Securitisation and the function of functional actors

ABSTRACT

This article examines the neglected category of functional actor in securitisation processes. I argue that functional actors are a useful analytical category only if such actors are functionally distinct from other actors. A close analysis of Security: A New Framework for Analysis reveals that this is not the case; the majority of the functions such actors have are covered by other actors. The exception is that they may contest securitisation; yet in securitisation studies this function has become associated not with functional actors but with audiences. I show that when the audience is conceived in line with its meaning in common usage (i.e. as the addressee of speech (acts)) only specific actors (most notably, referent objects who are promised protection via securitising moves) can object to securitisation, and only on securitisations (ostensibly) intended to save them. Given that actors other than referent objects/threateners regularly object to securitisation, I go on to locate the ability to veto/endorse securitisation on behalf of others with functional actors. The remainder of the article distils functional actors into different categories/roles. I show that scholars too are functional actors; ergo they do not need likeminded audiences to stage critical interventions.

KEYWORDS:

• Functional actors
• securitisation
• veto
• Copenhagen School
• audiences
• media

Introduction

In critical security studies the secondary literature on the Copenhagen School’s (hereafter CS) securitisation theory is booming. While not everyone accepts that the CS qualifies as a part of critical security studies (Hynek and Chandler 2013; Aradau 2018), following Mutimer (2019) I take critical here in the broadest sense, to wit as critical of traditional security studies’ positivism.

Moreover, this article shows that because securitisation analysts can double-up as functional actors, an emancipatory agenda is not antithetical to securitisation studies.

Securitisation theory holds that security threats do not simply exist ‘out there’, but rather security is a highly political process with issues turned into security threats via a sequence of events usually involving a securitising actor, a securitising speech act/securitising move (whereby a securitising actor declares a particular referent object threatened in its existence unless urgent action is taken right away), the audience (which has to ‘accept’ the threat narrative contained in the securitising move), and the enacting of extraordinary measures (the breaking of established rules) in order to deal with a (perceived) threat. Practically every aspect of the theory has been revisited and revised; it has been applied to a vast and diverse number of empirical contexts in the West and beyond, gaining traction in fields other than International Relations/Security Studies (e.g. Gearon and Parsons 2018; Jackson 2015). In spite of this flurry of activity however, nothing explicit has been written about functional actors,¹ which is to say ‘actors who affect the dynamics of a sector. Without being the referent object or the actor calling for security on behalf of the referent object, this is an actor who significantly influences decisions in the field of security’ (Buzan, Wæver, and de Wilde 1998, 36).

By posing the research question: ‘Who or what are functional actors in securitisation?’ this paper seeks to close this gap in the literature. I do this not for the sake of identifying yet another new area in securitisation studies, but to understand what role – if any – such actors play, why they have been overlooked and whether they should be ignored. In short, I do this to better our understanding of securitisation processes.

I commence my analysis from a standpoint of agnosticism regarding the specific relevance of functional actors. In its original formulation by the CS and Wæver, the relevance of the functional actor is ambiguous. On the one hand, they suggest that functional actors are significant (Ibid, 36), but on the other Wæver claims that they ‘are not involved in securitisation’ (Wæver 2003, 12). It is also noteworthy that in the 20-odd years since the publication of Security: A New Framework for Analysis (hereafter SANFFA), they have not developed this concept any further. From an email exchange with Barry Buzan in the autumn of 2018 I learned that the term functional actor was – during the co-writing of SANFFA – ‘pulled out of the air’; if it had some specific and deeper meaning in linguistics, he was unaware of it. I have been unable to find evidence of this, and I hasten to suggest that if there was a deeper meaning it would have been picked up by the many second-generation scholars skilled in the philosophy of language. Though pulled out of the air, the term functional actor implies something that we should not ignore. Curiously we can find this something in Buzan’s work on the English School (Buzan 2004), wherein he examines the presence of institutions in functional terms, i.e. the idea that we can ascertain institutions in terms of the function they fulfil in international society. His proposal focuses on whether or not prospective institutions serve – amongst other things – one or more of the three primary goals of international order, identified by Bull as: (1) security against violence, (2) observance of agreements, and (3) property rights (Buzan 2014, 174). This is not the place to elaborate on debates concerning institutions in the English school. Nevertheless, for our purposes here Buzan’s take on functionalism suggests that we can elucidate the nature of functional actors in securitisation if we try and uncover what distinct function, if any, these actors fulfil in this process, or else in the distinct sectors where securitisation dynamics play out. It should be noted here that a functionalist approach does not merely allow us to understand functional actors; within the CS’ securitisation theory all categories of actors are differentiated along the functions they fulfil in the securitisation process. This means that the term ‘functional actor’ is probably not the most precise to capture this category of actor. Nevertheless, I stop short of suggesting a more suitable name. After all, the point of this article is to explain this very category.

In order to uncover the function of functional actors I proceed as follows. The first section of this article examines what we definitively know about functional actors from SANFFA. A close textual analysis reveals that the different sector specific functional actors identified by the CS do not fulfil any distinct function; instead their role is already covered by one of the other analytical categories of actor identified by the CS. This being said, however, at least one passage of SANFFA does suggest a functionally distinct role for functional actors, namely when they assume the role of ‘veto actors’ during securitisation (Buzan, Wæver, and de Wilde 1998, 89). Importantly, veto actors here does not refer to actors that possess formal (i.e. legal or constitutional) powers of veto as, for example, those held by the United Kingdom and the other four permanent members of the United Nations Security Council. Instead, the term veto actor is derived from Porter and Brown’s (1991) work on environmental politics, who theorise inter alia actors that (1) lead on environmental protection initiatives, (2) actors that support lead actors and (3) actors that veto environmental efforts (Buzan, Wæver, and de Wilde 1998: 77ff). For the CS, but also in this paper, veto actors, or else veto powers, simply are actors that assume the capacity to oppose securitisation, including by ‘contesting or ignoring’ it (Buzan, Wæver, and de Wilde 1998, 77). Be that as it may, in securitisation studies this role is not generally associated with functional actors but instead with audiences who, as actor, legitimate securitisation through implicit, and latterly explicit, consent (Olesker 2018; Salter 2011; Roe 2008).

