Morality in Intelligence Practice: Lessons from the British Experience in Northern Ireland

Photo by Hayes MKII/Flickr |

By Natasia Kalajdziovski | 

The role of intelligence in successful counterterrorism efforts is undoubtedly critical. That role, however, can become complicated when counterterrorism efforts focus entirely on domestic terrorist groups. Questions of morality inevitably arise when liberal democratic states direct counterterrorism efforts against their own citizens; these questions stem from a state’s responsibility to protect both individual citizens and society as a whole. If one understands morality as the principles that help us distinguish between right and wrong, do such concepts have a place within the conduct of intelligence collection by a democratic state within its own borders?

At the heart of this intersection between moral conduct and security is the question of proportionality. If the security of its citizens is of paramount importance to any good government, that government’s ability to maintain security at a “bearable and sustainable cost” is critical.[1] Stansfield Turner, former Director of Central Intelligence, has posed the above conundrum in simple terms: the morality of intelligence conduct is a test of sound will, “that is, whether those approving [decisions] feel they could defend their decisions before the public if their actions became public.”^[2] Former director of Government Communications Headquarters (GCHQ) Sir David Omand, however, has framed the issue as one of “sets” of morality; that is, the conduct of intelligence “requires extraordinary methods, involving a different morality than that of everyday life.”^[3] While neither presents intelligence within a moral vacuum, what Omand suggests – this “different morality” – is key. He identifies the question of morality in intelligence practice not simply as an issue of public operational viability, but rather says that the “rules we hope govern our private conduct as individuals in society cannot fully apply.”[4] However, the crux of this conundrum for intelligence services and those who guide their policy and practice resides in distinguishing and defining where moral parameters must exist.

For the British security establishment in Northern Ireland throughout the Troubles, the issues in defining moral parameters in intelligence practice became acute. As their experience demonstrates, sometimes those parameters can be lost, muddled, or forgotten in the height of crisis, and definitions can be altered or changed as a conflict protracts. In an effort to uncover some of the moral questions faced by the British intelligence services throughout the Troubles, this article will chronologically investigate how these questions manifested and modified as the conflict unfolded.[5] It will examine three areas: initial intervention, service engagement with the “long war,” and the sophistication of the intelligence apparatus in the latter days of the conflict. This article will not attempt to judge the morality of the decisions made by the services, but rather demonstrate how the ambiguity between moral codes and effective security is inherently difficult to navigate.

Intervention and the Fallout of Moral Responsibility

What initially started as a civil rights movement in Northern Ireland during the late 1960s quickly became a very different beast. In this initial phase of the conflict, the security establishment formulated a situational appraisal using intelligence that was “hopelessly out of date,” hindering a detailed assessment for the logic of intervention.^[6] As then-Home Secretary James Callaghan noted at the outset of the Troubles, the British “government knew less about Northern Ireland than [it] knew about our distant colonies, on the far side of the earth.”^[7] The colonial undertone mentioned here is telling. During a phase in which the British state was withrawing from a number of its colonial holdings, it came to the conclusion (through preconceived notions about the Irish character and the intelligence assessment dervied from Northern Ireland) that intervention was the most logical response.^[8] The question of leaving was “seriously considered,” but the British government reached the conclusion that withdrawal would lead to sectarian violence as it had in British India.^[9] The British state initially articulated its interventionist role as one of mediation. It saw itself as being morally superior to the two warring sectarian groups within the province.^[10] Furthermore, it did not understand the conflict for what it was — “one of divided loyalties and overlapping territorial claims.”^[11] Intervention thus became, in the minds of the British, a moral responsibility to the citizens of the province — a decision that was predicated on a strategic misreading of the conflict and the state’s place within it.

