to influence others (e.g. Auvinen et al., 2013; Spear and Roper, 2016). For example, Auvinen et al. (2013) explore the manipulative adoption of framing by leaders. Manipulation, according to the researchers, is a ‘diverse phenomenon’; it can be ‘seductive’, ‘misrepresent[ing] reality in too positive a manner’, often by ‘exaggerating one single detail’ above others (Auvinen et al., 2013: 426); it can also be deployed in ‘pseudo-empathetic’ ways, taking advantage of the ‘human capacity to participate emotionally in the experiences of other people’ or through the pretence of ‘sharing in the feelings and emotions’ of others (Auvinen et al., 2013: 427). While manipulation is not a point of focus for this study, as establishing intent on the part of the speaker in our data was beyond the scope of the research, we do see value in learning from existing work in terms of how leaders seek to convince an audience of a course of action using a range of rhetorical ploys.

A major problem with leadership research is that, as leadership is a contested, if culturally familiar notion, ‘leadership researchers often do not know what they are studying’ (Sutherland, 2016: 2). Acknowledging this problem and also wanting to be more precise than interpreting leadership as things leaders do and say, we turn to ‘categorical’ (Kelly, 2008) studies of leadership. Such studies focus on how the signifier of leadership is ‘actually used’ (Smolović Jones et al., 2016: 439), seeking to avoid the projection of prior categorical assumptions of leadership upon data, and thus, the phenomenon of ‘academics exercising power over the meaning-making of research participants’ (Smolović Jones et al., 2016). From a categorical perspective, therefore, the ‘very familiarity [of researchers] with the term “leadership” may be one of the central problems of researching it’ (Kelly, 2008: 764; see also Pondy, 1978). The focus of categorical studies is upon the specific articulations of leadership, of research participants’ ‘own methods in the practical accomplishment of leadership’ (Kelly, 2008: 769). Such an approach requires the ‘set[ting] aside [of] explicit theories, models and assumptions as to the essential character of leadership’ (Kelly, 2008: 770) so that the researcher may take account of the affective and ideological effects of the signifier.

Categorical studies enable an analysis that brings to light the myriad associations made through the use of leadership in language (Pondy, 1978). In this vein, Alvesson and Sveningsson’s (2003b) interview study of R&D managers shows how the category of leadership can disappear under contradiction or ‘break down’ (p.278) when subjected to closer scrutiny. Such vagueness associated with the signifier of leadership can contribute to the mystification and opaqueness of power, making notions of leadership less tangible and therefore less democratically accountable (Sutherland et al., 2014). In a related study, Alvesson and Sveningsson (2003a) highlight a more obvious association between the signifier of leadership and the power and status of people in senior positions, positing that the signifier is drawn upon to make otherwise mundane managerial activities seem ‘extra-ordinary’ and that it is this act of ‘labelling’ that connotes something ‘significant, even “magical” being accomplished’ (p. 1454). These studies have yet to explicitly connect, however, the grand language and signifier of leadership to the discursive assembling and normalising of sovereign power. One of our contentions, the implications of which will be returned to in the discussion, as it holds promise for future research, is that sovereign leadership may be a discourse that exists prior to and beyond its articulation by particular actors.

Leader rhetoric and terrorism

Contemporary political writers have drawn attention to what they see as an extension of the sovereign power of public leaders in response to terrorism. Agamben (1998, 2005) refers to such situations as ‘states of exception’, exceptional because they allow leaders the discretion to act outside normal structures of democratic accountability, to make legally and ethically ambiguous decisions concerning the life of certain groups of people.

This notion of sovereign power within a context of exception has informed studies of the language of leaders. These studies demonstrate how leaders, through their rhetoric, seek to engender support for exception and, when achieved, to normalise the exception (Crockett, 2003; Winkler, 2007). Esch (2010), writing in a US context, illustrates how presidents tend to draw on surreptitious myths, ‘invisible’ in his terminology (p.360), to frame a certain view of nation that makes military action seem like a ‘civilized’ (p.358) act. Framing is an important word here, and seems closely connected to sovereign power, in as much as acts of framing by leaders in response to terrorist attacks, can be interpreted as ‘claim[ing] the right to define the … moment of exception’ (Maggio, 2007: 818), definitions that often single out ‘enemies’ against whom the leader proposes a military response (Maggio, 2007) and which can come packaged in ‘savage’ terms (Ivie, 1980). Hence, leaders may commit a ‘strategic misrepresentation’ bolstered by ‘exaggeration or fabrication’ in order to adhere to the familiar tropes of the war rhetoric genre (Winkler, 2007: 312).