The second section of this article examines whether audiences have rightly been ascribed that role or whether the veto securitisation scholars attribute to audiences rests with functional actors. In short, I examine whether some audiences are in fact functional actors. I commence by looking at what audiences are in everyday language, and from this basis I argue that the only actual audiences in securitisation are the referent object and the threatener,² for these are the addressees of securitising speech acts. Accordingly, audiences are not only delimited to specific, rigidly defined groups, but the influence of their veto is limited as securitising actors frequently override their objections in the name of security. From here I return to the role of the functional actor. I assume that the idea (popular in securitisation studies) that securitisation will be contested by a broader range of actors than simply ‘audiences-as-addressees’ is valid. After all, many different people regularly object to securitisations in which they are not referent object or threatener. I suggest that these actors are functional actors. The substantive difference between the functions of audiences and functional actors thus is that audiences object or endorse securitisation on behalf of themselves, while functional actors object or endorse securitisation on behalf of others.³

In the third section I go on to sketch out what kind of functional actors might exist in each sector, by looking at what kind of actors are likely to veto or endorse securitisation in which others are referent object or threatener. On the basis of the findings, section four aims to distil the many functional actors into a number of specific roles, namely: gatekeeper, (de-)legitimater, epistemic community, rebel and champion. By way of a conclusion I point out that all those securitisation scholars who place stock in the audience for normative reasons, have good reason to embrace the notion of functional actors. Thus, it allows scholars to enter securitisation processes – where they are not referent objects or threatener – in a truly critical capacity.

What do we concretely know about functional actors?

The description of functional actors in SANFFA takes the form of a juxtaposition, whereby little is said about functional actors other than what they are not. Specifically, they are neither securitising actors nor referent objects of securitisation. Securitising actors are described as ‘actors who securitize issues by declaring something – a referent object – existentially threatened’ (Ibid, 36); referent objects, in turn, are ‘things that are seen to be existentially threatened and that have a legitimate claim to survival’ (Ibid, 36). Given that we are told that ‘functional actors [are dealt with] in the sector chapters’ (Ibid, 36) I will begin my analysis by looking at what precisely we learn in each chapter about these actors. It should be noted that, in spite of the statement by Buzan, Wæver and de Wilde that functional actors will be discussed in the separate chapters, they do not – by that label – feature in the chapters on the societal and political sectors.⁴ By contrast, in the chapter on the military sector the discussion of functional actors is given a separate subheading because, we are told, this sector ‘is rich in actors that influence the dynamics of the sector without being either referent objects or securitising actors. Many of these actors are either the agencies of force, ranging from assassins and mercenary companies through defence bureaucracies to armies, or providers of the instruments of force, most notably the arms industry’ (Ibid: 56 my emphasis). While there can be no doubt that these actors can be found in the military sector, what is their primary function? Clearly the primary function of agencies of force is to provide security.⁵ In the military sector agencies of force execute securitisation (they may do this also on threats in other sectors), which is to say they carry out the extraordinary measures part of successful securitisation (Ibid, 25–26).⁶ The arms industry, in turn, does not execute securitisation; instead they identify security threats in need of securitisation. By speaking security, however, it is not their intention to initiate their own securitisation, instead they seek to convince other more powerful actors (most notably in this context governments) to securitise. They request securitisation – at times – very likely simply in order to sell their products, but at other times because they perceive a real threat. The ability to request securitisation is, if not by that label, mentioned in the CS’s original framework, because they recognise that while anyone can declare an entity existentially threatened, securitisation is successful only when it involves: ‘existential threats, emergency action, and effects on interunit relations by breaking free of rules’ (Buzan, Wæver, and de Wilde 1998, 26). Ultimately, however, the idea of securitisation for agenda-setting purposes was more fully developed by Vuori (2008, 76) who argues:

[…] not all securitizing speech is uttered by the powers that be, who also do not always have to be state powers. People outside official authority can utilize securitization speech to achieve certain aims, provided they have sufficient social capital (cf. Bourdieu 1991 [1982]). One such aim can be the raising of an issue on the agenda of decision makers. [….] The perlocutionary effect intended is convincing decision-makers of the urgency of a threat, so they will raise the issue on their agenda and put the suggested measures into effect. The illocutionary point of this type of securitization is directive; the point is to try to get other people to do things, to get the hearer to carry out the course of action represented by the propositional content, e.g. to do X in order to repel threat Y.

In the chapter on the environmental sector of security we learn that functional actors are ‘economic actors [industries etc.], whose activity is directly linked to the quality of the environment. These are functional actors whose behaviour affects ecosystems […]’ (Buzan, Wæver, and de Wilde 1998, 79). To describe these functional actors as such, however, detracts from the fact that they are one of those causing the threat by failing to factor environmental concerns into their conduct. Thus, while the degraded (e.g. toxic) environment is threatening to people living within it, the polluting companies are not simply functional actors but the agents at the source of the threat (for short: threateners). Another set of actors identified by the CS in this sector are those on the right side of the environmental security equation, for example, non-governmental organisations who work to protect the natural environment for the benefit of all. As with the arms industry examined above, their function is to request securitisation from other more powerful (including adequately strategically situated actors) who can affect emergency measures (notably states); for example, by becoming signatories to an extraordinary (in terms of enforcement mechanisms and demandingness) global climate regime.

In the economic sector the CS identifies notably fewer functional actors. Firms, states and IGOs are all actors that affect ‘the security dynamics within the sector, most notably as the demon or saviour of less-developed national economies’ (Ibid, 103). Another way of putting this, however, is to say that firms etc. can act as both the threatener and/or the executor of securitisation/securitising actor in this sector.

From what I have said so far, we can draw two preliminary conclusions. First, that functional actors are sector specific. Second, none of the actors that the CS identifies as functional actors fulfil a role that is not also already covered by other actors in the securitisation process. The entities picked out by the CS are either: a) securitisation requesters; b) securitising actors and/or executors of securitisation, or c) threateners. In short, it seems that, because functional actors are accounted for by other analytical categories, there is nothing else to say. Indeed, this may well explain why functional actors have practically played no role in securitisation studies.

