Fish swimming in denial: non-state armed groups, “propaganda wars”, and “performing” peace processes

ABSTRACT

Using the case study of statements of denial issued by the Irish Republican Army (IRA) over an extended 35-year period, this article critically examines how non-state armed groups (NSAGs) use statements of denial when engaging with various audiences across time and space. It posits that these statements are an integral part of how NSAGs communicate with different audiences during their armed campaigns, and subsequently during the process of transitioning out of political violence. At the same time that these statements feed into the macro-level “propaganda war” between the NSAG and the state, this article maintains that they also reflect the complex intimate relationship between NSAGs and the communities from which they emerge. Arguing that statements of denial help NSAGs to favourably frame how the conduct of its campaign, the character of its members and its internal cohesion are understood by proximate and distant audiences, the article tracks the qualitative changes to IRA statements that would eventually become a key component in the performance of the peace process by the late 1990s.

KEYWORDS:

• Political violence
• propaganda
• peace processes
• denial

Introduction

Even if there is disagreement about whether violence committed for political ends is primarily theatrical or rational in nature, there is nonetheless a consensus that it is communicative (Kydd and Walter 2006; Matusitz 2013; McClure 2014; Kearns, Conlon, and Young 2014). Whether labelled as political violence, armed struggle or “terrorism” (Breen Smyth 2008), this violence has been conceptualised as a form of “propaganda by the deed” (Hoffman 1998, 136). Although a form of propaganda in its own right, an act of political violence is often followed by a claim of responsibility from those behind it (Kearns 2021), usually through statements that justify the act and outline how it fits with their ideology and objectives (Hoffman 1997). Despite this, academic literature has traditionally focused on what organised politically violent non-state armed groups (NSAG) do rather than on what they say about their use of violence (Honig and Reichard 2019).

The vast body of scholarship on the Provisional Irish Republican Army (herein after IRA) is an instructive example.¹ Having fought a protracted armed campaign in pursuit of Irish reunification and national self-determination from 1970, endorsed a peace settlement (via its political wing Sinn Féin) in 1998, and then formally ended its campaign in July 2005, the IRA were the most heavily researched NSAG prior to 9/11 (Silke 2006). While the emphasis of many researchers has moved elsewhere, the IRA case study retains its relevance today (Morrison and Gill 2016). Although noted for being “pragmatic in their use of language” (Honig and Reichard 2019, 774), the extensive literature on the IRA, with a few exceptions (Wright 1990; Filardo-Llamas 2013; Molloy 2015; Cardillo 2017), has focused primarily on what they did rather than on what they said about what they did (Morrison 2016b). However, as Laura Filardo-Llamas (2013) has argued, a more nuanced examination of statements issued by the IRA can help us understand how they justified their actions through articulating their own ideological worldview.

Accepting that point, this article merges the literature on the sociology of denial, political violence, and peacebuilding to conduct a critical discourse analysis (CDA) of statements of denial issued by the IRA. Although silence is the most public form of denial (Zerubavel 2006, 4), and NSAGs often exercise silence when they choose not to claim certain actions (Pluchinsky 1997; Kearns 2021), this article is not a study of denial as silence. It critically examines those cases where the IRA chose to actively and publicly issue a statement premised on disclaiming, disowning or dissociating from a particular act or actor. In contrast to silence, this understands denial to be an active process of meaning making (Zerubavel 2006, 9). It draws on an analysis of hundreds of IRA statements sourced through An Phoblacht and Republican News,² the self-published newspaper(s) used by the IRA to engage with various audiences during the conflict and throughout the process of disengaging from armed struggle (Somerville and Purcell 2011; Molloy 2015; Hoey 2018). This dataset was subjected to CDA, an analytical framework that interrogates how language, power and ideology are inter-linked (Fairclough 1995, 23). While CDA explores how and why certain actors choose to articulate their reality to various audiences through specific words, discursive strategies and mechanisms (Wodak 1989), it also recognises that both texts and the social processes behind their production need to be critically examined (Cardillo 2017). Having previously adopted the CDA approach to IRA statements, Laura Filardo-Llamas (2013) argues that it is useful in showing how the IRA legitimised certain actions, how they interacted with proximate and distant audiences, and how they used particular words or phrases to enact their ideological worldview. Building on that approach, this article focuses on how the IRA, through statements of denial, legitimised itself and its actions while at the same time delegitimizing other actions and actors in accordance with its ideological worldview and self-image.

By doing so, this article seeks to rebalance scholarship on political violence in three important ways. First, it focuses exclusively on what the IRA said, rather than on what it did. Second, by turning a lens towards NSAG denial it provides a counter-weight to the considerable literature – including from the North of Ireland³ (Rolston and Gilmartin 2000; White 2015) – on discourses of denial surrounding state violence (Cohen 1993, 1996, 2001). However, a statement of denial should not be automatically assumed to be untrue; while an IRA apology in 2005 for the death of Derry teenager Kathleen Feeney⁴ contradicted a previous denial of responsibility in 1973 (Feeney 2015), a 1971 denial that the bombing of McGurks Bar⁵ in Belfast was the result of an IRA “own goal” (Republican News 1971e) transpired to be true. In some cases, like the denial of IRA involvement in the 1972 Abercorn Restaurant bombing (Republican News 1972a), indeterminacy still exists. Finally, it highlights how statements of denial are used by, and useful to, NSAGs during peace processes and transitions out of political violence. Little attention has been afforded to how and where NSAG denials fit into such processes. By including denials issued by the IRA following the cessation of its violence, the article critically examines how these statements became integral to the “performance” of the peace process to various audiences, particularly given how what actors in that process were doing (and saying) “front stage” did not always match what they were doing and saying behind the scenes (Dixon 2018, 4).

The article opens by exploring the divergent audiences that IRA statements were directed at during the conflict. Critiquing how both internal and external audiences were engaged through statements of denial, it examines the discursive “propaganda war” that accompanied the physical war in Northern Ireland (NI). It then proceeds to unpack the theoretical literature on denial, revealing how denial is a more complex and nuanced process than is often assumed. The remainder of the article then applies these theoretical insights to empirical data, positing that IRA denials fall into three categories: conduct-based denials about how the IRA fought its armed campaign; character-based denials about the moral integrity of the IRA; and organisational-based denials about the internal cohesion of the organisation. Acknowledging the importance of what Max Abrahms (2018, 9) calls the “branding” of NSAGs by the leadership, it is argued that these denials were integral to how the IRA interacted with both proximate and distant audiences during the conflict and during the peace process. Ultimately, by the end of their campaign in July 2005 statements of denial that had once epitomised a counter-hegemonic challenge to state narratives had themselves assumed a hegemonic character in the intra-republican battle over the peace process.