War rhetoric is usually saturated with the discourse of emergency (Honig, 2011) or crisis (Grint, 2005), the force of which is the closing down of alternative solutions as insufficient to a problem constructed in urgent and oversimplified terms (Winkler, 2007). Leaders can, and often do, therefore actively ‘interpret’ (Widmaier, 2007: 780) events as crises and influence public opinion with a framing of an appropriate response. ‘Times of crises’, according to Bligh et al. (2004: 212), in their study of the rhetoric of George W. Bush, ‘enhance the likelihood that followers will want to invest increased faith in leaders, see leaders as more powerful, and identify more with their leaders as a coping mechanism’.

Although valuable in making more visible the rhetorical framing employed by leaders, particularly in times of perceived crisis to justify their sovereign power, these studies tend to assume that the unit of analysis, ‘leadership’, equates to the words and actions of leaders. The effect of such an attribution of leadership to the words spoken by leaders may be to naturalise the power of leaders (Laclau, 2015: 231–241). A post-foundational analysis, we will now argue, enables a view of leadership and sovereign power as inhering in a contingent web of language, which may exist prior to and beyond any one individual leader.

A post-foundational approach to analysing sovereign power

To understand the value of a post-foundational approach, we must first, and briefly, explore the contributions of two of the most influential writers on sovereign power, Carl Schmitt and Giorgio Agamben. Infamously, Agamben (1998) sees the concentration camp as ‘the hidden paradigm of the political space of modernity’ (p. 123). To understand the basis of this assertion, we need to track back to the work of Carl Schmitt (2006 [1922]), later a Nazi jurist, who proposed that any act of founding a state seemed to require what we might consider to be a distinctly ‘undemocratic’ supplement. This supplement is that of sovereign power, the inauguration of a democratic state through the act of a leader who, at that point in time, has the required knowledge that will not be possessed by the citizenry. Thus, a contradiction exists at the heart of formally democratic states – that they rely upon the undemocratic action of a sovereign power for their founding. For Schmitt, this paradox is a kind of natural starting place for understanding leadership – all that follows from the sovereign, in the shape of constitutions, democratically elected parliaments and so on are a distraction from the potential of the exceptional rule of sovereign leaders. Schmitt and Agamben see the tendency of the states to revert to exception, in particular, during times of perceived crisis, when the norms of liberal democracy are suspended in favour of a return to exceptionalism. Agamben’s (2005) critical reading of exceptionalism is more nuanced than that of Schmitt, positing the exception as ‘a threshold of indeterminacy between democracy and absolutism’ (Loc 19). For Agamben (2005), the sovereign may act in exceptional ways that could be interpreted as outside law (e.g. in suspending freedom of speech or in acts of violence against individuals, groups or other states) and yet may also represent a pure ‘force of law’ (Loc 529) within an exception, and he has dedicated his career in search of ways in which exceptionalism may be resisted and alternative forms of life established. Nevertheless, in accepting the founding role of the sovereign, Agamben seems to also accept – or at least acknowledge the durability of – an ontology founded upon sovereign power.

From a post-foundational perspective, Ernesto Laclau (2015) claims that Agamben has been too hasty in assuming a foundational status for sovereign power. Laclau (2015) states that Agamben has confused ‘origin’ with ‘a determining priority’ (p. 231) and that Agamben ‘jumps too quickly from having established the geneology of a term’ (Laclau, 2015) to its functioning. For Laclau, the act of founding signals a range of subsequent possibilities, only one of which (exceptionalism) is pursued by Agamben. The act of such a closing down of possibility holds two effects for Laclau. The first is that it strips political agency or ‘any kind of collective identity’ (Laclau, 2015) from alternative ways of organising. The second is an accusation of depoliticisation. By this, Laclau (2015) means that Agamben does not account for political contestation, stating that in his work ‘politics disappears’ (p.235). Hence, his stinging closing retorts that ‘political nihilism is [Agamben’s] ultimate message’ (Laclau, 2015: 241).

