Informers and the Transition in Northern Ireland

Though criminological literature has paid attention to the use of informers in ordinary law enforcement, there is a research gap regarding their usage in contexts of conflict and political violence. This article explores the social, political and security functions of IRA informers in the transition from conflict in Northern Ireland. Based on that experience, it develops four heuristic models regarding informers that the paper argues may be of direct relevance to other conflicted and transitional societies. These are the informer as folk devil, the informer as rumour, the informer as political manipulator, and the informer as celebrity. All these themes demonstrate the long-term effects of the use of informers during the Northern Ireland conflict — an important finding given the increasing prevalence of the use of informers in a political context.


Introduction
The use of informers by police and other state agencies in situations of violent political conflict would appear at first glance as a natural topic for criminological research. However, surprisingly little has been written on the issue. A decent-sized literature on the operation of informers in ordinary law enforcement has emerged in the last two decades (e.g. Rosenfeld et al. 2003;Cooper and Murphy 1997;Dunningham and Norris 1998;Glover 2001;Hirsch 2002;Clark 2006;Billingsley 2001;Crous 2009). However, the usage of informers in the context of politically motivated violence, even in such a heavily researched site as Northern Ireland, remains under-explored. This research, which aims to begin fill this gap, forms part of a wider agenda connecting criminology, conflict transformation and 'dealing with the past' in Northern Ireland and elsewhere (see, e.g. McEvoy and Newburn 2003;Mulcahy 2006;Eriksson 2009). 1 This article focuses in particular upon the long-term effects of the use of informers on a transitional society. Many questions and ethical dilemmas involved in the use of informers are essentially similar in ordinary law enforcement and in political conflicts-most notably, in relation to problematic recruitment methods and handlers' tolerance or encouragement of the commission of offences by informers. 2 However, I will argue that the long-term effects on society are much more prominent in cases of

Informing and the IRA: Theoretical and Historical Framework
In his work on the modern intelligence informer, Hewitt suggests a simple definition of informers: 'anyone who furtively supplies information to a state security agency' (Hewitt 2010: 18). Such definition is, however, too broad and can include a wide range of activities. A more useful approach is taken by Greer (1995b), who focused on providing a taxonomy of different types of informers. He suggested distinguishing types of informers according to two factors. The first is the type of their relationship with the activities and people about whom they inform: whether they are outside observers or insiders who have some involvement with the people or acts they report. The second factor is whether they provide a one-off tip to the police ('single event informers' in Greer's terms) or have a longer-term relationship with state agencies ('multiple-event informers'). Within these categories, the majority of informers used against the IRA had long-term, rather than one-off, relations with state agencies, and were 'insiders'-formal members of the organization, as well as people from the wider circle of active or passive supporters in their communities and networks. Their status as insiders also leads to another key definitional element: at least in the context of this article, informing is a form of betrayal, in that it involves a violation of a norm of loyalty existing in the informer's immediate community (Akerstorm 1991;Ben-Yehuda 2001). To sum up, for the purpose of this work, the term 'informing' refers to actions by individuals who cooperated with state agencies (including the police, army, secret security services and the judicial system) in the agencies' struggle against armed groups, with such cooperation being considered deviance by their communities.
In general, state agencies use informers against armed groups for a variety of reasons. Their most prominent function, naturally, is to gain information, especially that which technological means, and the agencies' own personnel, will not be able to gain: 'Being members of groups and communities which are under surveillance allows informers access to information not easily available by other means of intelligence gathering' (Cohen and Dudai 2005: 230). But, in addition, informers are also used as a 'proxy', to obfuscate responsibility of the state for certain types of operations (Jamieson and McEvoy 2005). More broadly, the use of informers spreads mistrust, internal doubts and paranoia in the ranks of armed groups and thus weakens and undermines them (Hewitt 2010). From the perspective of an armed group, the actions of informers constitute both an ideological betrayal of 'the struggle' and a practical sabotage of operations. Consequently, during conflicts, reprisals against real and alleged informers are common in most contexts where armed groups operate (Dudai and Cohen 2007).
It is important to note that informers have featured significantly in virtually all previous episodes of violent Irish republican uprisings before the beginning of 'the troubles' in 1969. As Elliot has argued, 'the word ''informer'' has resonated through two centuries of Irish nationalist and republican narratives' (Elliott 2006: 325) and the failure of, among others, the 1798 rebellion and the Fenian movement in the nineteenth century have often been attributed by republicans to the operation of informers (see, e.g. Knox 1997;Jenkins 2009: 334). During the Irish War of Independence, IRA leader Michael Collins has viewed the use of informers as a key counter-insurgency strategy and, in turn, has prioritized attacks on informers as a key revolutionary strategy (Dwyer 2005;Coogan 1990: 164). This history forms an important background, especially in regard to community perception of informers (a point that will be picked up again below).
After the outbreak of violence in 1969, all state agencies-the police, the army, the judicial system and the secret security services-were involved in recruiting and running republican informers. Initially, the adoption of a colonial counter-insurgency model (Faligot 1983) as well as inadequate police intelligence (Ellison and Smyth 2000: xvii) led the army to take the lead in recruiting informers and 'turning' IRA militants (Urban 1992;Dillon 1990). From the mid 1970s, the introduction of the closely linked policies of 'Criminalisation' and 'Ulsterisation' (Ellison and Smyth 2000: 80-6) led to the primacy of the local police rather than the army, and to the police taking the lead in recruiting and operating informers (Bamford 2005: 593). Also, between 1981 and 1985, the strategy of deploying 'supergrasses' (state witnesses) in mass trials against fellow republican, as well as loyalist, militants has become prominent (Greer 1995a). 5 In addition, as the nature of the military struggle between the British Army and the IRA changed towards the end of the 1970s from open street battles to the increased deployment of covert special forces such as the SAS , ambush operations were often closely tied to the operation of informers (Moloney 2010: 266). Informers also played a key role in the interception of arms shipments (Moloney 2007), through the offices of MI5 in particular (Andrew 2009: 737-45, 826). In short, throughout all phases of the conflict, informers provided details of personnel, weapons and forthcoming operations. They led to arrests, foiling of operations and discovery of secret arms dumps and shipments. They also interfered with weapons or vehicles in order to sabotage operations (Bishop and Mallie 1988: 401).