NSAGs and audience

“Smart” NSAG leaderships, Max Abrahms (2018, 6) argues, recognise the importance of “branding” their organisation to multifarious audiences. NSAG statements can therefore speak to a “double audience” (Filardo-Llamas 2013) comprising its own membership and support base (Picard 1989), other actors in a conflict or peace process, and external observers. They can communicate different messages to different audiences, whether that be reassuring supporters of the organisation’s integrity, convincing international observers that the group is not morally toxic as the enemy claim, or demonstrating its resolve to the adversary (Honig and Reichard 2019). The willingness of, and necessity for, a NSAG to engage with new audiences might change over time. For example, while the IRA relied primarily on the “propaganda of the deed” during the initial stages of the conflict, by the 1980s it had become more propaganda savvy when engaging with a multitude of target audiences (Somerville and Purcell 2011). The increased involvement of international actors in the peace process broadened out the audience for IRA statements even further by the turn of the millennium (Cox 1997; Dixon 2018).

Although state discourses dismiss NSAGs as alienated purveyors of senseless violence, observed wisdom demonstrates that NSAGs are firmly embedded within the social structures of, and reflect grievances within, the communities from which they emerge (Beck 2009). Yet the fact remains that NSAGs are reliant on popular support for survival (Tse-Tung and Griffith 2005; Crenshaw 1998), particularly when they adopt guerrilla warfare tactics during asymmetrical conflicts. This reality forces NSAGs to incorporate Mao’s aphorism about the “fish and the sea” into their “branding”, organisational culture and discourse (Bangerter 2011; Abrahms 2018, 12). While this means not unduly imposing themselves on the local civilian population, it also requires convincing their communities to support the armed campaign just as the “sea” supports the “fish” (Tse-Tung and Griffith 2005). The civilian population must therefore be convinced that the cause is worth supporting and then further reassured that the righteous fight will be conducted within permissible moral and ideological boundaries (Wright 1990: xii). A sympathetic community, then, would see the NSAG as fighting for a just cause against an unjust aggressor who imposed conflict upon the group (Bar-Tal 2009). Whatever its detractors may have argued about political violence being a minority pursuit in the North of Ireland, it is simply inconceivable that the IRA could have sustained a 30 year campaign without considerable support among certain sections of the Nationalist community (Morrison 2016a, 625).

Although Irish republicanism is a complex ideology bridging disparate interests (McGlinchey 2021), in its simplest manifestation the justification offered for IRA violence was premised on historic tropes of nationalism and centuries of struggle for self-determination (English 2003; Morrison and Gill 2016). The IRA framed its armed campaign as a modern national liberation struggle to reverse the injustice of partition, expel the British occupier and reunite the island of Ireland into a single sovereign entity (Wright 1990). However, the nature of and motivation for IRA violence has always been bitterly contested; what was framed by the IRA as a legitimate national liberation struggle was decried by other conflict protagonists as sectarian “terrorism” and criminal violence (White 1997; Bruce 1997; Dingley 1998; White 1998; Kowalski 2018; McCleery 2021; McGlinchey 2021). Notwithstanding this, the ability to frame their campaign as a national liberation struggle fostered feelings of loyalty, community and political legitimacy among IRA volunteers and their wider support network (Cochrane 2007).

This did not mean, however, that the IRA was given a “blank cheque” of unconditional support by its support base. Rather, as English (2003, 30) has noted, broad sympathy for Irish republicanism and support for particular acts of IRA violence have always been very different things. Support ran along a continuum with active support that sheltered IRA activists and stored ammunition at one end to passive support that shared some ideological commitment to or sympathy with the IRA’s objectives at the other end (Smyth 1989). Indeed, the levels and type of support given to the IRA by the Nationalist community fluctuated throughout the conflict (Burton 1978; Sluka 1989), suggesting that a tacit understanding had emerged whereby the support base would continue to support the armed campaign provided the IRA conducted it within morally acceptable parameters. When these parameters were breached, the support base would, at least temporarily, withhold its support (Burton 1978, 85; Morrison 2016a, 625).

In capturing this reality, Wright (1990:xiii) has argued that IRA statements were directed at three primary internal audiences: the “uncommitted audience” that might enquire further as to why the organisation is engaged in political violence; the “sympathetic audience” that could come to identify with the organisation’s cause; and the “active audience” that had taken action in support of the group. The complexity of communal support that defies dichotomous poles of unconditional support and outright rejection means that an IRA denial might be premised on underpinning its ideological discourse for the “uncommitted” audience, defending the morality of its campaign to a “sympathetic” audience, and reassuring the “active” audience of its continued strength, capacity to operate and resolve to win.

An internal audience, however, was never reliant on IRA statements to shape their worldview of the conflict. They could, of course, assess the complex everyday reality they witnessed firsthand on the ground. This was not the case for more distant audiences outside the urban working-class districts and rural border hinterlands that the IRA primarily operated in. These audiences relied upon mainstream media reporting of IRA violence (Molloy 2015), making a compliant UK media a useful ally for the state in its bid to delegitimise and depoliticise IRA violence (Miller 1994). The Irish media also took a similarly hostile approach to IRA activity (Hanley 2018), making it more difficult for the IRA to appeal to “uncommitted actors” there who could potentially become “sympathetic” and/or “active” through fundraising, logistics efforts and billeting IRA volunteers on the run.