The above would be a premature conclusion however, for in the little the CS does say about functional actors one comment in particular stands out. In the discussion of the environmental sector they argue that a holistic assessment of environmental security dynamics includes an assessment of: ‘Who are the veto and other functional actors in this issue area?’ (Ibid, 89) In short, here, functional actors are also those actors that contest securitisation. This function comes through in the economic sector as well, where states can act as gatekeepers ‘regarding how permeable it allows its borders to be to economic transactions […]’ (Ibid, 103). While this function sets functional actors apart from securitising actors and securitisation requesters,⁷ there is one other entity allegedly in possession of veto power: the audience. In SANFFA the audience is not listed separately alongside referent objects, securitising actors and functional actors, nor does it feature in the sector chapters. Instead the audience features only once,⁸ then however decisively:

A discourse that takes the form of presenting something as an existential threat to a referent object does not by itself create securitization – this is a securitizing move, but the issue is securitized only if and when the audience accepts it as such. (Accept does not necessarily mean, in civilized, dominance-free discussion; it only means that an order always rests on coercion as well as consent. Since securitization can never only be imposed, there is some need to argue one’s case) (Ibid, 25, first emphasis in original, second added).

Over time – in securitisation studies – the nature of the audience has become clearer and its role has increased. In 2003, Ole Wæver described the audience as ‘those who have to be convinced in order for the securitising move to be successful. Although one often tends to think in terms of “the population” or citizenry being the audience (the ideal situation regarding “national security” in a democratic society), it actually varies according to the political system and the nature of the issue’ (Wæver 2003, 11–2). Moreover, critiques emerged that highlighted the impossibility of combining illocutionary speech with intersubjective agreement (e.g. Balzacq 2005) which tilted the scales further in favour of the audience. Today, many securitisation scholars consider audiences no longer as passive actors but rather active entities that decide securitisation (Williams 2003). We might say that for many securitisation scholars audiences have become quasi-formal veto powers. Moreover, audiences are also diverse in their nature, including electorates, parliament, technocratic or scientific bodies; plus, analysts often identify multiple audiences in one and the same securitisation (Salter 2011; Kaunert and Léonard 2010). Adam Côté, after a systematic review of the literature, summarises the role of the audience thus: ‘the securitization audience is best defined as the individual(s) or group(s) that has the capability to authorize the view of the issue presented by the securitizing actor and legitimize the treatment of the issue through security practice’ (2016: 8 emphasis in original). For our purposes here, all of this means that one of the two statements below is true.

1. Functional actors do not exist because their function is already covered by one of the other categories of actor, including audiences.

2. Some of the vetoing entities securitisation scholars refer to as audiences really are functional actors.

In the remainder of this article I will argue in support of the second statement.

Do audiences have veto power?

Regardless of what anyone in securitisation studies says audiences are and what their function is, let us consider here for a moment what the standard meaning of audience is. My trusted Oxford Compact Dictionary tells me that audiences are ‘the assembled spectators or listeners at an event’ (Soanes 2000, 60). At the theatre, during an opera, or at a public lecture, for example, the audience are all those people who have come to see the play, hear the music, or listen to the lecture. The function of the play/music/lecture is to entertain and perhaps to educate, the function of the audience is primarily to consume. While audiences are thus mainly external to the performance, they do influence events. Audiences’ screaming particularly loudly or for a prolonged period of time might convince a performer to reappear and play another song, while a mass walkout can terminate a public lecture. More subtly, performances are planned with a view to audiences’ reactions and often altered with a view to feedback from past performances. Indeed, without an audience there would be little point in the performance. What insights can we draw from this conceptualisation for securitisation? I think it makes sense to include the category of audience into securitisation theory. Especially if securitisation involves securitising speech, then this speech must be aimed at someone, otherwise what would be the point of it? This also means that if we can ascertain the addressee of the securitising speech act then we can know with certainty who the audience actually is. To do this, however, we need to figure out what securitising actors do when they utter securitising moves. We know that when they declare something/someone existentially threatened, they do this – usually ˗ by identifying a point of no return etc., yet what do they do in doing this? In short, what precisely is the performative element in doing this? Wæver answered this question some years ago when he argued:

What is the illocutionary act in relation to security? It is to define the particular case as one belonging to a specific category (‘security’) where the state tends to use all available means to combat it. It is partly a threat but also a kind of promise since more is staked on the particular issue. (1989:42, my emphases)

If then the securitising speech act is either a warning and/or a promise, it seems to me that there are two possible addressees of securitising speech: 1) agents at the identified source of the threat who are being warned off by means of the speech act; and 2) referent objects who are promised protection with the same.⁹ From this position, however, the intersubjectively shared view – among securitisation scholars – that securitising actors utter securitising moves as a form of legitimation for subsequent action makes little sense.¹⁰ Instead securitising speech acts are self-executing statements that serve to highlight the securitising actors’ position on, or recognition of, a threat; to be valid they do not require acceptance by relevant audiences. However, this is not to say that referent objects cannot object to securitisation. What it does mean is that their veto does not categorically influence whether securitisation will go ahead. Consider that, in virtue of the social contract states have a duty of care towards their citizens, which might make it necessary that objections by referent objects are overridden in the name of security. During the Ebola crisis in West Africa across 2014–2015, people living in the affected areas, who were clearly the immediate referent object of securitisation, were strongly opposed to securitisation. Initially many were in denial regarding the disease and its lethality, and many went on to fear executors of securitisation and to resent necessary practices (notably, some emergency workers were violently attacked while carrying out save burials) (Linn 2015). In spite of this rejection of securitisation, however, it continued.