To counter these measures, the IRA developed a multi-faceted response to informing. From early on, it killed suspected informers (Dillon and Lehane 1973) and, in some cases, secretly buried their corpses, rendering them 'disappeared'. Around 1977, it established a special internal security unit to seek out, interrogate and execute informers. Around that period, in response to the threat of informers, the IRA also changed its entire paramilitary structure from battalion-based (mirroring the British army) to one based on small cells, again in order to diminish the damage an informer could create (Moloney 2007;Taylor 1997). This critical change coincided with the rise in prominence of a younger generation of IRA leadership-especially Gerry Adams and Martin McGuiness-who had promoted the view within republicanism that the organization had to plan for a 'long war' (Moloney 2007). As an internal IRA document from that time spelled out, 'we must gear ourselves towards a long term armed struggle based on putting unknown men and new recruits into a new structure' (cited in Coogan 2002: 466). Of course, it is possible that the threat from informers was also useful for those 'modernizers' to justify and legitimize changes that were being opposed by more traditional members. In either case, the widespread use of informers by the state was a catalyst for this crucial development of the IRA.
In addition, as part of its response to informing, the IRA offered periodic amnesties to informers who confessed to their activities, and also ordered hundreds of suspected informers to leave the country (as an alternative to execution). 6 At least in some cases, 5 While the open testimony in court makes supergrasses different from the usual informer who acts secretly, they could be seen as a sub-case of informer, as suggested by Greer (1995b). Some of the supergrasses have been acting as secret informers prior to testifying. Most importantly for the context of this article, the supergrasses have been labelled as informers by members of their communities (see, e.g. Boyd 1984;Sinn Fein 1983). the IRA held 'court martial' to its members suspected of informing. In addition to these acts, republicans also engaged in public campaigning, mainly through the pages of republican publications, demonizing informers, warning their supporters of recruitment attempts and showing positive examples of individuals who reported recruitment attempts or confessed that they had agreed to become informers. 7 The supergrass period saw a particularly intense mobilization against informers, and the informers who served as supergrasses were a very visible target for community anger (de Baroid 2000: 260). All these forms of 'counter-informing'-killing, threatening, exposing recruitment attempts, public demonizing-continued up to, and after, the 1997 IRA ceasefire and the 1998 Belfast Agreement-the key points in Northern Ireland's transition from conflict.
With this brief historical background in mind, I will now explore in greater detail the four themes regarding attitudes to IRA informers identified during the period of the transition from conflict in particular. This in no sense represents an exhaustive treatment of all aspects of informer-related discourses during that period. Rather, it represents my attempt to draw out heuristic illustrations of (sometimes overlapping) discourses on informing. As will be detailed below, the images of the informer range from an unforgiven folk devil, a rumour designed to slur political opponents, a cunning Svengali who covertly manipulates political processes and, finally, a media-friendly celebrity.

The Informer as Folk Devil
The first theme in public discourse on informers in transitional Northern Ireland explored here is of the informer as 'folk devil'. As Cohen (1972) had famously identified, folk devils are individuals who become the subject of intense moral concern by the community, the embodiment of a serious threat to collective social values and interests (Cohen 1972). 8 In this guise, informers are seen as the ultimate deviants, the most reviled figure, 'a despised creature' (Harnden 1999: 201). To be an informer in republican communities is to be the 'lowest of the low' (Smyth and Fay 2000: 27). The nomenclature is extremely loaded. As McKay has argued, the term 'informer' is the deepest insult in republican vocabulary (McKay 2009: 235). One prominent community activist and republican ex-combatant told the author 'I would rather be called a paedophile than an informer'. 9 The successful creation of folk devils 'rests on their stereo-typical portrayal as atypical actors against a backdrop that is over-typical' (Cohen 1972: 61) and on establishing the folk devil as 'a sort of alter-ego for Virtue' (Hall et al. 1978: 161). Indeed, in the republican case, the informer has been projected as the antithesis and the polar opposite of the venerated IRA 'rebel' (Hart 1998: 293;Bell 1991: 108).