Yet there is narrative interaction between the story that the enemy tells about a NSAG and the story the group tells about itself, meaning that IRA denials usefully challenged the media framing of its campaign as indiscriminate “terrorism” and/or wanton criminality. This became increasingly important for “branding” the group to more distant audiences south of the border, in Britain and in the US who did not have the more rounded view of political violence that those living in the Six Counties had (Molloy 2015). Likewise, statements of denial proved useful in the “rebel diplomacy” of seeking funding, resources and political support from international actors. Not only did these statements “brand” the organisation internationally as a serious force with attainable objectives and an effective strategy to achieve them (Abrahms 2018, 6), but they were also a useful counter-balance against media claims of involvement in activity that offended the political, social or cultural sensitivities of possibly supportive external audiences. This is particularly true, for example, in distancing itself from international communism lest it upset a significant Irish-American support base. At the same time, though, it also had to distance itself from Western imperial powers to retain support from non-Western sponsor states like Libya (Cox 1997). An IRA denial, then, might have spoken as much to an “uncommitted” actor in Cork looking on at the IRA campaign or a “sympathetic” actor putting money into collection buckets in a Chicago bar, as it did to “active” volunteers operating in Crossmaglen.

Unpacking denial

Denial is more complex than simply contesting whether or not a certain thing happened. Where acts of violence are concerned, it can relate to the existence, nature or significance of that act (Cottee 2005). While denial might involve a refusal to “accept facts as facts”, it can also manifest itself through “complex strategies of relativisation” that are excusatory and/or justificatory in nature (Ignatieff 1996, 118). Stan Cohen (1993) argued that denial can develop through three stages: outright denial of any wrongdoing; reclassification of what has taken place; and an admission of sorts that is usually tempered by justificatory language. Cohen (2001) later developed this into three forms of denial: literal denial premised on contesting the existence of wrongdoing (i.e. “nothing happened”); interpretive denial premised on reframing the meaning of what happened (i.e. “something did happen but not as you think”); and implicatory denial premised on justifying what would otherwise appear to be an act of wrongdoing (i.e. “what happened has happened for a good reason”).

Interpretive and implicatory denial are closely bound to techniques of neutralisation that justify certain actions against certain persons by deflecting blame away from those who perpetrated these acts (Matza and Sykes 1957). In its crudest form, this can involve using legalistic or militaristic euphemisms whereby civilian deaths and casualties quickly become “collateral damage” (Welch 2004). The responsible actor therefore draws from an established “vocabulary of excuses” that is contextualised within “situational ethics” (Labeff 1983, 29). When the IRA targeted contractors working on military installations, for example, it did not deny targeting these people but chose to frame it as the targeting of “legitimate targets” who were assisting the “British war machine” rather than the targeting of civilians (Hearty 2019). Similarly, NSAGs might deny deliberately targeting civilians by citing the “technical fault excuse” that attributes civilian deaths to a technical fault or resorting to the “rogue soldier” excuse that attributes acts to individual members acting without the sanction of the group leadership (Honig and Reichard 2019, 770–2).

Existing literature on NSAGs shows that they routinely use the various forms of denial to deny their presence in a certain area, their activities, or their past, current and future intentions (Wirtz 2008). While it is natural that they would deny actions that prove ideologically damaging and publicly unpopular (Kearns 2021), they may also deny otherwise embarrassing intelligence failures or mechanical faults (Pluchinsky 1997). NSAGs can also use denial as a “strategic instrument” to shape their environment so it is more conducive to attaining their core goals (Godson & Wirtz 2000, 425). “Strategic denial” relating to the “bigger picture” can also be fostered by, and within, the senior ranks of the NSAG (Godson & Wirtz 2000, 425), something which takes on heightened relevance during peace negotiations and movement out of conflict.

Categorising IRA denials

Adhering conceptually to Cohen’s typologies of literal, interpretive and implicatory denial, IRA denials can be empirically differentiated as conduct-based, character-based or organisational-based: conduct-based denial is a denial that the IRA were involved in a certain action that transgressed the accepted moral code of a national liberation struggle (i.e. a denial of deliberately targeting civilians); character-based denial is a denial that IRA volunteers were motivated by anything other than ideology (i.e. a denial that IRA personnel were misappropriating funds for private use); and organisational-based denial is a denial related to the IRA’s internal workings (i.e. a denial about splits within the IRA). Each of these forms of denial exhibit traits identified by Cohen, mapping usefully onto the “branding” of the IRA at different junctures throughout the conflict and the peace process.

Conduct-based denial

The ideology of a NSAG often determines how it conducts its campaign (Wright 1990: xi), either by narrowing or expanding the range of actors deemed “legitimate targets” (Hearty 2019) or by placing or removing limitations on the amount of force it is prepared to use (de la Calle and Sanchez-Cuenca 2006). The belief that it is engaged in a legitimate war of national liberation can take root within a NSAG and condition both its practice and its discourse (McEvoy 2003), leading it to, rhetorically at least, conduct its campaign in loose accordance with International Humanitarian Law (IHL) norms (Noortmann 2002; Bangerter 2011; Dudai 2011; Dudai and McEvoy 2012). Often, however, NSAGs will afford themselves greater “moral elbow room” (Hearty 2019, 1127) than IHL does when it comes to determining just who and what represents a “legitimate target” (de Burca 2014, 57).

According to the IRA’s Green Book, its campaign was directed against “foreign occupation forces and domestic collaborators” (CAIN 2020b). Within this framework the concept of “legitimate target” included British soldiers, members of the home-grown Ulster Defence Regiment (UDR), police officers in the Royal Ulster Constabulary (RUC), government financiers, members of the UK government, loyalist paramilitaries, contractors working on military installations, and those passing on information to the security forces (Dudai 2011; Hearty 2019). This expansion of the “legitimate target” concept was hardly unique to the IRA; Frente Revolucionária de Timor-Leste Independente (the Revolutionary Front for an Independent East Timor) (Fretilin) in Timor Leste (Stanley 2009, 76), the Front de Libération Nationale (National Liberation Front) (FLN) in Algeria (Branche 2007), and Euskadi Ta Askatasuna (ETA) in the Basque Country (de la Calle and Sanchez-Cuenca 2006) all targeted collaborators, opposition activists and leaders, the enemy administration and local politicians and officials during their armed campaigns.