This being said, at least in liberal democracies – and concerning less harmful threats than deadly epidemics – when the number of objectors is too numerous (for example, when mass protests are held) referent objects can in principle succeed in vetoing securitisation.¹¹ Yet audiences rejecting the securitising move can also have the opposite effect and (seemingly) necessitate securitisation, notably, when threateners do not heed warnings to stop threatening behaviour. In summary, while referent objects and threateners can veto securitisation, their view is not decisive to the securitisation process. This discussion raises the question: why has the audience become so important in securitisation studies? The answer is that most securitisation scholars believe in the value of democratic deliberation whereby securitisation is decided intersubjectively, and not by an all-powerful sovereign actor (Roe 2012). Moreover, most are anti-securitisation – largely because of the adverse consequences that securitisation has for beneficiaries, innocent bystanders and threateners. The assumption is that democratic deliberation on the need for securitisation in any given case reduces its likelihood; after all, those in favour can still be convinced of the wisdom of desecuritization (Hansen 2012). What these analysts do not see, however, is that this theorising transforms the audience from an apolitical analytical lens into a normative category in which the hopes and views of securitisation analysts are staked. We can see this clearly from the fact that securitisation no longer depends on simply informing the audience (Buzan, Wæver, and de Wilde 1998), but that the goal posts have shifted to actual consent, including of security measures (Roe 2008; Salter 2011). It is also evident from the fact that, in contemporary securitisation studies, audiences pertain to any one group that might object to securitisation and not to the actual addressees of speech acts by securitising actors.

So, what can we take from all this? It seems to me that while securitisation scholars are incorrect to label as audiences actors that are not the addressees of the securitising speech acts, they are correct in the observation that actors other than audiences can and will object to securitisation. That is, there are plenty of actors who will object to securitisation on behalf of others. In addition, it seems to me that those actors (i.e. actors that veto securitisation in which they are not referent object or threatener) are correctly – and in line with SANFFA – labelled as functional actors.¹² This formulation also suddenly shows why Wæver and the CS’s claim that functional actors are simultaneously important, yet not involved in securitisation is apt. In the following section I will sketch out what kind of functional actors may exist in the different sectors of security.

A sketch of functional actors in different security sectors

So far, I have argued that many of the actors securitisation scholars refer to as audiences are in fact functional actors. I also think, however, that securitisation scholars are right in their view that securitisation is often vetoed and that it is important and often justified that some securitisations are contested (cf. Floyd 2019a). We have established that while both audiences and functional actors fulfil the function of objecting to securitisation, audiences object on behalf of themselves, while functional actors object on behalf of others. Splitting functional actors away from audiences enables acuter theoretical precision. Notably, with the audience now clearly defined it is much easier to ascertain whether or not securitising moves lead to audience acceptance. Going forward, securitisation scholars could go on to examine the interplay between audiences and functional actors as well as that between securitising actors and functional actors.

In this section, I want to try and sketch out what kind of functional actors we may find in the different sectors. Before I can get to this, three further things need to be said. First, the idea of a consenting audience in the CS original formulation of securitisation means that we must assume that functional actors seek to influence the process of securitisation negatively as well as positively. In other words, functional actors may act not only to veto securitisation but also to endorse it (c.f. Balzacq, Léonard, and Ruzicka 2016).¹³

Second, the number of functional actors is likely to be larger in liberal democracies than in autocracies. This is because in free societies the right to freedom of speech and expression translates into the right to veto/endorse securitisation.¹⁴ By contrast, in autocracies no comparable right to veto securitisation exists; functional actors veto/endorse securitisation antithetical to the ideology of the regime, often at a high personal cost to themselves. Moreover, in autocracies endorsement of regime-led securitisation often rests not on consent but on coercion. Given that the remainder of this article aims to provide a broad overview of possible functional actors – unless otherwise stated – it refers to free societies.

Third, it is important to notice that the right to veto does not automatically translate into success. In other words, functional actors, like audiences-as-addressees, can try and influence securitisation; whether they succeed is another issue.

In the military sector many groups of people who are not referent objects will object to securitisation, simply because in this sector securitisation can have the worst possible consequences: war. Pacifists, who can be made up of all sections of society including members of the general public, public intellectuals, academics and celebrities, object to all wars – including humanitarian interventions – as a matter of principle, while any one of these might also object to simply a particular war effort. Politicians might also object to war efforts, especially after recent war efforts gone wrong. Notably, in Britain the intervention in Libya influenced parliamentarians’ decision not to support military intervention in Syria after the chemical attacks in the summer of 2013. Defence industries and enforcers of force in turn are likely to endorse securitisation, stressing their readiness as well as the long-term strategic importance of acting on a threat with kinetic force.

The astute reader will notice that defence industries were mentioned above when I suggested that their function in SANFFA is one of securitisation requester (i.e. an actor who requests securitisation from a more capable/appropriate actor (cf. Floyd 2018)). This raises the question, what is the relationship between securitisation requesters and functional actors; specifically, do we really need both of these analytical categories? My answer is affirmative. While both types of actor seek to influence securitisation without being involved in it (Wæver 2003, 12), they do so at different times in the process. Securitising requesters seek to influence the initiation of securitisation, while functional actors are active during securitisation. While it is possible, and likely, that requesters of securitisation (e.g. the defence industry) double-up as functional actors down the line, not all functional actors will have requested securitisation. Most obviously those functional actors that veto securitisation won’t have requested it. In short, while there is overlap between the kind of entities that will request securitisation from more powerful/appropriate entities and those that – later on in the process- endorse it (the media comes to mind here), these actors fulfil a different function in the wider securitisation process and ought to be recognised as such (see Table 1 below).

Table 1. Schema of functionally distinct actors in securitisation processes¹⁵

Actor	Function
Securitisation requester*	Seeks to convince more powerful/strategically positioned actors of the need to securitise.
Securitisation pre-emptor*	Seeks to pre-empt securitisation by discouraging would-be securitising actors from initiating securitisation.
Securitising actor	Utters a securitising move and enacts emergency measures, either by changing its own behaviour, or by instructing executors of securitisation, they are in a position of power over (i.e. government over police or military) to address the threat.
Audiences	The referent object and/or the threatener who are the addressees of the securitising speech act. Audiences veto or endorse securitisation on behalf of themselves.
Functional actors*	Actors that veto or endorse securitisation on behalf of others.
Executors of securitisation	In states and collectives of states, security professionals (police, border guards etc.) charged with the execution of securitisation. In securitisations by non-state actors likely to be same as the securitising actor.
Desecuritizing actor	An actor able to undo the institutional facts of securitisation.