The image of the informer as folk devil has sociological and political functions for anti-state armed groups. As Erikson (1966) famously argued in his study of witch-hunts, the quest to unearth hidden deviants can contribute to a group's solidarity, cohesion and boundary maintenance. Heresy is very useful in defending orthodoxy. More specifically, an imperfect but suggestive analogy could be made with the perception of 7 See, e.g. 'IRA threatens to kill informers ', Daily Telegraph, 19 October 1972;'Provos Renew Threats to ''Informers''', Irish Press, 14 August 1975;'Falls Man Targeted to Inform', An Phoblacht, 7 June 1990. 8 'The Folk Devil-on to whom all our most intense feelings about things go wrong, and all our fears about what might undermine our fragile securities are projected . . . when things threaten to disintegrate, the Folk Devil not only becomes the bearer of all our social anxieties, but we turn against him the full wrath of our indignation' (Hall et al. 1978: 161). 9 Interview with a republican ex-prisoner and community activist, Belfast, April 2009. informers by prisoners and street criminals, who share-through their conflicts with state authorities-some of the attributes of armed groups and the communities that support them. The hostility to informers by such groups is well documented: '. . . for street criminals . . . cooperating with the authorities is tantamount to treason' (Rosenfeld et al. 2003: 306). In the 'social code' of prisoners, 'the most inflexible directive . . . is concerned with the betrayal of a fellow captive to the institutional officials: Never rat on a con' (Sykes and Messinger 1970: 402, emphasis in the original). The function of this hostility is, to a large degree, the creation of solidarity (Brown 1993: 58;Marquart and Roebuck 1985: 218;Sykes and Messinger 1970: 407): '. . . pursuing snitches is seen as a manifestation of the common bonds of the prisoners or of the prison community in itself' (Akerstorm 1991: 98). Similarly, the image of the informer as folk devil serves an important function for paramilitaries and their supporters: the deviant whose rejection by the community demonstrates the importance of loyalty and solidarity. The identification of informers as folk devils also plays a more specific function within the armed groups' process of nation-building: the establishment and maintenance of an imagined community (Anderson 2006) to which loyalty should be owed (with the group in question claiming to be the legitimate authority to lead this community). The ability to define and punish betrayal is central to these ambitions: armed groups that seek to take over a state simultaneously claim the legitimacy to punish betrayal (Cohen 2008: 5;Fletcher 1993: 58). For emerging states, the power to punish treason is among the 'marks of sovereignty' (Orr 2002) and, for armed groups, which seek to assert the real or symbolic authority of a state, such a method of asserting sovereignty is particularly appealing. The construction of the informer as folk devil is itself, then, a mark of sovereignty and thus plays an important function for the armed group. The modalities in which this power was asserted by the IRA-the use of 'court martial', the periodic offer of amnesties-also bore the hallmarks of a quasi-state institution, a variant of what James C. Scott (1998) has referred to as 'seeing like a state'. The function of the informer as a marker of a group's boundaries and loyalties remained important-perhaps even intensified-in the peace-process years. This is because mainstream republicanism still struggles to maintain loyalty and cohesion among its supporters, even if, this time, it is in the realm of electoral political rather than the military sphere. During the transition, retaining the informer as a constant enemy helps to maintain continuity-a particularly useful feature given the inevitable uncertainty of a peace process, and the ongoing questioning of the leadership and their strategy.
Indeed, the important point to this article's argument is that the hostility to informers has not just been a feature of the conflict period-when the operation of informers posed a concrete direct threat to the IRA's armed campaign-but has continued almost unabated well into the peace-process years. A key insight is that informers seem to have remained perhaps the last 'unforgiven' category of actors in the landscape of post-conflict Northern Ireland. This is manifested in several ways.
The first is the absence of informers from a discourse and practice of reconciliation, dialogue and political acceptance between former protagonists of the conflict. Since 1998, numerous high-profile dialogue or reconciliation meetings and initiatives have taken place in Northern Ireland involving republicans, loyalists and security forces members, as well as victims of violence by these different actors. Such activities included, for example, public events and tours, TV programmes and conferences organized by NGOs (see McKay 2009: 225-54 for an overview). To illustrate, meetings between survivors and perpetrators included, for example, joint public appearances of a Derry Catholic who was blinded by a rubber bullet and the soldier who shot him; 10 of Patrick Magee, who was responsible for the 1984 Brighton bombing, and the daughter of one of those killed in the bombing; 11 and of loyalist paramilitary Michael Stone and the widow and brother of a man he is convicted of murdering. 12 In addition, there have been several joint projects between republican and loyalist ex-combatants. 13 There were also various forms of meetings and joint work by former conflict protagonists in the political arena-most memorably, the amicable work relationship of republican leader Martin McGuiness and the leader of extreme unionism, Ian Paisley. However, what is very interesting for current purposes is the complete lack of engagement with informers in such meetings, initiatives or projects. In other words, while at least many in the affected communities were willing to hold some form of normalized contact with the groups with whom they fought for decades, such willingness to 'move on' was not extended to informers. Though 'the question of meeting the enemy' has become prominent in transitional Northern Ireland (McKay 2009: 231), meeting those who were considered the 'enemy within' has been conspicuously absent. Informing remains beyond the pale even years after the events, and even while members of security forces-those who operated the informers-are accepted. As one republican activist told the author: I had discussions with loyalists, I work with loyalist ex-prisoners, I meet members of the RUC, we've met former British soldiers. Loyalists regularly would be in West Belfast and they feel free to come and go. But I don't think an [ex] informer could do that. 14 Another activist similarly acknowledged: We can have ex-soldiers in this office, and we've had, and I'm involved in roundtables with police officers, whatever, that's all possible-but the supergrasses, I just can't see them returning to this community here. Just can't see it. 15 One of the most concrete manifestations of the 'unforgiven' theme is the continuing problem of the 'exiles'. During the conflict, hundreds of suspected informers were ordered by the IRA to leave Northern Ireland (Coogan 2002: 519;Morrison 2001: 67; Consultative Group on the Past 2009: 80). In addition, many informers were spirited out by their handlers, and state witnesses were re-settled elsewhere as part of their deal with prosecutors. By and large, no organized solution has been found for these people, and there has been no mechanism to facilitate a safe return to the community of those exiled (Consultative Group on the Past 2009: 138). In one instructive example, in 2007, Raymond Gilmour, one of the prominent ex-informers, made a public appeal requesting to come back to his native Derry. He sought assurances from republican leader and Deputy  won plaudits across the political spectrum in Northern Ireland for his commitment to the peace process, was decidedly frosty. He would only say that Gilmour must decide for himself whether or not it was safe to return to Derry. 16 Put in more concrete terms, while McGuinness was ready to shake hands amicably with Ian Paisley, a bitter enemy for decades, and to provide broader leadership to his constituency regarding other former enemies (McEvoy and Shirlow 2009), he was not yet ready even to support the return to the country of informers such as Gilmour. Another important manifestation of this attitude is the continuing ostracism of the families of alleged informers (see, e.g. Ardoyne Commemoration Project 2002; Smyth and Fay 2000: 26-8;McKay 2009: 174). The relatives of informers are often shunned by many in their communities, as if they are contaminated by association with the informer/folk devil. This is a form of what Goffman described as 'courtesy stigma' or stigma by association, where stigmas not only affect the individuals bearing them, but also those who are in close association with these individuals (Goffman 1963: 30). As Toolis described, 'the waters cleave and the life of the informer, and their kith and kin, diverges from the tribe' (Toolis 1995: 194-5). Finally, although relatively sporadic, it is also important to note that the IRA seems to have continued to plan reprisals against informers even after the halting of violence against security forces. 17 Some attacks-such as the attempted killing of Martin McGartland in 1999, the killing of Eamon Collins in the same year and the killing of Denis Donaldson in 2006 18 -were against already-exposed informers and seem to be 'settling accounts' of historical cases. Significantly, during the same period, there were no similar reprisals and 'settling of accounts' against individual security forces members who were involved in the past conflict.