Laclau’s critique can be understood as partly informed by Foucault’s (1980) suspicion of sovereign power and its preference for ‘prohibition’ (Foucault, 1980: 119) as a disciplinary tactic, which he held as less relevant for understanding contemporary, ‘productive’ forms of power. Hence, his quote ‘We need to cut off the King’s head’ (Foucault, 1980: 121), meaning that theory needs to move beyond the prohibitive power of sovereign leaders, to understand power as a matter of governmentality, as something that ‘circulates … through a net-like organisation’ (Foucault, 1980: 98). Butler (2006) holds that governmentality and sovereign power are not mutually exclusive, but co-dependent and contingent, as recent responses to terror have brought into focus for her how sovereign forms of power ‘assume shape within the field of governmentality, and are fundamentally transformed by appearing within that field’ (Butler, 2006: 66). What she means by this is that sovereign power, as expressed through the power to instigate exception, manifests within an apparatus of governmentality, of diffuse systems and agents that enforce power beyond the law, as is the case with the US government’s use of ‘indefinite detention’ (Butler, 2006: 50) in relation to terror suspects captured abroad. While it is beyond the scope of this study to explore the overlaps of sovereign power within diffuse systems of governmentality, it is worth highlighting the imbrication of these forms of power in the practices of states as important context and noting the underlying point that sovereign power is contingent upon other forms of power and other discursive associations.

Laclau offers a way of interpreting sovereign power as contingent, postulating that while we must ultimately know the world through language, our language is incapable of ever capturing ‘fullness’ (Laclau, 2014: 48). This is because language is contingent to its roots: it is ‘contingent all the way down’ (Cederström and Spicer, 2014: 179). The signifiers we deploy in our articulatory acts are always dependent on other referents, and onwards indefinitely. We cannot speak meaningfully of leadership in isolation, therefore, outside its dynamic relationship to countless other concepts – followership, charisma, power and so on – but this system of language never fully closes; it is contested and keeps adapting. Notions such as leadership and sovereign power are understood by post-foundationalists as constituted through contingent articulations that rely upon an alliance of interests and discourses (Laclau, 1990, 2007; Laclau and Mouffe, 1985; see also Connolly, 2005; Honig, 2011). Under this interpretation, sovereign power would also need frequent affirmation and regrounding in language, quite plausibly beyond the career or lifetime of any one leader or articulator.

To summarise the review of literature thus far, CLS offers a salient area of research for this study, in that it is interested in analysing how meaning and power are exercised through the language of leadership. Categorical analyses of leadership do not take for granted the presence of leadership but prefer to explore the ways in which the specific word ‘leadership’ is used, particularly to enhance and mystify the power of certain people and groups. To date, research into leader rhetoric helps us understand some of the rhetorical strategies used to construct a case for war but in attributing leadership to the speaking leader – who may or may not use that word – risk further naturalising sovereign power. Post-foundational analysis offers a way of interpreting leadership and sovereign power as temporary accomplishments, needing to be articulated in a contingent web of language. In what follows, we outline our methodological approach to operationalising these conceptual underpinnings for an analysis of Blair’s texts.

Methodology and case description

Tony Blair became leader of the UK Labour Party in 1994. Blair’s political legacy is contested, in domestic and international terms. In foreign policy, Blair was an advocate for eliminating poverty and debt in the Global South but was also at the forefront of advocating military action to combat terrorism in Afghanistan and Iraq – and later, out of office, supported military interventions in Libya and Syria. Blair was chosen as the focus for this study because of his centrality in the military coalition posited as a response to terrorism. The invasions of Afghanistan and Iraq are notable as symbols of the militaristic strategy in relation to terrorism, and Blair played a significant role in assembling and supporting the US-led campaigns. Significantly, however, Blair represented a broad coalition of voters, many of whom felt a significant degree of disappointment at the militaristic turn in his premiership, in a way that was not matched in the US context, where the president, his advisors and cabinet arrived with more ‘hawkish’ expectations and credentials. Militarism was therefore a more controversial and contested strategy for Blair, whose rhetoric would have had to do extra work to appeal to increasingly hostile voters. Relatedly, Blair, who continues to play a prominent role in UK public life, represents an important moment in terms of how people relate to the words and actions of public leaders, at least within a UK context: he is a socially significant symbol of leadership.