In the environmental sector securitisation looks quite different depending on who is to be made secure. In the Global North environmentalists in academia, Think Tanks and NGOs with a genuine interest in the natural environment (who are also usually aware of its manifold interconnections with human well-being), have a long history in vetoing securitisation of the environment by the military. After all, securitisation of the environment by the military may simply take the form of protecting defence departments from adhering to increasingly strict environmental legislation (Barnett 2001). Yet those very same environmentalists, but also powerful individuals (the former US Vice President Al Gore or the Hollywood actor Leonardo DiCaprio come to mind) endorse securitisation of the environment by governments in the Global South provided that it means making the natural environment safe for human beings. The latter kind of securitisation has its objectors too. The ‘skeptical environmentalist’ and academic Lomborg (2013), for example, argues that the plight of the poor is such that developing states ought to prioritise economic development, even at the expense of climate security.

One question that arises in the context of vetoing is whether functional actors are not simply desecuritizing actors and thus already accounted for in the wider theory of securitisation. Vuori has suggested, for example, that acts of resistance by referent objects are desecuritizing moves (Vuori 2011a, 191). I do not deny the veracity of this claim, however, if successful securitisation necessarily involves policy change as opposed to ‘merely’ securitising language, it follows that securitisation cannot be undone with words alone. This also means that desecuritization is likely to be subject to the same power relations as securitisation and that the only meaningful actors that can undo securitisation, are actors sufficiently powerful to undo ‘the institutional facts’ (Vuori 2011b, 116). In other words, while anyone can request desecuritization not everyone is also a desecuritizing actor.¹⁶ Actors who request desecuritization during securitisation are functional actors; after all this is one way of objecting to securitisation. Yet more needs to be said here. Bourbeau and Vuori (2015) have argued that desecuritizing moves can pre-empt securitisation altogether, i.e. before it occurs. This observation is important in the context of this article as it suggests that the conceptual space occupied by securitisation requesters must be shared with securitisation pre-emptors. And just as the former can double up as functional actors during securitisation so can the latter. For greater clarity I depict distinct actors and their functions in Table 1 above.

In the economic sector securitisation might take the form of state protectionism. Such measures will be vetoed worldwide by ideological neo-liberals believing in the good of market forces. The consensus across much of the British news-media (The Guardian 2018; The Economist 2018),¹⁷ for example, seems to be that US President Donald Trump’s national security policy of imposing economic sanctions on, for example, EU and Chinese steel, is harmful. Conversely, such securitisation will be endorsed by individuals who believe that sectors of national economies need to be secured. The academic and realist security studies scholar Schweller (2018), for example, endorses this policy in an article for Foreign Affairs.

In SANFFA the CS does not – by label – identify functional actors in the societal or the political sectors of security, yet here too different groups and individuals will endorse/veto securitisation. Recall that the referent object of societal securitisation is identity (Buzan, Wæver, and de Wilde 1998, 119). Securitisation in this sector can take the form of minority groups securing their identity against (perceived) threats, while dominant identities secure themselves against dilution, most prominently by migration. Globally, the latter has endorsers and objectors. In the US 2016 presentational election, Trump campaigned (and won) at least in part on a platform of securitisation against migration from Mexico; his proposals (in-)famously include the building of a wall to stem the flow of illegal migration- a move that has received much criticism by non-referent object actors around the world. In 2016, for example, then German Foreign Minister Frank-Walter Steinmeier referred to Trump’s wall plans as ‘a bad idea’ and as a ‘politics of fear’ (in Oreskes 2016).

The political sector is about ‘threats to state sovereignty’ (Buzan, Wæver, and de Wilde 1998, 141), given that such threats can take the form of a move to larger or bigger entities (e.g. for instance into the EU), this sector is hard to disentangle from the societal sector (Ibid, 141 − 145). One threat to sovereign states in this sector is the secession of territories. Globally the securitisation of secession in other countries has endorsers and objectors; many will be driven by their own aspirations for secession, or else by fears of secession in the respective domestic context. However, objectors are also fuelled by how the issue is securitised. The violent crack-down by Spanish police at the Catalan referendum in the autumn of 2017, was widely vetoed by a large number of high-profile European politicians, including Angela Merkel, Jeremy Corbyn and Charles Michel (Rolandi 2017).

This concludes my short sketch of functional actors in the five established sectors of security as advanced by the CS. Many more actors abound in each sector, and the aim here is not to provide a full list but rather to make some general observations about functional actors across the different sectors. My observations, in no particular order, can be summarised as follows:

1. Functional actors are individuals and groups that seek to influence the trajectory of securitisation positively or negatively.

2. Functional actors veto/endorse securitisations in which they are not the referent object of securitisation or the threatener. In short they contest/sanction securitisation on behalf of others. As such these actors are functionally distinct from audiences-as-addressees of securitising speech acts.

3. Functional actors do not seek to initiate or pre-empt securitisation either by themselves or by someone else; instead they comment on existing securitisation processes. As such these actors are functionally distinct from securitising actors, securitisation requesters and securitisation pre-emptors.

4. Some functional actors are sector specific, but many functional actors are present in all sectors.

5. The number of functional actors increases relative to how free a given society is.¹⁸

Different categories of functional actors

So far, I have argued that many of the actors identified by securitisation scholars as audiences really are functional actors, that is: actors who seek to positively or negatively influence the trajectory of securitisation – in which they are not the referent object or the threatener – while it is already underway. We have seen that everyone from ordinary people, celebrities, foreign politicians, parliament, to knowledge-based experts (including academics) and the media can be a functional actor in securitisation processes. In order to offer a more refined theoretical contribution still, this final section attempts to distil the disparate functional actors into the wider purposes or tasks they fulfil. To this end I aim to focus on the social/political roles distinct groups of functional actors fulfil by drawing out their powers and how they operate (their methods). By invoking some further contemporary empirical examples than those discussed already,¹⁹ I suggest in this section that functional actors can be categorised into: gatekeeper, (de)legitimater, epistemic community, rebel and champion.