How to account for the strength and durability of the hostility to informers? Hatred of those considered to be 'traitors' by a community has been identified as a trait across cultures (Ben-Yehuda 2001) and is especially strong in relation to those who are considered to have betrayed national armed struggles (Dudai and Cohen 2007). The continued hostility towards those seen as traitors in the aftermath of a conflict is, in itself, not unique to Northern Ireland: the resentment of perceived traitors, even during reconciliation with the direct enemy, has been a feature of many other contexts, from South Africa after apartheid (Cherry 2000: 140), to Algeria after independence from France (Evans 2002). Nils Christie has described how in Norway the hatred of Norwegians who served as informers and collaborators with the Nazi occupiers persists decades after the events. For example, when, in 2002, Norwegians invited Second World War veteran soldiers from all countries, including Germany, to a reconciliation 16 'IRA ''Supergrass'' Wants to Return', BBC website, 2 February 2007. Others have been more explicit, e.g. a member of the Irish Republican Socialist Party, who said he believed the informer would not find dissident republican groups in a forgiving mood. Graffiti has also appeared in the Bogside declaring: 'Gilmour you dare come back!'-see C. Weir, 'IRSP Warns Gilmour over Derry Return', Belfast Telegraph, 6 February 2007. 17 The Independent Monitoring Commission (IMC) has found that, while the IRA has stopped planning attacks on members of the security forces, 'PIRA does continue to investigate suspected informers or others in its own ranks' (IMC, Twelve Report, October 2006, para. 2.18. Similar assessments were also made in subsequent IMC reports). It should be noted that the IMC remains a controversial body in Northern Ireland, accused of bias and of not presenting sources of evidence for its findings. However, I am not aware of a specific challenge relating to the IMC's finding on informers. 18 All of these attacks are reliably believed to be carried out by republicans, although it remains unclear whether they were formally ordered by the leadership of any group. No one was convicted for any of these attacks. In 2009, the dissident splinter group, the Real IRA, claimed responsibility for the killing of Donaldson over three years earlier (see S. Breen, 'How Real IRA Killed Denis Donaldson', Sunday Tribune, 12 April 2009), although the veracity of this claim is hard to assess.

INFORMERS AND THE TRANSITION IN NORTHERN IRELAND
meeting in Norway, Norwegian collaborators were the only category excluded from the invitation (Christie 2004: 248).
That said, I would argue that there are several factors that have likely contributed to the particular intensity of attitudes to informers in the Irish republican context. As noted above, the historical importance, across two centuries, 19 of using informers as a successful counter-insurgency strategy against republicanism, and of the informer as a hate-figure, is directly relevant. It serves to amplify the importance of contemporary informers and legitimize reprisals against them. Another important factor is the localized and close-knit nature of republican groups and their supporting communities, often based on extended families and neighbourhood networks (Taylor 1997: 256). In such a setting, informing becomes a particularly 'intimate betrayal' and the negative status of informers intensifies. 20 The practice of recruiting informers from among criminals or using blackmail based on martial infidelity or other behaviours considered deviant (Bamford 2005: 591) has also probably contributed to the construction of the informer as folk devil. It has arguably created an association and a blurring of boundaries between criminals, those who engage in anti-social behaviour and informers, with various forms of deviancy mixed together to create an enemy of the community. This image of informers as low-level criminals also often combines the despicable with the pathetic, such as with the common term 'ten pound touts', 21 resulting in a portrayal of agents far removed from any James Bond-like association and contributing further to the undiluted hostility to informers. I would suggest that an amalgam of these factors resulted in a strong mix with lasting effects.
The implications to Northern Ireland, however, might be more immediate than in a state like Norway. With sectarianism, and even sporadic political violence, still a feature of the Northern Ireland society, leaving informing as an open wound from the past conflict means that the process of conflict transformation cannot be completed. The exiles and the informers' families remain a reminder of an unfinished process. To retain informers as an unforgiven category, while interacting cordially with other types of ex-combatants, victims and former political enemies, including those responsible for abuses, can also send an implicit symbolic message. It implies that betrayal of a community is inherently and categorically worse than other violent acts, suggesting that unconditional attachment to one's community should always trump other attachments. Attempts to maintain community boundaries even after transition out of conflict could be seen as part of what Davis (2009) described as 'new imagined communities': the dynamic of seeking allegiances and loyalty by non-state actors even where such actors no longer seek to take over a state. The hostility towards those who acted as informers during the conflict-hostility that is premised on the notion of loyalty to a community rather than to the state-is a useful symbolic asset for those seeking allegiance, all the more so when the political situation is such that overt hostility to the state and its institutions (including the current police force) is no longer a legitimate option for mainstream republicanism. 19 'In Republican communities an atavistic hatred of informers runs back to the Dublin Castle spies who betrayed the 1798 Uprising' (Toolis 1995: 194). 20 The general literature on betrayal finds that the intensity of disapproval of betrayal rises in correlation to the close nature of the betrayed in-group (Akerstorm 1991;Ben-Yehuda 2001). 21 That is, individuals who receive only a meagre sum of money in exchange for their betrayal.