We chose Blair with the informal foreknowledge that ‘leadership’ was a signifier he often used in public appearances. An initial convenience sample of available online speeches revealed the prominence of leadership in his talk. Furthermore, an initial literature search indicated that leadership was not a topic that had been covered in any depth in relation to Blair. The rhetoric of the broader New Labour movement and governments of Blair has been studied, notably by Fairclough (2000) and Finlayson (2003), whose work will be drawn upon in the analysis. Neither, however, offers an in-depth analysis of the use of ‘leadership’.

In terms of leadership and Blair, Smolović Jones and Jackson (2015) explore the visual dimensions of Blair as a leader, from a Lacanian perspective, but do not interrogate Blair’s use of the signifier ‘leadership’. Finlayson and Martin (2008: 457), in their analysis of Blair’s final party conference speech as leader, provide some empirical substance as to how he constitutes the task of being a leader, with ‘personal character [as] the substance of his argument’, specifically, ‘qualities of leadership’ such as ‘self-belief and “courage” rather than “caution”; a capacity to make unpopular decisions rather than no decisions at all’. The primary contribution of the authors, however, is of demonstrating the significance of an analysis of a set-piece political speech for insight into ideological, institutional and strategic manoeuvrings, rather than upon the narrower focus of better understanding how the word leadership is deployed on such occasions.

Our data set comprises a purposive sample of Blair’s speeches, press conferences and writing to address the research question, texts that allow for an in-depth analysis of the use of leadership within a broader rhetorical context. We accessed the extensive Blair archive at the UK People’s History Museum, Manchester and analysed the 910 documents available, which cover the full range of Blair’s leadership of the Labour Party (1994–2007) and supplemented the archive with a reading of Blair’s (2010) memoirs. In terms of the museum documents, 205 of them were from the pre-11 September 2001 period and the remaining 705 from the post-11 September 2001 period. They cover large, set-piece conference speeches (party and trade union), campaign speeches and articles, speeches to interest groups and public bodies and articles penned for newspapers. These documents varied in length from longer speeches of around 20–30 pages to tighter ‘doorstep’ statements of a couple of pages. We gave primacy to such set-piece events due to their well-established value in illuminating the ideological work and strategy at play in the rhetoric of the speaker (e.g. Finlayson, 2003; Finlayson and Martin, 2008). Although we could have added further to the sample with parliamentary speeches, debates and broadcast interview transcripts, we noted a significant degree of saturation having analysed the 910 documents, and our sample compared well with other studies pursuing similar approaches: for example, Bligh et al. (2004), 516 speeches and media sources; Esch (2010), 50 texts; Finlayson and Martin (2008), 1 speech and Maggio (2007), 6 speeches.

Two concepts from post-foundationalism were put to work in analysing the texts: the empty signifier and the chain of equivalence. Post-foundationalism introduces the notion of the empty signifier as a means of better understanding how some words seem to gain prominence and significance in shaping discourse around particular dominant understandings, such as tendencies towards sovereign power. Empty signifiers hold a paradoxical status in that they buck conventional understanding of the word ‘empty’ because they are affectively salient for groups, yet in and of themselves seem to mean little, instead relying on an abundance of investment and support from other signifiers, moments and discourses of value to a group: they are simultaneously the subject of ‘overdetermination’ and ‘underdetermination’ (Laclau, 2015: 66), meaning-full but also meaning-light. Empty signifiers hold great affective and organisational value for groups as they are the core around which people make sense of things that matter to them; they signify and constitute ‘an image of a pre-given totality’ (Laclau, 2007: 162), a totality that is due to the contingent nature of language, always beyond reach. The association of empty signifiers with an (unreachable) ‘totality’, ‘an impossibility of signification as such’ (Islam et al., 2017: 4), indicates the significance which groups attribute to them. Empty signifiers thus reach beyond utility to signal salient, but ultimately unobtainable affective and political commitments (Jones and Spicer, 2005), a ‘compelling utopia’ (Kenny and Scriver, 2012: 616). The concept of the empty signifier has been drawn upon in organisation studies to highlight the cultural salience and ideologically loaded constitution, but also vagueness, of the concepts of entrepreneurship (Jones and Spicer, 2005; Kenny and Scriver, 2012), mindfulness at work (Islam et al., 2017) and, indeed, leadership (Kelly, 2014; Smolović Jones et al., 2016). In terms of leadership as an empty signifier in relation to terrorism, such a focus may indicate and organise a certain set of available ethical and political responses. However, because leadership is also in and of itself a contingent signifier, it will make apparent the limits of this response, for example, when terrorist attacks keep occurring despite articulations of leadership.