Most of the functional actors invoked in the previous section I am aware of from the news media. In other words, the media is a vehicle for other functional actors to be known and heard. Conversely, however, the media itself is easily one of the most powerful functional actors in securitisation processes. Research suggests that newspapers and television channels have the power to shape public opinion regarding securitisation (cf. Vultee 2011). Moreover, given that media outlets actively control what becomes public knowledge, by deciding what is prioritised and most importantly how information is relayed, the media itself, fulfils the role of gatekeeper in securitisation processes. This gatekeeping function can be used to endorse securitisation (for example, Fox News reported mainly positively on President Trump’s securitisation of borders in the face of the 2018 migrant caravan that saw thousands of people from Latin America arrive at the US border with Mexico), including by arguing the case for its necessity (for example, by showing, or at least often restating Trump’s claim that criminals are among the migrants), and also by defending securitisation as justified and legal (e.g. by making reference to relevant legislation or simply by referring to the rights of sovereign states).

When opposed to any given securitisation, however, gatekeepers can also seek emancipation away from existing securitisation. The left-leaning UK newspaper The Guardian, for example, seeks emancipation away from the securitisation of migration and of borders. To this end it reported on the migrant caravan by focusing on the plight of individuals. From this basis it went on to advance the view that US President Trump is wrong and unjustified in constructing these vulnerable people as security threats (e.g. Mealer 2018).

In addition to traditional forms of media, social media also works as a gatekeeper. Although, for example, Facebook does not produce news content, it polices its content (Milan 2015, 3). Moreover, it uses algorithms that select and channel information in line with values held by their ‘human designers’ (Koene 2016). During the 2016 American election, for example, evidence suggests that ‘Facebook’s News Feed algorithm suppressed users’ access to credible journalism in the run-up to Trump’s election’ (Grygiel 2019).

In liberal democracies parliament is another powerful functional actor.²⁰ It acts to (de-)legitimise securitisations that are put to vote or at least debated in that forum.²¹ In short, unlike other functional actors, this actor has – if consulted – formal power of the veto. In order to explore the role of functional actors as (de-)legitimater further it is first of all necessary to examine the meaning of legitimacy. This is necessary because legitimacy and justice are often confused, including by political scientists. Some, as Peter (2017) points out wrongly claim that ‘only a just state is morally acceptable and legitimate’. The problem is that this obfuscates that legitimacy is about ‘the justification of coercive power’ which is bound up with the right to rule (political authority) (Ibid.). Legitimate political authority is a function of how political authority is awarded. It is widely accepted that only democratic governments are legitimate. This also means that when democratically elected governments make unpopular decisions concerning the use of force (we may think of the 2003 Iraq war with the British public, where the Iraqi people were the referent object) then this does not render the decision, or indeed the government, illegitimate. Democratically elected governments become illegitimate only when they erode the basis of their legitimate political authority; for example, by banning free and fair elections, or by dismantling the separation of powers (Turkey’s current President Recep Tayyip Erdoğan comes to mind here). While, as just explained, legitimacy is about the justification of coercive authority and consequently securitisations that bypass parliament (for example, by not putting an issue to the vote) not automatically illegitimate, it is also the case that democracies that systematically stifle debate on securitisations have potentially eroded the basis of their political authority (cf. Balzacq 2015, 7).

Be that as it may, when it comes to legitimacy and securitisation perception matters more than the accurate use of the terminology. In liberal democracies many people consider securitisations that do not have parliament’s consent illegitimate, erroneously equating legitimacy with justice (i.e. rightness). Important for our purposes here is that it does not matter whether these people’s view is correct and securitisation is indeed illegitimate, because for elected leaders it matters whether they are perceived to be legitimate. In other words, because it is generally accepted that parliament (or in any case, the legislature) (de)-legitimises securitisation, parliament acts as (de-)legitimater, regardless of whether this is so. Indeed, there is more. The dominant, yet skewed, view of legitimacy as justice also means that other actors can fulfil the role of (de-)legitimater. In liberal democracies at least there is widespread consensus that securitisation is legitimate only when those that are most affected by securitisation have consented to securitisation or to being saved,²² including the means employed. Concretely this means that groups of people disproportionally affected by securitisation can operate as (de-)legitimaters of securitisation.²³ As a case in point, consider that when the Italian Five Star movement overturned the mandatory vaccination programme the preceding government had introduced for Measles, Mumps and Rubella in order to secure unvaccinated children and people with poor immune systems in 2018, it stressed the right of parents (who would have to live with the potential health implications from vaccinating their child) to choose whether or not to vaccinate (Winfield and Murru 2018). In short, functional actors disproportionally affected by securitisation are especially able to influence the general public’s view of whether securitisation is necessary and justified. And the general public’s view in turn often influences what politicians dependent on public approval will do.

Finally, from the above discussion we know that foreign politicians (including ministers, state leaders but also opposition leaders) veto/endorse securitisation processes elsewhere as functional actors. Given that they do so not as individuals but in their official capacity as legitimate political authorities, it seems to me that these actors too belong into the wider category of (de-)legitimater. Thus, they provide, or withhold, external legitimation.

Academics but also Think Tanks, NGOs and relevant knowledge-based industry/businesses variously veto or endorse securitisation. As knowledge-based experts they act as epistemic communities, who endorse or veto securitisation not in line with ideological leanings (unlike many media outlets), but rather on the basis of reasoned argument. This said, we must acknowledge here that this is not always true. Defence experts, for example, may be lobbied by the defence industry passing off vested interests as objective knowledge. Nevertheless, this type of functional actor is especially important when it comes to science-based securitisations (notably, of the environment, anti-microbial resistance or infectious disease). With regards to global climate change, for example, the Intergovernmental Panel on Climate Change (IPCC) has not only established the link between man-made carbon emissions and climate change, instead the 5^th Assessment report features a chapter on Human Security. This chapter is authored by academics and draws on existing peer-review (Gleditsch & Nordås 2014). Its concluding paragraph does endorse securitisation of climate change by the International Organisation for Migration and the U.S. National Intelligence Council, at least in so far as this pertains to furthering human security in developing countries (Adger et al. 2014, 777). Epistemic communities thus fulfil the purpose of authorising or vetoing securitisation through reason and facts on the basis of (objective) knowledge.²⁴