The Informer as Rumour
A second important theme is the contemporary use of the informer tag as a rumour, slur or 'accusatory practice' (Fitzpatrick and Gellately 1997) in public life. Years after political violence has been diminished, the practice of anonymously 'outing' alleged informers has become a common refrain, a staple of political, social and cultural life in Northern Ireland. 22 In the majority of cases, such allegations were not supported by transparent evidence, and the study of informer-naming in contemporary Northern Ireland is thus, to a large degree, a study of rumours. Rumours were defined in Knapp's seminal work as a particular form of information, 'a proposition for belief of topical reference disseminated without official verification' (Knapp 1944: 22). Thus, rumours are defined by their mode of production, not by their veracity as such: rumours may or may not have an accurate factual element. Scholars of political violence and its aftermath have often noted the importance of rumours (Das 1998;Guha 1983). Rumours both reflect and reproduce political violence and conflict, and are an important vehicle through which protagonists in a conflict interact (Kirsch 2002: 57). Rumours are, in themselves, an important source for understanding political conflict and violence, especially in times of uncertainties and political transitions (Fujii 2010). Two informer allegations from the last decade have been particularly high-profile. In the first case, from 2003, that was because the person in question-Freddie Scappaticci, a veteran IRA member, who escaped out of the country after being named in the media as an informer-was a key figure of the IRA's internal security unit. In other words, the dramatic claim has been that a key person within the IRA tasked with rooting out informers was himself an informer, actively involved in interrogating and executing wrongly accused 'loyal' IRA members. 23 The second prominent revelation was that Denis Donaldson, a senior Sinn Fein official with a long past in the IRA, was an informer. In this case, the chain of events leading to his exposure was remarkable: he was arrested by the police and charged with operating a Sinn Fein 'spy ring' in the Northern Ireland assembly, where republicans were allegedly spying on their political opponents. This caused a collapse of the power-sharing assembly, the key feature of the Belfast Agreement. Later, the charges against him were dropped without explanation. Following that, he was exposed as a British agent. He confessed in a televised press conference that he was an informer, and left public life to live in a secluded house in a remote part of Ireland, were he was shot and killed a few months afterwards (Moloney 2007: 580). Republicans have repeatedly alleged that the collapse of this government on the basis of a supposed Sinn Fein spy-ring that turned out to have a British agent at its centre amounted to a coup d'état instigated by elements of the British security establishment who were opposed to the peace process.
But, in addition to the Donaldson case, where the evidence was indisputable, and the Scappaittici case, where there was strong circumstantial evidence, allegations have also been made against a string of prominent republicans, in most cases without credible evidence. Some of the cases of allegations of informing made since 2003 include those against Joe Haughey, a former IRA intelligence officer and a well-known republican 22 S. Breen, 'Nobody Knows Who to Trust', Sunday Tribune, 9 January 2006; Moloney 2007: 580. 23 The claim has been the Scappattici has been a military agent code-named Stakeknife. See, e.g. R. Cowan, 'He did the IRA's dirty work for 25 years-and was paid £80,000 a year by the government', The Guardian, In a marked minority of cases, the people named have confessed to being informers, but in the majority of these cases, no credible evidence has been presented and it is therefore difficult to assess the veracity of the allegations. Indeed, I make no claims in this article as to whether or not any individual is or was an informer. What is more important from my perspective is the ways in which the term is deployed across the spectrum for apparent political and ideological reasons. Two questions are of particular interest to this article: what functions the prominence of informer labelling plays (what are the political properties of the rumours) and what factors facilitate the potency and spread of the allegations.
The first, and fairly obvious, function is the discrediting of the individuals named. In addition, there is a form of what can be termed indirect discrediting, where the aim of the rumour is not so much the alleged informers themselves, but the armed groups and political movements associated with them. This can have several objects: one is the discrediting of leaders who, according the narrative, were proven incompetent by the alleged uninterrupted operation of informers close to them. 32 A second object can be an attempt to discredit the entire nature of the movement and its ideology. 33 In either formulation, such claims correspond to Knapp's category of a 'wedge-driving or aggression rumour', aimed to 'divide groups and undermine loyalties' (Knapp 1944: 24). Finally, another format and set of motivations around the public discourse on 'outing' of informers is the way former security forces personnel are using it to threaten further exposure unless some demands of theirs are met. 34 Thus, the imputed knowledge about the identity of secret informers becomes an 'informational advantage' (Mui 1999)-a resource used in order to demand other goods. 35 Beyond these functions, the question remains as to what factors facilitate the potency and effectiveness of rumours and threats about informer naming. One important factor is the virtual inability of the target of an allegation to refute it. Not unlike witchcraft accusations, once someone is alleged to be an informer, it is almost impossible to shake off the accusation and any action could be interpreted as 'confirming' suspicion. 36 This is further facilitated by the state authorities' refusal to confirm or deny that individuals have been informers 37 and the lack of any public access to state files on this issue. Both these factors are a result of the particular nature of transition in Northern Ireland, which-unlike typical transitions-did not involve a complete break with the 'old regime' (Campbell et al. 2003): the security services that operated during the conflict still maintain control over the relevant information.