The second concept of analytical value to this study is that of the chain of equivalence, which helps us to understand the associations assembled to pull off an articulation of sovereign leadership. The chain of equivalence has been adopted in research to better understand how coalitions are articulated, assembled and adapted, such as campaigns opposing airport expansion (Griggs and Howarth, 2004, 2008) or within labour disputes (McLaughlin and Bridgman, 2017), offering a deep view of the positive and antagonistic bonds that bind these groups discursively. A chain of equivalence simply means the range of discourses and people anchored by an empty signifier. They are actively assembled – they are articulated – and do not exist independently of our actions (Moon, 2012). In terms of their power effects, chains enable a view of power as one articulated between a coalition of interests and actors, rather than being something possessed by leaders. Post-foundationalists are thus interested in ‘sedimentations of power’ (Marchart, 2007: 139) within (temporary, if often stubbornly durable) hegemonic constellations. One important power effect of chains of equivalence lies in their very contingency. Because such chains are contingent accomplishments, they are always reliant upon an ‘antagonistic’ outside element for their coherence (Laclau and Mouffe, 1985: 144). Equivalential chains must always ‘expel or externalise those others to which they are opposed’ (Laclau and Mouffe, 1985: 165) because their unity depends as much upon the construction of a negative ‘other’ as it does upon any positive, internal content. We need the bogeyman of the terrorist for our response (assembled through a chain of equivalence) to make sense. Nevertheless, due to their foundational contingency, chains can be contested and their vulnerabilities exposed – people may develop a different chain of association in relation to leadership, sovereign power or terrorism.

The data were read in three stages. A first reading was concerned with identifying the empty signifiers put to work in Blair’s texts. We discovered a marked turn in Blair’s rhetoric after 9/11: prior to the attack, his rhetoric was dominated by language without the same connotations of sovereign power. The empty signifier most noted in this period is ‘partnership’, whereas after 9/11, ‘leadership’ was pre-eminent. A total of 423 references to leadership were identified in the latter, post-11 September period of Blair’s leadership, with only 43 found in the earlier period. We bracketed off each section of the documents where leadership was used, to read this signifier in the context of the broader rhetorical point made within the text.

A second reading was concerned with exploring the status of leadership as an empty signifier (Jørgensen and Phillips, 2002: 29). We read the documents iteratively, asking whether and how leadership was deployed as an empty signifier and whether it was being used as a node, bringing together other discourses in a coherent form of meaningful argumentation. Through such a strategy, we identified three moments at work: emergency, positivity and vulnerability. These moments were interleaving in the texts and separating them out for neat presentation in a paper is perhaps a necessarily synthetic dissociation, something which will be elaborated upon in the discussion. Also of particular note was the way in which morality (through informal theorisations of moral leadership, establishing what it means to be a moral leader and equating morality with being a normal person) seemed to weave in association with all of the moments.

We performed a third reading, a close textual analysis drawing on post-foundational discourse analysis (Cederström and Spicer, 2014; Griggs and Howarth, 2004, 2008; Howarth, 2013; Jørgensen and Phillips, 2002; Moon, 2012). Our analysis focused on the empty signifier within its textual context (Jørgensen and Phillips, 2002) and proceeded in three further stages. First, we explored what was placed in a chain of equivalence with the empty signifier, ‘the linking together of signifiers in chains of equivalence’ (Jørgensen and Phillips, 2002: 42). Identifying the moments and signifiers assembled around the empty signifier allows a reading of the affective force lent to an articulation (Cederström and Spicer, 2014: 191). Second, we explored who was included in the chain of equivalence, the people pointed to by Blair as sharing common cause (Laclau, 2007). Furthermore, focusing on who was included in a chain enabled an analysis of those who were excluded by the articulation, actors constructed as antagonistic, on the other side of the ‘political frontier’ proffered by the articulator (Griggs and Howarth, 2008: 129). Third, we focused on points of fragility, and hence also contestation, in the texts (Cederström and Spicer, 2014; Howarth, 2013; Kelly, 2014), as a means of identifying limits to the articulatory logic presented. Such points can be indicated by pauses or half-completed sentences (Smolović Jones et al., 2016). A focus on fragility also entailed exploring the non-meaning of articulated signifiers, instances where the meaning might be understood tacitly, by inference.