Even if a NSAG restricts its campaign to targeting “legitimate targets” in loose accordance with IHL norms, the reality remains that this has never entirely prevented the death and injury of civilians (Rothbart, Korostelina, and Cherkaoui 2012, 3). The problem becomes more pronounced where NSAGs find themselves on the wrong side of asymmetrical warfare, largely due to the fact that it is difficult to gauge the proportionality of “collateral damage” against their military objectives and/or gains (Benvenisiti 2009). To problematise this even further, the support base will often have a lower tolerance for particular types of violence than the NSAG will have (de la Calle and Sanchez-Cuenca 2006). Exceeding the threshold for this by, for example, using proxy human bombs, can cause unease within the wider support base (de Burca 2014, 85). Certainly, in high-profile cases where the IRA went beyond what their support base deemed acceptable, the “sympathetic” and “active” audiences, as leading republican Danny Morrison readily conceded (Morrison 2016a, 265), did not hesitate to push back.

Not only are civilian deaths injurious to morale among the “active” and “sympathetic” audiences, but they also enable the state to create further distance between a NSAG and the “uncommitted audience” by framing these deaths as less the inevitable consequences of war and more the natural outworking of “terrorism”. “Smart” NSAG leaderships recognise as much, endeavouring to steer the group clear of these unpopular activities and subsequently mitigating against reputational costs where the “branding” of the organisation has been publicly deviated from (Abrahms 2018, 9). Unsurprisingly, then, NSAGs will often move to publicly deny that civilians were knowingly and deliberately targeted during the planning of their operations (Honig and Reichard 2019). This, of course, allows the NSAG to maintain its self-image as a legitimate national liberation movement adhering to the standards expected of such a body, to mitigate against the damage to morale caused by operational blunders, and to counter state criticism of their campaign as indiscriminate “terrorism”.

The IRA denied responsibility for the deaths of those who could not be deemed “legitimate targets”. For instance, they denied responsibility for the death of local West Belfast businessman Sean McNamee during an armed robbery on his factory (Republican News 1975).⁶ Although it seems only natural that the IRA would deny publicly damaging actions like this (Kearns 2021), it is difficult to ascertain the sincerity of denials early on in the conflict where the intensity of the conflict, the sporadic nature of the violence, and the toxic combination of rumour, personal grievance, and retribution often obscured the motivation behind a particular death and/or the party responsible for it (Burke 2018: viii; Andrew and Wood 2012, 6). Likewise, the natural difficulty of accurately collating such frenetic activity was further compounded by the rapid influx of new recruits to the IRA after internment in August 1971 and Bloody Sunday in 1972 (White 2017) that made it more difficult for the IRA leadership to tightly regulate the activities of all its members. Accordingly, some members could have taken unsanctioned action that was subsequently denied by the organisation. During this period, the state had also adopted the Kitsonian counterinsurgency strategy of using counter-gangs, agents provocateurs and “black flag” operations (Kitson 1971), including plain clothes members of specialist British Army units launching attacks on unarmed civilians in Nationalist districts in order to discredit the IRA. Hence, the IRA denied responsibility for a bomb attack on the Ulster Brewery that was subsequently claimed by the obscure group The Irish Freedom Fighters. Denying that any such group existed, the IRA argued that the attack was part of “British black operations” designed to destabilise and smear the IRA (Republican News 1978b).

In other cases, the IRA accepted responsibility for civilian deaths but denied that they were intended. For example, while it took responsibility for civilian deaths as a result of the July 1972 Bloody Friday bombings, it nonetheless dismissed claims that warnings had not been given in advance as “an attempt to discredit the Republican Movement” by “the British propaganda machine” (Republican News 1972c).⁷ Regrettable though these deaths may have been, they seemed to reflect, from the IRA’s perspective at least, what Honig and Reichard (2019, 772) call the “unpredictable nature of the battlefield” excuse. Here, denial is rooted in the “propaganda war” over how IRA violence is framed; a national liberation war defined by the inescapable, yet sad, reality that civilians will die or as wanton “terrorism” that thrives on this very threat to civilian life. Implicatory denial was also discernible where the IRA defended targeting certain people as “legitimate targets”. For example, they justified killing a contractor who had serviced the security forces by arguing that the victim had “more connections with the RUC hierarchy than most RUC constables and was more important than many officers” (An Phoblacht 1987).

Further to this, the IRA also denied that certain operations had unnecessarily endangered civilians. In one such example, they denied RUC claims that an attack on a British Army patrol near the Park Centre shopping centre in West Belfast had placed civilians at risk; the bomb detonated at 10.45 pm while the shopping centre had closed at 9 pm (An Phoblacht 1989). However, the organisation also defended the conduct of its campaign more generally, dismissing claims that it had used dum dum bullets in contravention of the Geneva Convention (Republican News 1971b); labelling RUC claims that they had used an 80-year-old woman as bait to lure the security forces into a bomb attack in West Belfast as “absurd” (An Phoblacht 1991a); denying that an RUC officer had been shot in Dungannon while tending an injured woman (Republican News 1973a); and arguing that marks on the bodies of three men shot dead as informers were “caused as a result of the execution” and were not signs of torture as reported by the media (An Phoblacht 1992b). Just as with denials about deliberately targeting civilians, these examples show the “pull” of IHL norms whereby the IRA frames its conduct within the assumed moral code of a national liberation struggle.

Notably, the IRA also denied being engaged in conflict-related activity south of the border, including involvement in bombs and hoaxes (An Phoblacht 1976), robberies (An Phoblacht 1977), a bomb attack on the chief forensics officer of the Department of Justice (An Phoblacht 1982) and arson attacks on businesses there (An Phoblacht 1981b). They claimed that while the Dublin government were aware that “99 per cent” of this activity was unconnected to the IRA they were nonetheless using it as “a pretext for intensifying the harassment and repression of the Republican Movement” (An Phoblacht 1980b). Where IRA activity did result in members of the Irish security forces being killed this was framed in interpretive denial terms of being unintended (An Phoblacht 1983b), and in other cases, like the killing of Garda Jerry McCabe, unsanctioned (An Phoblacht 1996). The willingness of the IRA to deny undertaking activity south of the border speaks to the peculiarity of the relationship they had with the state and its agents there (Hanley 2018; Reinisch 2021); arguing that they were not legitimate but a repressive partitionist apparatus, yet at the same time having standing orders prohibiting their members from militarily attacking it. These denials were also useful in offsetting any disquiet hostile media coverage may have stoked up in the “sympathetic” and “active” audiences south of the border who were providing moral, logistical and material support to those in the Northern war zone.