Another important factor is the central role of the media in facilitating such rumours. The speculative nature of reporting on the issue shares characteristics with the way the Northern Irish media handled the issue of sex crimes (Greer 2003). One prominent element in media reporting on informer allegations has been uncritical reliance on information from unnamed security sources. 38 Another has been a pattern of predicting more revelations to come when reporting on a particular case 39 -a feature that amplifies the resonance of the stories. Often, newspapers simply reported the existence of rumours, thus lending them credibility. 40 In one emblematic example, the Sunday Times published allegations that Martin McGuinness was an informer. Such a dramatic claim-which, if true, would change the entire record of modern Irish history-was published without any transparent source to support it. The story was based on unnamed sources, both from ex-security personnel and from 'republican veterans', on interviewing ex-informer Raymond Gilmour and on insinuations such as pointing out that McGuinness has 'never been shot or injured nor served a serious prison sentence in the UK' as an implied indication that he was an informer. 41 While stressing again that I do not pertain to know whether or not any individual was an informer, the important point is that, with no senior republicans or security services personnel speaking publicly and openly on the issue, and little documentary evidence being presented, almost the entire public record on this crucial issue is based, therefore, on rumours. 42 To sum up, informer naming provides an easy and powerful tool in political life in contemporary Northern Ireland: a readily available vocabulary and a clear template to discredit opponents. The combination of the hostility to informers, the limited capacity of the accused to fight it back and the public receptiveness to such claims makes such 36 Joe Cahill, a former IRA chief of staff, described how, once one IRA member was suspected as being an informer, a next step was to ask 'if the informer appeared to be very conscientious; if everything he did appeared to be done well', since that is 'trait of an informer' (Anderson 2002: 143). The fact that the suspect was later arrested by British authorities seems only to confirm the suspicion (Anderson 2002: 144). 37 See 'In the Matter of an Application by Freddie Scappaticci for Judicial Review', CARC3995, 18.08.2003, Neutral Citation no. (2003 NIQB 56. In this case, Freddie Scappaticci, mentioned above, appealed to the government to deny that he was an informer, claiming that his life was in danger. The government maintained its position neither to confirm nor deny that any individual was an informer, and the court accepted the state's position. Some of these rumours and allegations have been relatively ephemeral, while others have been more enduring. It will be difficult to quantify and explain the acceptance level of each allegation and rumour, but the key to the argument presented here is the accumulated effects of the different rumours on political and social life.

INFORMERS AND THE TRANSITION IN NORTHERN IRELAND
accusations a particularly strong currency. As one former republican prisoner told the author, 'if you want to discredit someone you imply that they were an informer in the past and their names are ruined, because how do those people disprove that they weren't informers'. 43 It is beyond the scope and ambition of this article to determine the exact motivations and actors behind all these rumours, although it is interesting that many (although not all) of the targets have been from the pro-peace process camp of republicanism. Thus, those ultimately benefiting from this discourse are mostly those who resent or oppose this camp. In any case, it is clear that the prevalence of this 'accusatory practice' is, on the whole, a destabilizing factor in the attempt to build healthy political structures and processes in the transition out of the conflict. This represents an indisputable negative consequence of the pervasive use of informers during the conflict and the suspicion, mistrust, resentment and secrecy it has generated.

The Informer as Political Manipulator
Any historian of warfare knows it is in good part a comedy of errors and a museum of incompetence; but if for every error and every act of incompetence one can substitute an act of treason, many points of fascinating interpretation are open to the paranoid imagination. (Hofstadter 1965: 24-5) Military setbacks often lead to a search for 'internal enemies', which serve as a convenient explanation for defeats (Keen 2008: 63). One of the most prominent examples is the 'legend of the ''stab in the back''' popularized in Germany in the aftermath of the First World War, in which the defeat in the war was explained not by the weakness of the army and leadership's mistakes, but by the secret treasonous work of German Jews-'the enemy within'-undermining the war effort (Bartov 2000: 95-6). The best description of such tendencies remains Hofstadter's (1965) oft-cited analysis of the 'paranoid style' in American politics. This style culminated in the McCarthy era, in which the work of secret traitors and the infiltration of hidden communist agents to high level of government-the 'betrayal from high'-was put forward as an explanation for American policy failures. In Hofstadter's analysis, the term 'paranoid' is not used in a clinical sense, but rather to denote a political tendency, a mode of expression. It is common for members and supporters of rebel armed groups to employ such a style and explain failures to achieve the group's goals by the activity of secret traitors, rather than by strategic and tactical mistakes or by broad structural changes that make the group's ideology no longer relevant. Thus, one of the main reasons for a fixation with real or imagined informers among rebel groups is that it gives an opposition armed movement: . . . an excuse for its defeat or for its failure to win, through putting the blame on 'traitors', 'fifth columns' and so on . . .. Blaming collaborators helps an opposition movement maintain its image and power even when it is failing to reach its goals. (Dudai and Cohen 2007: 43) This general trend has featured in the Irish republican case as well, where it has been a common refrain to explain defeats of past and present by the work of informers (Bean 2007: 180). As noted above, in Republican communities, 'the informer is . . . a convenient scapegoat for centuries of glorious failure' (Toolis 1995: 194; and see also 43 Interview with a republican ex-prisoner and community activist, Belfast, April 2009. Morrison 2001: 66). Yet, the practice of explaining failures by the secret work of informers acquired a new particular character and resonance in the years after the IRA's ceasefire, on the background of the embrace by the republican mainstream-led by Gerry Adams and Martin McGuiness-of the peace process and its gradual rejection of the 'armed struggle'. A new theme emerging since the peace process has matured is that informers are to be blamed not only for the foiling of military operations, but for directly shifting the political strategy of the republican movement into accepting the peace process. In this discourse, the informer is a cunning political manipulator, a Svengali who secretly exerts power and influence, and shapes historical events. Thus, there have been claims that informers 'helped sideline opponents of the peace process in favour of others more favourably disposed towards the Adams leadership' 44 and that 'Over the years, MI5 used Donaldson to strengthen the Adams-McGuinness leadership and to weaken its critics'. 45 Anthony McIntyre, a former IRA prisoner and strident critic of the Adams/McGuiness leadership, has described informers as 'touting for peace' and argued that agents have been central to the British state attempts to 'nudge [the IRA] towards a peace process' (McIntyre 2008: 188). The 'ready accommodation with Ian Paisley's DUP at Stormont' was at least implicitly linked to 'The stream of informers being exposed in high places in the republican movement' in an account of an IRA ex-combatant (Bradley and Feeney 2009: 8-9). In another account by a former IRA leader, concerns and mistrust about the republican leadership's new peaceful direction were intermingled with claims about informers and their toleration by the leadership (Moloney 2010: 276, 290).