Moving on, we already know that in free societies, ordinary individuals can – as audiences – object to securitisation that concerns them, hence there should be no reason why they cannot also object to securitisations that concern others. Many UK citizens, for example, have vocally objected to President Trump’s securitisation of borders, including on social media platforms and in protest movements. Indeed, from Maria Barkardjieva’s work on social media we can infer that social media increases the number of ordinary individuals seeking to influence securitisation. Thus whereas traditional forms of activism take time, many people do not have, social media activism has infused ordinary people with ‘the sense of agency and efficiency – “I/we can do something”’ (Bakardjieva 2015, 985). Moreover, far from being a purely individualistic exercise, social media has facilitated the emergence of new social movements, ‘collective “we”s [making] claims for recognition in the political process’ (Ibid, 986).

While mass protests against President Trump and his security policies during Trump’s state visit to the UK in 2018 failed to alter his securitisation of borders; in principle at least, protests (especially violent ones) can help to keep the powers that be in check. In short, ordinary people can assume the role of rebel.

Individuals with large amounts of symbolic capital can also act to successfully champion securitisation. Former Vice President of the US Al Gore comes to mind here regarding his widely publicised efforts to first request and later endorse the securitisation of climate change to safeguard future generations. This example also shows the difference between on the one hand epistemic communities and champions on the other, thus the latter may make use of scientific facts to make their case, but they are not engaged in the production of scientific knowledge. That is, they are not an authority on e.g. climate change, but because of their celebrity status they can have sufficient symbolic capital.

This brings me to the end of the different categories of functional actor. Much like the CS’s sectors of security, these categories are not absolute, nor is this list meant to be complete. Moreover, it is possible for (groups of) people to occupy more than one category of functional actor at the same time. Academics in security studies operate as epistemic community, when they endorse desecuritisation (as do securitisation scholars), or – as do the Welsh School and human security scholars – endorse the securitisation of the individual from a large variety of threats, many of which are non-military in nature. Yet academics are also gatekeepers. Academic journals are subject to peer review and security studies journals control the knowledge published on their pages. For instance, even without evidence to support my point I am certain that readers will agree that no reputable journal in security studies would publish an article that objects to the securitisation of climate change on the grounds that climate change is entirely natural. In any case, future research on functional actors must consider to what extent public opinion influences the vetoing of securitisation by distinct functional actors, while it must also consider the conditions under which functional actors are most likely to succeed.

Conclusion

This article started from the observation that nothing has been written about functional actors in an otherwise burgeoning literature on securitisation theory. A closer examination, however, revealed this to be untrue; hitherto functional actors have been referred to as audiences. Notably, in securitisation studies audiences are not merely the addressees of securitising speech acts, but a wide array of actors able to authorise securitisation. Given that securitisation scholars are mostly sceptical of securitisation, audiences are normatively important. Ultimately, evidence of audiences’ dissent problematises securitisation.

By denying that audiences are categorically decisive for the success of securitisation as well as by limiting audiences to addressees of securitising speech acts this article goes against accepted wisdom. As such I anticipate that many securitisation scholars will object to the arguments advanced in this article. I want to end my analysis by explaining why they should embrace the view offered here instead. My first reason is that academics must realise that in cases where they are (part of) the referent object of securitisation (for example, when the referent is the electorate), they themselves are audiences and as such can veto securitisation on behalf of themselves.²⁵ Second, in all other instances of securitisation, securitisation scholars can be, and already are, functional actors. In other words, securitisation scholars can enter securitisation processes in which they are not referent object or threatener in a critical capacity directly, without relying on ‘audiences’. As academics with access to privileged platforms (including by giving public lectures, appearing on the radio, political TV shows, by writing opinion pieces in newspapers and by giving evidence to government select committees) they are already better positioned to veto or endorse securitisation on behalf of others than an average member of the general public, notably – and especially when like-minded academics come together – they serve as epistemic communities.

To conclude, it is time that securitisation scholars stopped staking their normative hopes on audiences and to mistake functional actors for audiences. Instead scholars ought to realise that they too are functional actors with – in liberal democracies – the right to veto or endorse securitisation on behalf of others, and hence – at least ˗ the potential power to change its occurrence and form.

Acknowledgements

I would like to thank Thierry Balzacq, Barry Buzan, Ian Paterson, Roxanna Sjöstedt and Timothy Potenz for their excellent and superbly helpful comments on earlier versions of this article. Special thanks to Gareth Perham Jonas for copyediting the piece and for his insightful feedback, as well as to Adam Quinn and Jonathan Parry for some last minute assistance. Finally, my thanks to the reviewers for their very instructive feedback and to the editors of CsOS, especially David Mutimer, for allowing me the time and space to submit a final version.

Disclosure statement

No potential conflict of interest was reported by the author.

Additional information

Notes on contributors

Rita Floyd

Rita Floyd (PhD) is a Senior Lecturer in Conflict and Security at the University of Birmingham, UK. Her work on securitisation and environmental security has appeared in a number of peer-reviewed journals. Her 2019 book The Morality of Security: A theory of Just Securitization (CUP) examines when securitisation is morally permissible. Dr Floyd is currently working on a follow-up monograph. Provisionally entitled The Duty to secure: From just to mandatory securitization, her current project examines whether actors have a moral obligation to securitise, to whom and on what issues.

Notes

1. Some scholars explicitly identify functional actors as part of their empirical research. Vuori 2011b focuses on the media in this capacity, others (Sjöstedt 2011; Vultee 2011) clearly mean functional actors when they write about the media, but – to my knowledge – no theoretical/conceptual work that probes this category exists.

2. Person or persons at the source of the threat. Traditionally this might have been called aggressor, however, not all threats are agent-intended. Climate change for example is agent-caused but not intended (cf. Floyd 2019b, 394).