Data analysis

In what follows, we introduce our three moments in the articulation of leadership by Blair, which, when placed together, weave a tapestry that seeks to make sovereign power understandable, sensible and desirable and draw in more depth on the twin concepts of the empty signifier and chain of equivalence. The first moment, ‘emergency’, involves a generation of urgency that simplifies the problem of terrorism and offers straightforward moral leadership as the solution. The second, ‘positivity’, articulates leadership as a purely favourable category which contrasts positive moral leadership with an ‘outside’ that is posited as threatening and immoral. The third, ‘vulnerability’, invites empathy with the ‘normal’ and moral person of the leader, conveying the sense of a subject who struggles through adversity, emerging with a stronger faith in the necessity of sovereign leadership. We refer to the rhetorical tactic of transiting between vulnerability and foreclosed sovereign leadership as one of ‘expose and close’.

Discussion and conclusion

We have sought to uncover a system of foreclosing moments within Blair’s discursive chains, cohering upon the empty signifier of leadership and constructed through chains of equivalence, which knit together contingent but affectively and morally loaded moments that push towards sovereign power as the solution to terrorism. We called such an imbrication of the empty signifier leadership and the discourse of sovereign power, ‘sovereign leadership’. We will now consider our contribution in relation to the three moments highlighted in the data, before moving on to reflect on some of the broader implications of our work.

First, sovereign leadership was achieved through aligning the empty signifier of leadership with moments of emergency. Grint’s (2005) account of the relationship between crisis and command, as well as the contribution of war rhetoric studies (e.g. Honig, 2011; Maggio, 2007), draws attention to the fact that leaders often heighten a sense of emergency through language, facilitating hierarchical command solutions to problems. We show how such constructions of emergency are pieced together via chains of equivalence, but we also draw attention to their heavy moral overtones. Such a presentation underlines that emergency moments not only seek to heighten anxiety and urgency but also make a case that sovereign responses are morally just ones. The irony of this presentation lies, however, in the point that the moral presuppositions made were in themselves contingent, even empty – democracy, freedom and the rule of law are all contested and contingent signifiers. Likewise, the target upon which sovereign leadership was directed, terrorism, is also well established as a contested and contingent term (e.g. Fisk, 2001: 438–441). In contrast to Agamben’s (1998, 2005) reading of sovereign power, our case is that although the moment of emergency seeks to foreclose meaning, as these are contingent and co-dependent articulations of sovereign leadership, they are also more susceptible to being scrutinised, unpicked and challenged.

The second moment showed how meaning was foreclosed around sovereign leadership by equating it to a positive and moral attribute – those who oppose are simply labelled as something else: misplaced, eccentric, intolerant or terroristic, and expelled outside the chain of equivalence as antagonists. Within CLS, a tendency towards excess positivity has been highlighted by Collinson (2012) in relation to the decision to invade Iraq and the subsequent resignation of the cabinet minister, Clare Short, who complained of the silencing of dissenting voices within government. The broader implication of the work of Collinson (2012), and of Fineman (2006), is that excess positivity creates an environment, indeed, a seductive narrative (Auvinen et al., 2013) that restricts learning from negative experiences and failure. A key finding of our article is that we may take insight into excess positivity in relation to leadership a stage further and connect it explicitly to the generation of shadowy ‘others’ in leadership talk, who are articulated in this study via chains of equivalence. These antagonistic others – terrorists, the threat of terrorism, the media, opponents of war and protestors – are primarily opaque and dynamic presences who serve the function of justifying the need for sovereign leadership. However, without the presence of excess positivity, the need for these figures would appear compromised at best, as the signifying system around leadership would contain within it more capacity for critical self-reflection and learning. Positivity and antagonistic others are co-constitutive of the kind of leadership constructed, as if the uncertainties, unspoken dark spots and doubts concerning militarism and sovereign power are packaged up and placed onto the figure of the unreasonable or dangerous ‘other’. While the construction of ‘others’ in a co-dependent relation to positivity may serve as a function of foreclosure, it also, ipso facto, reveals the contingent nature of the construction of leadership, which is dependent upon an external, and usually imprecise, supplement for its coherence and persuasive force. Future research might explore whether such constitutive ‘othering’ manifests in other contexts, outside the formal political arena. To date, CLS has largely overlooked the role of the ‘other’ in constituting leadership discursively, perhaps because to do so entails exploring the ultra-dark recesses of organisations and language, of marginalisation and violence, or, more provocatively, entails disrupting and challenging the value of the broader leadership industry.