While certainly informers have been operating in republican ranks (and communist agents did operate in 1950s America), what is distinctive about this mode of political expression is the weight placed on the work of informers, their view as omnipotent and the presentation of their actions as the most important, even exclusive, reason for the republican political change. 46 As Hofstadter describes, an important part of the 'paranoid style' is that the interpretation of history is 'distinctly personal': 'decisive events are not taken as part of the stream of history, but as the consequences of someone's will' (Hofstadter 1965: 32). Thus, the critical role ascribed to individual informers in changing the course of the IRA, 47 ignoring the wide range of structural and contingent developments leading to the IRA's termination of its 'armed struggle' (ranging from the 9/11 attacks to European integration: see O'Leary 2005), is an example of such a mode of thinking. 48 On one level, this mode of representing informers might be a coping mechanism, solving a cognitive dissonance of those activists who cannot come to terms with the leadership's political changes. Subscribing to a view that the republican leadership's embrace of the peace process was a result of manipulation by informers helps explain what would otherwise be an inexplicable change of heart. On a more political level,  Table', Irish Echo, 19-25 August 2009. 48 In turn, it is important to note that pro-peace-process republicans have ascribed a critical role for informers in the intensification of operation by dissident groups. It has been often suggested that dissident groups are heavily infiltrated by British security agencies, and that informers are behind the mobilization and operation of these groups (acting to realize, according to this rationale, the British aim of re-militarization of Northern Ireland). See, e.g. 'The Militarist Campaign: ''Who's Pulling the Strings?'' Asks Ex-POW Bobby Storey', An Phoblacht, 28 June 2010. the dissemination of such views, if seen as credible even by small segments of the community, obviously undermines the credibility of the republican leadership's peace policy. It renders the moves towards peace as deprived of real republican agency-a passive tool in the hand of hidden conspirational forces. On either level, the combination of the secrecy and of the popular negative perceptions of informers makes informing a useful resource easily tapped for excuses and explanations. The potential of this mode of expression to erode the public legitimacy of the peace process is another problematic longterm consequence of the pervasive use of informers during the conflict. 49 The Informer as Celebrity Finally, the last theme identified here stands in contrast to the others: this is the informer as celebrity, or respected pundit.
While retaining the pariah status in republican communities, for the media and for sections of society that oppose an armed group's goals, the figure of an informer who proudly shares his experience can have a special appeal. Members of armed groups who turn informers and later publicly denounce these groups combine several features that make them attractive to the media and to those who oppose these groups. First, they provide rare insider knowledge on unlawful behaviour and an associated 'celebrity of infamy' (Oleson 2003). Their involvement with both armed groups and security forces can make them appear as experts on the issues-'authorized knowers' (Ericson et al. 1989) whom the media and others can rely on. Second, a shift from militant to public critic of paramilitary violence provides a compelling story of conversion, combining the appeal of 'redemption narratives' (Maruna and Ramsden 2004) of ex-addicts or ex-criminals, with the importance attached to the 'figure of the renegade from the enemy cause', apparent for example in the prominent use of ex-communists in anti-communist campaigns, or of ex-Catholics in anti-Catholic movements (Hofstadter 1965: 34-5). Finally, the exploits of armed groups' informers simply make good subjects for the press and publishing houses, much like informers such as Joe Valachi or Vinnie Teresa, in the context of organized crime, made 'good copy' for the press (Potter 1994: 31). Like the other themes identified here, this theme has contributed to the prominence and visibility of the issue of informers in transitional Northern Ireland.
The aftermath of the IRA ceasefires saw the emergence into public view of several exinformers from republican ranks, who spoke unashamedly about their past and attained high media and public profile. This has been a development of the peace process years, and was not evident at the height of the conflict. This new development also involved the emergence of a literary genre that may be unique to Northern Ireland and can be termed the 'informer memoir'. No fewer than five informers from the ranks of republican groups have published their memoirs-a trend that has no parallel in other analogous situations such as the Israeli-Palestinian or South African conflicts. They include Eamon Collins (1997), Martin McGartland (1997, Sean O'Callaghan (1998), Raymond Gilmour (1998) and Kevin Fulton (2006). Their high public profile was also a result of threats and violence against some of them, culminating in the killing of Collins in 1999; McGartland survived an assassination attack in the same year. 50 Some of these ex-informers also acquired a respectable public role. Sean O'Callaghan, a prominent former IRA informer, worked as a consultant to the Ulster Unionist Party (UUP) and was often interviewed as a media pundit, and wrote op-eds for British newspapers. 51 A documentary about him was screened by the BBC. 52 In another example, McGartland's first book-Fifty Dead Man, a reference to the number of people he claims to have saved from death through his operation as an agent-was adapted into a Hollywood-style movie starring Ben Kingsley, which was on general release in 2008 (shown in, among others, Belfast). McGartland himself was also the subject of a BBC documentary. 53 Collins was cited as an honourable example of 'leaving terrorism behind' (Bjørgo and Horgan 2009: 22) and O'Callaghan's account of changing his heart and deciding to become an agent following indiscriminate killings by the IRA was also commended (e.g. Mulholland 2007: 410). A typical framing was given in a by-line to an article by O'Callaghan published in the National Review: 'A former terrorist describes his life in the IRA, and looks at the current peace negotiations through the prism of what he learned in his old life.' 54 Another emerging trend in Northern Ireland, the publication of memoirs by former police officers who were involved in the conflict, also included often flattering portrayals of informers; one of these memoirs, for example, was dedicated to 'Raymond Gilmour and all the other agents' (Barker 2004: 10).