3. In the UK the phrase ‘on behalf of’ encompasses both doing something for another person’s benefit or interest, as well as acting as a representative of another. In the US the former is often referred to as ‘in behalf of’ and only the latter as ‘on behalf of’ (cf. Soanes 2000, 87). I have decided to stick with the UK’s ‘on behalf of’ here. I use it to mean that functional actors object or endorse securitisation for the sake of other people/groups/entities (notably, referent objects and putative threateners). Moreover, while it is possible that functional actors object on behalf of others with the intention to primarily help or benefit themselves, the objection will almost certainly need to draw on the security equation in question (e.g. by stating that x is not really threatened).

4. A justification is provided for this imbalance on page 42 of SANFFA. Moreover, in their analysis of the societal sector the CS writes ‘[…] the media is an important actor that contributes significantly to the definition of situations. Who are the parties to conflicts; what are the conflicts about? With its attraction to simple stories, the media will often tell the news in terms of “us” versus “them” or, in the case of foreign news, of “Serbs” and “Muslims”. When ethnic or religious categories are established as the interpretative instruments for understanding a situation, the media has often played a role in this’ (Buzan, Wæver, and de Wilde 1998, 124). This raises the following question: is – for the CS – the media a functional or securitising actor? In my view this statement leaves it unclear. In support of the fact that the media is a functional actor here is the fact that it ‘significantly influences decisions in the field of security’ (cf. Buzan, Wæver, and de Wilde 1998, 36); against and thus for the status of securitising actor (or – in some cases – securitising requester), stand the fact that by invoking ‘us’ versus ‘them’ dynamics the media ‘calls for security on behalf of the referent’ (ibid, 36).

5. Here in the widest sense, thus a strategy of aggression also is aimed at providing security, if by expansion and domination.

6. The CS argue: ‘Securitization is not fulfilled only by the breaking of rules (which can take many forms) nor solely by existential threats (which can lead to nothing) but by cases of existential threats that legitimize the breaking of rules. Still, we have a problem of size and significance. Many actions can take this form on a small scale […] a successful securitization thus has three components (or steps): existential threats, emergency action, and effects on interunit relations by breaking free of rules’. (Buzan, Wæver, and de Wilde 1998: 25–26, emphases added). Extrapolating from this we can say that securitisation is successful when countermeasures are adopted, while securitisation exists when an issue/object/entity has been framed as a security issue by a securitising actor (cf. Floyd 2016). For many securitisation scholars, the latter includes that the threat narrative has been ‘accepted’ by a relevant audience.

7. As I argue below threateners are one type of audience of securitising moves, as such they can object to securitisation.

8. Not literally once, as this point is restated a couple of times throughout.

9. It is not necessary that both audiences are present in all securitisations. In securitisations of natural disasters, for example, there is no threatener.

10. I return to the issue of legitimation later on.

11. In the same way as audiences can terminate, for instance, public lectures by walking-out.

12. To be clear, other issues influence securitisation. For example, regardless of whether they speak their view on securitisation an arms industry important for exports and balance of payments can become an important factor in securitisation. This raises the question as to whether functional actors need to be active in the way I propose. I think that they do, because while the arms industry can gain the status of actor, unless they vocally endorse or veto securitisation they do not realise that agency. In plain English, factors are not (functional) actors.

13. This is evident also from the categories of lead and support actors in the environmental sector (cf. above and Buzan, Wæver, and de Wilde 1998, 78).

14. Alas, in liberal democracies the right to free speech is in danger of becoming a right only on paper. Especially where the securitisation of identity (by numerous disparate groups) is concerned, many sections of society (including public intellectuals, academics, but also public sector employees) are no longer fully able to voice objections (cf. Fukuyama 2018; Haidt 2017).

15. Note that some scholars deny the possibility of requesting desecuritization as any linkage between security and an issue recreates securitisation (Behnke 2006, 65). More recently a similar point has been advanced by Oren and Solomon (2015) who argue that audiences have accepted securitising moves when they chant slogans/logics contained within such moves, regardless of whether they believe in them. For more detail on who can and indeed should desecuritize see (Floyd 2019a, 184–189).

16. Actors marked with a star are vital to a fuller understanding of securitisation processes; however, not all of these actors will necessarily feature in each and every securitisation.

17. Please note The Independent, The Financial Times, The Times and many others feature almost exclusively stories condemning Trump’s decision. I do not list them here because they are not editorials. But see for example: https://www.bbc.co.uk/news/business-44324565.

18. Note here that the number of factors influencing securitisation in more or less free societies may well be no different (cf. FN12 above).

19. This also enables me to give yet further examples of functional actors within a limited word count.

20. Given that parliament – in matters of national security – is not the referent object of securitisation, I hold that parliament is not the audience but instead a functional actor in the same securitisation process. However, considering that parliamentarians represent the general public – who may well be the referent object of securitisation – I acknowledge that the border between the two actors can be fluid (here, parallels with the media abound). Importantly, however, and as the discussion shows, the categorisation as either audience or functional actor does not alter the fact that they can object to securitisation processes.

21. It is important here not to overstate this power. Research has shown that in times of emergency, all branches of government tend to pull together and everyone – at the point of the emergence of the threat – is already on the side of securitisation (see Evangelista 2008; De Londras 2011; Neal 2013; Jarvis and Legrand 2016). In other words, few will have to be convinced of the need for securitisation. Up for debate (especially since the 2003 Iraq war), is rather the nature of the precise security measures to be used (note that this further explains the increasingly popular view (in securitisation studies) that audiences have to consent to the nature of the emergency measures).

22. Hence also the widespread view of that the audience acts to (de-)legitimatise securitisation (cf. above). This comes out strongly in debates on humanitarian intervention, thus intervention is generally considered legitimate only when those to be rescued have consented to being rescued by means of war (cf. Pattison 2010, see also Parry (2017) on the role of consent in other-defence).

23. This includes audiences as referent objects of securitisation, who as argued above can succeed in vetoing securitisation.

24. Of course this knowledge is not always objective (cf. Villumsen Berling 2011), as with (de-)legitimaters who – as argued above – bar one exception are not technically in the position to (de-)legitimise, what matters is that epistemic communities are perceived to bring scientific and hence objective knowledge to the table.

25. This also applies in cases where they are threateners.