The third moment of ‘vulnerability’ showed how Blair sought to position himself as a ‘normal’ (Fairclough, 2000) person, within a system of contingent identifications, where all ‘decent’ people are assumed to be flawed and incomplete. This was a foreclosing moment, as vulnerability was drawn upon as a kind of rite of passage that demonstrated strength in and through vulnerability, as the sovereign emerged on the other side emboldened by his moral commitments and judgement. Studies of the war rhetoric of leaders have focused on the aspects of leader presentation that emphasise strength in the face of crisis: the ‘commanding’ (Grint, 2005), ‘savage’ (Ivie, 1980), ‘civilising’ (Esch, 2010) and ‘charismatic’ (Bligh et al., 2004) aspects of leaders’ talk. However, our study raises the possibility that weakness – or at least the willingness to communicate a less than perfect façade – seems significant. We can connect vulnerability to the findings of Grint (2010) and Śliwa et al. (2012), who both highlight the salience of self-sacrifice in the discourse relating to heroic leaders and also to the notion of ‘pseudo-empathy’ (Auvinen et al., 2013) as a manipulative act. Our study contributes by drawing attention to the tactic of expose and close. The tactic heightens a leader’s vulnerable position within a contingent system, only to posit such vulnerability, the leader’s willingness for self-sacrifice and the ability to survive attack, as a further qualification for sovereign leadership. We also contribute here by highlighting how notions of self-sacrifice, via vulnerability, are assembled within chains of equivalence that position the leader in relation to other ‘normal’ people. However, we also show how such moments of vulnerability and willingness to self-sacrifice again contain a dark side, a positing of an antagonistic ‘other’, without which the leader could not appear as normal or moral in the first place.

As previously noted, separating out these moments in some ways is a synthetic exercise, as in practice, they seem to overlap, coming together to foreclose meaning around sovereign leadership. It is difficult to envisage any contemporary pull towards sovereign leadership being as compelling without the presence of all three: convincing people of the need for a strong and decisive moral leader without an obvious antagonist and the threat of emergency seems implausible. In particular, in an era of post-deference towards leaders, at least in the context of liberal democracies, displays of self-depreciating vulnerability – acknowledging contingency but also, through that very contingency, transcending it – might be particularly important in building a case for sovereign leadership. We might also note the fact that each of these moments is in itself somewhat ‘empty’. They are salient but flexible moments that might carry a quite different connotation under alternative circumstances. The relative emptiness of the moments presented, as well as the notion of sovereign leadership itself, points to their contingency and thus also, in theory, to the possibility that each of these contingent connections could have been otherwise articulated and received. Such a reading stands in contrast to the more ‘determining priority’ (Laclau, 2015: 231) given to sovereign power by Agamben’s (1998, 2005) genealogy and to the tacit assumption in studies of leader rhetoric that the speech of leaders equates to leadership.

In terms of our primary contribution, in the area of CLS more generally, we hold that the concept of sovereign leadership could be an important, yet overlooked, explanation of the preference for ‘strong’ individuals in the broader leadership literature and, indeed, in organisational and political life. Our study shares in common with the work of Śliwa et al. (2012) and Grint (2010), a concern for highlighting the cultural familiarity and appeal, the ‘sacredness’, of investing in the sovereign leader as the source of a group or society’s problems and solutions, but differs in that it moves beyond the figure of the leader to make more visible the system of language that constitutes such sacralisation. An underlying assumption of the value of our research is that systems of language pre-date and live on after a leader has left office, a point we regard as significant given that terrorist attacks continue but the political leaders charged with addressing them change. While the analysis can never be entirely divorced from the articulating individual, we see great value for CLS in exploring more specifically and systemically uses of the signifier of leadership, how it is used to foreclose but also, perhaps, to open up critique and dialogue.