Character-based denial

Unsurprisingly, with the emergence of illicit war economies in any conflict zone (Pugh 2004),⁸ the state sought to depict the IRA as an organisation driven by criminality. Naturally, in the throes of the “propaganda war”, the IRA pushed back against this, mapping such claims onto a wider state criminalisation discourse:

The IRA totally rejects allegations of involvement in pirate videos, pornographic or otherwise. This is the latest baseless allegation in the propaganda campaign which has alleged IRA involvement in everything from drug dealing to Angel Dust and dog fighting. Not a shred of evidence has ever been produced to support any of these claims. People should see these smears for what they are – part of the British government’s dirty tricks campaign which has the eager collaboration of the Gardai. (An Phoblacht 1992a).

In particular, the IRA strenuously and consistently denied any involvement in the drugs trade, dismissing such claims as “psychological warfare”, “anti-republican black propaganda” and “misinformation” (An Phoblacht 1991b). While this allowed the IRA to challenge the state criminalisation discourse aimed at the “uncommitted audience”, it also reassured both the “sympathetic” and “active” audience that the IRA campaign was driven by ideology rather than criminality. Resistance to criminalisation, after all, has long been a central plank of the self-identity of Irish republicans, reflected in their readiness to resist attempts to treat them as “ordinary decent criminals” within the jails (Campbell, McKeown, and O’Hagan 2006; MacIonnrachtaigh 2013; McConville 2021; Reinisch 2021). Even today, former IRA members regard themselves as communal defenders whose use of violence was qualitatively quite different from the violence of the common criminal (McEvoy and Shirlow 2009).

Interestingly, while the IRA never denied fundraising via armed robbery, it did deny carrying out robberies that were “against the interests of our people” (An Phoblacht 1979b). Such actions would, as a denial of involvement in the robbery of Twinbrook post office pointed out, only increase the “hardship” already faced by working-class Nationalist communities (An Phoblacht 1981c). Reaffirming its image as communal defenders, the IRA denied involvement in the robbery of GAA clubs in Armagh city (An Phoblacht 1986), the robbery of a chapel renovation fund (An Phoblacht 1981a), and the robbery of a till in a Coalisland shop during a commercial bombing (Republican News 1978a). The fact that the IRA were prepared to publicly deny many small scale, localised acts like this demonstrates just how receptive they were to the sensitivities of the “sympathetic” and “active” audiences. While these actors might be somewhat ambivalent about the IRA financing its campaign at the expense of multi-national corporations or government agencies, they would not tolerate it being funded at cost to their own community.

In addition to denying involvement in criminality, the IRA also denied claims that its members were involved in petty thuggery and anti-social behaviour. It denied that its members were responsible for wrecking a South Armagh ballroom (Republican News 1973b), for a hatchet attack on an Alliance Party member (Republican News 1973c), and for attacking the home of a priest following disagreements with him (An Phoblacht 1980a). These denials cut against macro-level state narratives that dehumanised those engaged in political violence as criminally deviant and psychopathic (Sluka 2012, 285) and that claimed the IRA was bullying its community into submission (Sluka 1989, 166). By moving to reassure the “sympathetic” audience that the IRA did not engage in these anti-community activities, the organisation showed acute awareness of the Maoist warning that the NSAG “fish” should not upset the mood of the “sea”.

However, disassociation from certain problematic acts and/or actors was also for the benefit of external audiences too. In particular, the organisation was keen to dispel claims that it was a fellow traveller of certain other politically violent movements elsewhere. Notably, this involved denying that it was linked to left-wing movements on the continent (An Phoblacht 1984), lest this upset the political sensitivities of Irish-America. The organisation said that it existed as “a national liberation organisation aimed at ending British rule in Ireland and establishing a united Ireland”, and that claims to the contrary were British propaganda (An Phoblacht 1981d). As such, the organisation denied claims they were responsible for a series of bomb attacks in France, saying that “such actions are unhelpful to our struggle” and “play into the hands of the Imperialist Governments” (An Phoblacht 1979a). It also denied claims by the West German police that they were linked to left-wing groups that had attacked US personnel stationed there (An Phoblacht 1984). The importance of assuaging concerns over links to left-wing groups hostile to US interests was again shown in 2001 when the organisation denied training FARC guerrillas in Colombia (An Phoblacht 2001a). Such was the importance of maintaining an ideologically and morally pure image to the Irish-American support base that the organisation even rejected as “smear” accusations made in a US courtroom that it had been providing protection to John De Lorean who was then on trial for cocaine smuggling (An Phoblacht 1983a).

Just as with criminality, self-gain cuts against the grain of the Irish republican self-image; an image premised on hardship, sacrifice and struggle rather than personal enrichment (Campbell, McKeown, and O’Hagan 2006). Unsurprisingly, then, the IRA denied that its members were involved in corruptive self-aggrandisement. This included dismissing sensationalist media claims that a well-known Belfast republican family were set to move to a newly built house south of the border paid for by the IRA (Republican News 1972d) and rubbishing tabloid stories of IRA members getting paid a bounty for killing RUC members (Republican News 1971d).

Evidently, then, there was a symbiosis between the “divisive propaganda” of the state and the “cohesive propaganda” of the IRA (Sharma 2007). Whereas the state’s “divisive propaganda”, alongside its criminalisation policy, depoliticised IRA violence and dehumanised those involved in it as criminal deviants, bullies and corrupt opportunists, the IRA’s “cohesive propaganda” reasserted the ideological nature of IRA violence, the self-image as communal defenders, and narrative tropes of sacrifice and struggle for community, cause and country. This “cohesive propaganda” spoke simultaneously to multiple audiences: to “uncommitted” and international audiences who may be influenced by the state criminalisation narrative, to local support bases who would not tolerate attacks on the local community, and to prospective recruits that criminality, anti-social behaviour and self-gain would not be tolerated within the IRA.

Organisational-based denial

Often blurring the lines between literal and interpretive denial, the IRA issued a number of organisational-based denials about how its campaign was progressing, its internal cohesion and its future direction of travel. Underpinning the leadership’s “branding”, these statements projected a public image of a disciplined and united organisation that was committed to its objectives and confident of attaining them. By the late 1990s, however, these statements had become a key component in the “performance” of the peace process (Dixon 2018), as the organisation framed itself as the key driver of conflict resolution facing politically motivated hostility from the British government and political Unionism (Cardillo 2017).