It is important to remember that the articulated and well-connected memoir-writing informers are not necessarily representative of the hundreds of other informers. Their accounts are often characterized by an over-privileging of their heroic agency 55 and do not, for example, include references to the reality of coercive recruitment that has been a feature of most informer careers. Rather than creating a more nuanced perception of informers in republican communities, the media profile of these informers has probably further cemented the hostility towards them as a whole: the informer as celebrity, singing from the same hymn sheet as the republican political opponents, thus reinforcing rather than undermining the informer as folk devil. 56 The same features that make these informers' tales attractive to the media-the revealing of insider knowledge and the authority of redemption narratives-entrench the image of the informer as a pariah in the communities from which they originally came.

Conclusion: Informers, International Human Rights and Dealing with the Past
This article has identified the long-term effects of the use of informers during the Northern Ireland conflict, evident long after the beginning of the peace process. During the conflict, the recruitment and operation of informers have been key elements of almost all types of encounter between state agencies and republican communities-50 Thus, becoming a public figure cannot be seen as a survival strategy on behalf of these individuals. As such, these memoirs have also played into the narrative of the informer as an extremely influential manipulator, as described in the previous section. 56 E.g., Danny Morrison wrote about 'the celebrity touts, those twice-rewarded Judases whose kiss-and-kill was quickly followed by kiss-and-tell'. See 'Exiles Cannot Expect a Welcome Home', available online at http://sluggerotoole.com/2006/01/18/exiles_can-not_expect_a_welcome_home/. arrests, interrogations, trials, military operations, policing of ordinary crimes-and dealing with the risk of informers permeated many facets of republican activities. The result, an informer-related 'culture of suspicion' (cf. Cohen and Dudai 2005), was carried over from the conflict years to the peace-process phase, complicating conflict transformation. The secrecy surrounding informers, the intense hostility towards them in their communities and a general 'fascination of betrayal' (Simmel 1950: 333) made them a useful target for diverse actors with diverse motives, as shown above. The phenomenon of informing has been used to discredit and undermine the political shift from violence and suspected informers and their families remain unreconciled with their communities-a symbolic reminder of the durability of conflict-era stances and a lingering open wound.
The themes discussed here-intense resentment and social ostracism, the central role of rumours, the importance of the search for hidden conspirational explanations-do not lend themselves easily to interventions anchored in international human rights. While human rights advocates favour fact-based, legalistic methods that reach clear and precise findings (Dudai 2006)-see, for example, their positive reception of the report of the Saville Inquiry on the events of Bloody Sunday (Bell 2010)-such methods cannot adequately deal with the ambiguity and murkiness involved in informing and its effects. Indeed, the issue of informing also exposes several 'blind spots' in the discourse and practice of human rights and 'transitional justice' (Cohen 1995;Teitel 2000), which has become the dominant framework and lens for dealing with the past in post-conflict societies such as Northern Ireland (Bell 2003).
While, as was demonstrated above, informing has permeated many aspects of political and social life in transitional Northern Ireland, with overall negative implications for the transition from violence, the phenomenon and its effects have not featured in most discussions of transitional justice in Northern Ireland (see, e.g. Duffy 2010; Aiken 2010). One reason is that, because of the universal tenets of international human rights, human-rights advocates are not sensitive to the importance attached by many to the perceived betrayal of communities. The fact that such acts have not been a violation of international law, but of community trust, means that human rights advocates may underestimate their importance during political violence and its aftermath. 57 Another blind spot results from the tendency among human rights and transitional justice scholars and advocates to think in binary categories in which individuals are either victims or perpetrators of human rights violations, while overlooking more complex identities, such as informers (Borer 2003). Yet another blind spot is a result of the exclusion of informers from civil society projects, as described above. The enthusiasm of many transitional justice scholars for initiatives dealing with the past established by civil society groups and for 'transitional justice from below' (McEvoy and McGregor 2008) has perhaps obscured the fact that civil society groups can retain patterns of exclusion. Lastly, as the human-rights framework has followed criminal justice in de-personalizing responses to violence (Christie 1977), it has overlooked the importance that personal emotions and perspectives could have in political violence and policies aimed to respond to the legacy of such violence. Informing is one major example of the lingering importance of 57 A related point was made by Fletcher, who compared the offence of treason, which he described as protecting 'parochial' interests, to the 'universal' offences in international criminal law, such as crimes against humanity and war crimes, which protect universal interests (Fletcher 2005). the personal in political conflict and political reconciliation. As one ex-combatant described to the author: You think back: you stayed in their house, you had them in yours, you maybe went to holiday with them. At that time you thought you know them, but you didn't, they led a double life. And all that sense of betrayal is personal. It's not just organizational. 58 The importance of these issues resonates beyond Northern Ireland itself. The use of informers, in various forms, is a key element of responses by Western states to the contemporary real and alleged threat from Islamist terrorism (Hewitt 2010: 121-46). 59 Therefore, appreciating the right lessons from the Northern Ireland experience should become crucial. For years, techniques of surveillance and control were 'tested' in Northern Ireland before being normalized in a wider British context (Hillyard and Percy-Smith 1988) and, in recent years, policing models from Northern Ireland have been presented as 'best practice' and recommended or transferred to other settings of political violence, including Iraq (Ellison and O'Reilly 2008) or Afghanistan (English and Johnson 2008: 280). There is often even a direct move of intelligence units and personnel from Northern Ireland to Great Britain and abroad. 60 A recognition of the long-term effects informing has on communities, rather than narrowly seeing it only as a security tool, should become part of policy design and of public debate on such policies. Drawing attention to such effects can be an important contribution of criminology to conflict transformation. While exact policy recommendations are beyond the remit of this article, the analysis suggests that responding to the legacy of informing will necessitate interventions that move beyond an exclusive human rights lens, to involve techniques of restorative justice and extra-legal approaches (see, more generally, McEvoy 2003; for dealing with the past.

Funding
The research was assisted by a grant from the Stephen Livingstone Memorial Fund.