The connection we have made between sovereign power and leadership, and the conceptual and methodological approach we have adopted for their analysis, we believe holds potential for understanding how leadership is articulated and circulates in organisations and political life beyond the specific utterances of an individual leader. After all, the call for ‘strong leadership’ is familiar across political and organisational contexts and it usually comes packaged with foreclosing tendencies. For example, in the UK political context, Conservative PM Theresa May fought the 2017 general election campaign on the premise of ‘strong and stable leadership’. This was a mantra that was accompanied by minimising questioning from journalists and voters. The signifier of leadership seemed to be once more deployed as a mechanism of foreclosure, a culturally loaded signal sent out to voters that the leader ought to be able to exercise sovereign power in a time of uncertainty and crisis, this time the United Kingdom’s withdrawal from the European Union (EU). Such a commonplace association of leadership with sovereign power is not simply something pulled off by leaders themselves but by a broader web of actors – among followers and the media. In a recent edition of the BBC’s political panel discussion programme, Question Time, for example, the solution of three separate (white male) audience members to the impasse and complexity of the UK’s negotiations over withdrawal from the EU was ‘a strong leader’ or ‘strong leadership’, with the details of what such leaders or leadership might accomplish left unspecified (BBC, 2018). Although a more dispersed analysis of sovereign leadership is beyond the scope of this article, future research might explore how notions of leadership and sovereign power are entangled and manifest among groups in their everyday practices. Relatedly, we also note that the focus of our research on language overlooked a more embodied account of the relationship between sovereign power and leadership. In terms of both of these alternative foci, considerable scope exists for exploring the material, gendered and racialised connotations of sovereign leadership, deferring as such constructions usually do to a particular type of privileged white masculinity – although, as noted with reference to Theresa May, being a man need not be a prerequisite for adopting and communicating sovereign leadership.

Remaining within the bounds of our broader CLS contribution, making the system of language surrounding sovereign leadership more visible and moving the focus away from the figure of the leader holds potential for better understanding the dynamics of resistance leadership (Zoller and Fairhurst, 2007). The key facet of a post-foundational understanding of sovereign power is its focus on the contingency, and therefore, inherent contestability, of that power, allowing for the possibility that with a more nuanced view of the system of language surrounding sovereign leadership, researchers seeking to theorise resistance leadership – and indeed resisting groups in practice – could be better positioned to offer alternative articulations. After all, if we can identify and map the language of sovereign leadership, its complicated weaving of moments, performed through the deployment of empty signifiers and chains of equivalence, we can also begin to patiently unpick and challenge these articulations. One way in which our research might contribute to a better understanding of resistance leadership is in its differentiation between leader and leadership as a signifier, which in turn allows a differentiation to be drawn between resisting a sovereign leader and resisting sovereign leadership, as practice and discourse. While some resisting groups may successfully resist individual sovereign leaders, they may yet perpetuate sovereign leadership through adopting some or all of its discursive and practice characteristics (Collinson et al., 2017).

In terms of our secondary contribution, in the broader area of post-foundational research, we note that post-foundational approaches to sovereign power have remained conceptual in nature and separated from the rich body of work within CLS. While post-foundational thought has provided a theoretical lens through which to interpret the ontological contingency of the very notion of sovereign power, and concepts that can be put to work in analysing articulations of sovereign power, it has yet to explore and refine these propositions in a contextually rich manner, drawing on insights from contemporary critical leadership theory. In particular, we suggest that there is a gap present between the theorising of sovereign power as an ‘uncertain circulation’ (Connolly, 2005: 141) and a more stubborn empirical experience, where organisations and groups seem to revert to sovereignty. Approaching sovereign power via various foreclosing moments may help enrich further studies and contribute to better understanding how, despite its contingent foundations, it manages to persevere as a dominant discourse in political and organisational life, as well as pointing to ways in which alternative signifiers and contingencies can be established.