Early organisational-based denials refuted British claims that the IRA were on the brink of defeat. The IRA denied media reports that they were secretly burying their dead in Milltown Cemetery (Republican News 1971a), were hiding the bodies of dead members in sewers across Belfast (Republican News 1971c), and had suffered dozens of casualties when the 1972 truce ended (Republican News 1972b). Likewise, organisational-based denials also refuted allegations of internal tensions within the prisons. Tensions between prominent figures – driven at times as much by personality as by policy differences – are common within NSAGs, and the IRA were hardly any exception (Hearty 2016; White 2017). Yet, a 1974 statement denied that IRA prisoners had expressed their lack of confidence in the newly elected OC in Long Kesh (An Phoblacht 1974; Republican News 1974). Here again, denial usefully upholds the self-image of the IRA as a unified and disciplined entity, thus assuaging any concerns among the “active” and “sympathetic” audiences that the organisation was teetering on the brink of defeat or descending into distractive infighting.

Given the historic propensity of Irish republican organisations to split (Sanders 2011; Morrison 2013), organisational-based denials have unsurprisingly addressed internal disquiet within the organisation. From as early as 1977, the IRA denied that a split was emerging between the “hawks” and the “doves” within its ranks (Republican News 1977). With the emergence of the “Armalite and ballot box” strategy following Sinn Féin’s post-1981 hungerstrike entry into electoral politics (English 2003; Bean 2007), tension between the “hawks” and the “doves” grew. Two opposing positions became discernible within the organisation by this stage: one that favoured merging armed struggle by the IRA with an electoral intervention by Sinn Féin, and a more hardline position that saw engagement in elections as a distraction from, if not detriment to, the armed struggle (Sanders 2011; Morrison 2013; Hearty 2016; White 2017). The IRA said that media reports of a split within the organisation over this matter in April 1985 were “totally untrue” and had arisen after the dismissal of two IRA volunteers in Belfast for disciplinary reasons (An Phoblacht 1985).⁹

However, internal division soon publicly manifest itself when some traditionalists resigned from Sinn Féin after the party dropped its longstanding policy of abstention to the Irish parliament in 1986. Although several of the veterans central to the formation of the Provisional IRA left at this juncture (Sanders 2011; Morrison 2013; White 2017), the impact was felt much more acutely in Sinn Féin than in the IRA. Sanders (2011, 137) argues that the arrival of Libyan weaponry enabled the IRA leadership to placate militarists who might otherwise have defected, largely because it gave grassroots volunteers the impression that the IRA campaign was being intensified rather than wound down. While the small number of defectors did form a breakaway group (later to become the Continuity IRA), it was several years before the group publicly revealed its existence and its operational record was never comparable to the IRA’s (Hearty 2016). Notably, this juncture sees IRA statements become more inward facing as the IRA leadership tries to manage the “active” and “sympathetic” audiences through a changing political constellation on the island of Ireland (Somerville and Purcell 2011). As back channel negotiations with the British and other actors intensified and media reports of a possible cessation of IRA violence emerged by the early 1990s (O’Dochartaigh 2021), this “bigger picture” strategic denial became more important.

Entry into peace processes exposes how NSAGs are not monolithic organisations (Darby and Mac Ginty 2003), forcing leaderships to carefully manage their constituency into accepting a cessation of hostilities (Zahar 2008, 165). By the 1990s, this was reflected in the pages of An Phoblacht (Cardillo 2017; Hoey 2018), as IRA statements became integral to how the nascent peace process was framed by the leadership for the “active” and “sympathetic” audiences. Although it has since become clear that secret negotiations had been taking place for a number of years (O’Dochartaigh 2021), the Hume-Adams statement in 1993 confirmed for many republicans that the leadership had begun a political dialogue about ending violence. The IRA leadership, though, had to handle this reality sensitively, thus rejecting a “preposterous claim by the London government” that the organisation had asked for its help in bringing their campaign to an end (An Phoblacht 1993). Whether these claims were accurate or otherwise, by August 1994, the IRA had nevertheless announced “a complete cessation of military operations” (CAIN 2020a) following a number of earlier temporary cessations.

Despite growing frustration over political demands being made of the IRA during the peace process – resulting at one point in the IRA ceasefire being temporarily suspended from February 1996 to July 1997 – an IRA spokesperson told An Phoblacht (1996) that there was “no question of a split within the IRA”. However, fracture lines within the organisation became obvious following public resignations from Sinn Féin – accompanied by resignations from the IRA – in 1997 over support for the Mitchell Principles of non-violence in advance of the all-party talks that would lead to the Good Friday Agreement (GFA) (Sanders 2011; Horgan 2013). This saw a shift from literal denial of internal disagreement to interpretive denial that minimised the split. The IRA acknowledged that there had been a number of “regrettable” resignations, yet maintained that because those who resigned were few in number and closely associated with one another, “the IRA remains intact, united and committed”, despite media speculation to the contrary (An Phoblacht 1997). Indeed, when the moment of reckoning came, the GFA was endorsed by a significant majority of the republican movement (i.e. the IRA & Sinn Féin) in 1998.

While acceptance of the GFA marked an irrevocable milestone in the IRA’s transition out of political violence, peace processes are protracted events that require a constant management of core constituencies (McEvoy and Shirlow 2009). Moving too fast or too far with further post-GFA concessions risked pushing the “active” and “sympathetic” audiences into the hands of newly emergent splinter groups. Even if the politically-minded IRA leadership was always significantly ahead of its rank and file throughout the peace process, careful choreography of gradual and measured change carried the bulk of the movement with it by portraying this as the grassroots driven empowerment of the movement (Maillot 2008, 309). The pages of An Phoblacht became a key site for “performing” the peace process to an internal audience through “strategic denial”, particularly on sensitive issues like decommissioning prior to October 2001 (An Phoblacht 2001b). Although the IRA would subsequently move to put its weapons “beyond use”, and republicans would in time make further concessions on policing, by this stage a number of violent and non-violent rival groups had emerged (Frampton 2011; Sanders 2011; Horgan 2013; Morrison 2013; Whiting 2015; McGlinchey 2019).

As had been the case with the IRA itself, these rivals developed counter-media to contest how the peace process was framed and interpreted (Hoey 2018), thus pitching the “active” and “sympathetic” audiences at the centre of a new intra-republican “propaganda war” between those promoting and those opposing the peace process. This new “propaganda war” – which continues today – has seen pro-GFA republicans borrowing from state discourses previously used against the IRA by depoliticising the violence of rivals, arguing that rivals have no support, and claiming that rivals are working in the interests of British “securocrats” (Hearty 2016). For their part, anti-GFA groups have cited the unmasking of senior Sinn Féin figures like Denis Donaldson as British agents to support their claims that the Sinn Féin peace strategy was heavily compromised from its inception (Leahy 2020; McGlinchey 2021). Ironically, IRA statements that were once counter-hegemonic in the “propaganda war” had now become hegemonic in the new “propaganda war”.

Inasmuch as the peace process was being “performed” to an internal audience via “strategic denial”, it was also performed to external audiences too (Dixon 2018, 2). This saw IRA statements being used to legitimise the organisation’s actions throughout the peace process (Filardo-Llamas 2013) and to challenge what it said was a misrepresentation of these by others (Cardillo 2017). This involved framing the IRA as the key driver of the peace process while its political opponents remained entrenched in conflict mode. Consequentially, the organisation denied that it had breached its ceasefire on several occasions, attributing claims to the contrary as “misinformation” (An Phoblacht 2002) and “an effort to undermine public confidence” in the peace process (An Phoblacht 2003). Notable post-GFA denials include denials of importing arms from America (An Phoblacht 1999), training FARC cadres in Colombia (An Phoblacht 2001a), and breaking into RUC headquarters in Castlereagh (An Phoblacht 2002).

Perhaps the most noteworthy performative use of denial came just months before the formal ending of the IRA campaign when the republican movement faced what Maillot (2008, 298) described as the “closest [thing] … to a public relations disaster since the beginning of the peace process”. Having robustly denied involvement in a high-profile multi-million-pound raid on the Northern Bank at a politically sensitive time in late 2004 (An Phoblacht 2005a), the organisation soon found itself denying any involvement in the death of Robert McCartney following a Belfast bar brawl in January 2005. Initially, this involved a literal denial that “the IRA was not involved in the brutal killing of Robert McCartney … which runs contrary to republican ideals”. Responding to allegations of intimidation against witnesses, they further instructed that “no one should hinder or impede the McCartney family in their search for truth and justice” (An Phoblacht 2005b). A public campaign by the McCartney family alleging the involvement of IRA members from the Markets area – picked up by the local, national and international media – later drew an interpretive denial from the IRA that while “some republicans were involved … no materials under the control of or belonging to the IRA were produced or used at any time”. Informing that disciplinary action (i.e. expulsion) had been taken against those members involved, the statement warned that “anyone who brings the IRA into disrepute will be held accountable” (An Phoblacht 2005c). Eventually, the IRA disclosed the findings of an internal inquiry into the incident, rejecting media claims that 12 IRA volunteers were involved in the incident by maintaining that “of the four people directly involved in the attacks … two were IRA volunteers. The other two were not” (An Phoblacht 2005d).

The McCartney example chimes with Abrahms (2018, 12) observation that organisational denial is a “proven crisis management strategy” for NSAG leaderships following “face-threatening behaviours” by its members. Notwithstanding this, the organisational-based denials of 2005 pose an interesting question in relation to their reception by various audiences. Obviously, the McCartney killing denials spoke to a localised and national audience well aware of the identity of those involved and their links to the IRA. Rather than persist with the original literal denial, subsequent denials were premised on reassuring these audiences that the killing was neither sanctioned nor condoned by the organisation. Moreover, the “sympathetic” and “active” audiences were being told that the organisation was not prepared to shield the killers. Indeed, the decision to publicly encourage witnesses to assist the McCartney family sat uneasily with the republican movement’s then rejection of policing. The organisational-based denials of 2005 seemed to have nonetheless successfully placated the local and national and the “active” and “sympathetic” audiences, as Sinn Féin showed remarkable political resilience to weather the crises of 2005 (Maillot 2008, 298); it increased its share of the vote in both the Westminster and local elections that year (including in Belfast).

However, external audiences appear to have given the organisation shorter shrift. Notably, a chill factor from the US,¹⁰ as well as growing impatience in London and Dublin, towards the republican movement had defined the prelude to the July 2005 statement. In contrast to the generosity shown by the “active” and “sympathetic” audiences, these actors expected words to be followed by action. Even after July 2005, the IRA sought to convince these actors that it had “honoured all public commitments made in July last year”, despite “politically motivated” reports to contrary (An Phoblacht 2006). This projected a republican willingness to work the peace process in the face of hostility from others (i.e. political Unionism and British “securocrats”). The problem, according to the IRA, was not a failure on its part to commit to peace but rather the fabrication of such a failure by its political opponents in order to weaken republicans and undermine the peace process.

This also helped to “brand” the IRA as the drivers of conflict resolution to the “uncommitted” audience, particularly the post-GFA Nationalist electorate previously reluctant to lend any support to the republican movement during the conflict. Whereas in the past these denials would have been premised on making them sympathetic to the IRA campaign, during the peace process they needed to encourage these actors to vote for Sinn Féin rather than the moderate Social Democratic & Labour Party (SDLP). IRA denials subsequently boosted the electability of Sinn Féin by convincing Nationalist voters that republicans were not only committed to the peace process but that they were best placed to work that process to the advantage of Nationalists. Evidently, this political pitch worked, as Sinn Féin began seriously challenging the SDLP for the dominant electoral position within post-GFA Nationalism (Bean 2007), eventually coming to surpass them by some considerable margin. This was only possible because middle-class Nationalists had reasoned that even though Sinn Féin had shed its support for political violence it could still win concessions for Nationalists within the post-GFA state by taking a hard-line approach to political Unionism (and the British government) (Evans and Tonge 2013).

This appeal to Nationalist voters also overlaps with the “new propaganda war” because both hinge on maximising the apparent success of Sinn Fein’s post-GFA political project. The continued electoral growth of Sinn Féin is contrasted with the sporadic violence of rivals as evidence that the leadership were right to convert the armed struggle into an unarmed political struggle. Rather than being an ideological “sell out”, this was a prescient change of tactics that afforded republicans a “stepping stone” to build towards their goal in the post-GFA state while detractors persist with a futile armed campaign (Hearty 2016). Sinn Fein’s current position as joint partners in the powersharing executive at Stormont and lead party of opposition in Dublin may well vindicate those who subscribed to the “stepping stones” argument, yet purist critics are never slow to point out that holding political office in these “partitionist” institutions was not what the armed struggle was about (White 2017; McGlinchey 2021).

Conclusion

A CDA of IRA denials reveals the complex and diverse roles that such statements can play for NSAGs throughout their armed campaigns and the process of exiting political violence. It shows, too, the need for NSAGs to be “branded” to a multitude of audiences during conflicts and peace processes, whether that is their own members and/or local support base, possibly sympathetic audiences abroad, international actors with influence during political negotiations or the post-conflict electorate. With the emergence of the digital age opening up new fronts of “imagefare” (Ayalon, Popovich, and Yarchi 2016) to accompany traditional discourse based “propaganda wars”, “smart” NSAG leaderships will be receptive to ensuring that their “branding” is on point for local and international audiences (Abrahms 2018, 12). These different audiences are important to the NSAG in different ways for different reasons at different points in time; inasmuch as a NSAG needs operatives and material assistance during conflict, it also needs other actors and audiences prepared to hear its political message, prepared to enter into a political dialogue with it, and prepared to provide political cover as the organisation slowly guides itself away from political violence.

A critical examination of IRA statements over an extended period of time demonstrates the intimacy and complexity of the organisation’s relationship with the community from which it emerged. Reiterating again the point long made by others (Burton 1978; Sluka 1989; Morrison 2016a), the IRA did not, indeed could not, attract, and then retain, the communal support needed to sustain a 30-year campaign through crude intimidation. Rather, to borrow from Mao’s oft-cited analogy (Tse-Tung and Griffith 2005), the fact that the IRA invested considerable time and effort into consistently engaging with the “active” and “sympathetic” audiences through statements of denial evidences their acute awareness that the “fish” were at the mercy of the “sea” rather than the “sea” being subservient to the “fish”. Mirroring the often-localised nature of political violence (Sluka 1989), IRA denials of involvement in petty acts of localised anti-social behaviour were thus part of its “branding” as communal defenders to the “active” and “sympathetic” audiences. Yet, at the same time, as a “smart” NSAG leadership, the IRA leadership’s “branding” also involved denials of links with other groups and of involvement in organised crime that would resonate favourably with “uncommitted” and international audiences. Denials of transgressing the accepted moral code of a national liberation struggle by deliberately targeting civilians were, of course, cross-cutting (Abrahms 2018, 9). How successful or otherwise these denials were is difficult to discern, but the longevity of the IRA campaign suggests that at least the “active” and “sympathetic” audiences were assuaged by them.

Extending the temporal study of IRA statements also reveals the different purposes these statements served as the conflict progressed and then evolved into a protracted peace process. Remarkably, IRA statements that had in the early 1970s maintained that the armed campaign would continue unabated had by the late 1990s been recalibrated to reassure various audiences that it was over. Here, too, it becomes apparent that statements of denial were integral to a precarious balancing act of persuading external audiences of the republican commitment to the peace process while simultaneously reassuring the “active” and “sympathetic” audiences that there was no ideological “sell out”. The emergence of various post-GFA rivals willing to level such a charge (Sanders 2011; Horgan 2013; Morrison 2013) gave birth to a new “propaganda war” within post-GFA Irish republicanism over how to define the peace process (Hearty 2016).

Yet different analyses of the conflict and peace process have not been restricted to this new “propaganda war” between purists and pragmatists. They are now increasingly being played out in the post-conflict memoirs of former IRA activists. Those supportive of – if not involved in formulating – the Sinn Féin peace strategy have framed it as the natural evolution of armed struggle by other means in the face of a mutually hurting stalemate; but for more critical former IRA volunteers, this represents the culmination of years of the rank and file being deliberately misled by the leadership into accepting an outcome it had always strenuously denied seeking. For these critics, organisational-based denial in the latter stages of the conflict typifies the political manipulation of the “active” and “sympathetic” audiences by a Machiavellian leadership staring at military defeat while still clinging to its own political ambitions (Hopkins 2013). The IRA case study, then, has shown how, far from being unthinking statements to deny involvement in any given act, denials issued by NSAGs are the manifestation of complex and nuanced discursive strategies aimed at various audiences and fulfiling different functions across time and space.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. Following a split in the IRA in 1969/70, the breakaway faction became known as the Provisionals and would quickly become the dominant organisation within Irish republicanism. While several subsequent splits are briefly mentioned throughout this paper, an extensive overview of the emergence, evolution and fracturing of the Provisionals is beyond this paper. For fuller accounts see: English (2003); Sanders (2011); Morrison (2013) & White (2017).

2. Although initially separate Dublin and Belfast based papers they merged into a single An Phoblacht/Republican News publication in January 1979.

3. Language and labelling can be contentious in contexts of political conflict; this article therefore uses the terms the North of Ireland, Northern Ireland and the Six Counties interchangeably throughout.

4. Kathleen Feeney was accidentally hit when an IRA unit opened fire on a British Army patrol in Derry.

5. The bombing has since been shown to be the work of pro-state loyalist paramilitaries, possibly with the assistance of elements of the security forces.

6. In this instance, the denial appears to be accurate as the CAIN database on conflict-related deaths records McNamee as having been killed by a separate republican organisation.

7. The IRA would subsequently issue an apology to the victims of the attack, and to all non-combatants harmed by its campaign, on the 30^th anniversary of the attack.

8. For a fuller discussion of this in the Irish context, see: Maguire (1993) & Taylor & Horgan (1999).

9. Apparently, these members were the then Chief of Staff and the Belfast Brigade OC who were reportedly set to oust politically minded figures from the leadership before their plot was uncovered and averted.

10. For example, the Bush administration met with the McCartney sisters, rather than Gerry Adams, during the traditional White House St Patrick’s event in 2005.