One key issue immediately emerged: as violence increased from 1969 onward, the need for accurate, timely, and actionable intelligence became critical. With this need, however, came moral quandaries, specifically surrounding the assessment of threats and the collection of information linked to the 1971 policy of internment. In their work on state-terrorist engagement, John Bew et al argue that interment was, “at best, a necessary but extremely high-risk evil,” as the security establishment found itself in a position where it could envision no other option available.^[12] However, the initial problem posed for the services by internment was one of threat assessment. Due to the lack of up-to-date information, the services found it difficult to distinguish between those individuals who were actively in the process of planning terrorist activity and those who shared “the religious or political beliefs or ethnic backgrounds of the terrorists.”^[13] Therefore, the services found themselves unduly targeting entire communities rather than individual perpetrators. Despite this, of the 357 people initially detained during internment’s implementation, only 12 were selected as part of Operation CALABA for submission to what were known as the Five Techniques.^[14] British intelligence expert Bradley Bamford has assessed that the use of the Techniques at this point in the conflict was the only way the services could collect actionable intelligence,^[15] although this proved to be at the cost of radicalizing previously unengaged members of the Catholic community due to a “one-sided application” of internment policy to Catholics.^[16]

British intelligence services have a long history of developed interrogation methods, honed through the various colonial campaigns of the early twentieth century. However, these were never codified until the creation of what vernacularly came to be known as The Bible, a “fully-fledged interrogation doctrine.”^[17] What was found therein were the previously mentioned Five Techniques, a form of “in-depth” interrogation that included “wall-standing, hooding, subjection to noise, deprivation of sleep, and restrictions on food and drink.”^[18] The Bible “advocated the highest standards of behaviour on grounds of both morality and effectiveness,” although its use during the initial phase of the Troubles demonstrated a correlation between peaks of violence and the sporadic abandonment of the guidelines.^[19] In the lead-up to the codification of The Bible, however, the Joint Intelligence Committee (JIC) had some difficulty in deciding morally acceptable parameters of in-depth interrogation. As Eunan O’Halpin illustrates in his work on the JIC during this period, they settled on “a range of techniques designed to break a detainee’s will to resist interrogation,” which, in the view of the British state, “fell short of physical abuse.”^[20]

While CALABA was initially deemed successful, its methods were critiqued as highly dubious and immoral. Despite a lack of criticism when performed in the colonial context,[21] when the Techniques were applied to British citizens “just 300 miles from Whitehall rather than nameless locals in the remote fastnesses of a dissolving empire,” the reaction was somewhat more severe.^[22] Moreover, many of the serious accusations of complicity in torture came from outside of the United Kingdom. For example, the European Court of Human Rights initially ruled in 1975 that the Techniques were a form of torture, although this was downgraded two years later to “degrading treatment.”^[23] The outcry resulting from this episode led to a reformation of The Bible. The resulting methods, which would be used throughout the remaining phases of the Troubles, now centred on “achieving empathy with the subject by building up a relationship over time,” rather than the previous focus of quickly obtaining intelligence by any means necessary.^[24]

Internment and the Techniques are demonstrative of some of the uncharted moral territory through which the intelligence services navigated in the initial phase of the Troubles. With violence on the rise and actionable intelligence still scant, British intelligence was faced with the question of how to balance its critical and acute need for information with the means through which that information would be acquired. Although leaders in the security establishment were responsible for how these moral questions would be handled, on-the-ground operational realities were hindered by “higher-level” hesitancy on behalf of policymakers who initially had a “lack of clearly defined direction” to intervene.^[25] In this sense, moral responsibility was little more than a grandiose notion which lacked any real strategic objective or operational plan. It was derived from a weak initial assessment of the conflict, which resulted in poorly-constructed and haphazard decisions on the ground. Moreover, despite the government assessing that its intervention “would make it possible to inject the supposed virtues of British political culture […] into the province,” by 1972, the security situation became untenable for the Stormont government. Direct Rule was implemented on March 24, 1972.^[26] As terrorism expert Peter Neumann has argued, in strategic terms Direct Rule “represented no fundamental shift in British government policy,” and is thus derived from the initial assessment of intervention as a moral responsibility.^[27] In this sense, we are able to see one facet of the complex relationship between intelligence and moral action — assessment is rendered poor when intelligence is outdated. This in turn had a subsequent impact on the morality of the decisions derived from intelligence assessments.

Enemy Engagement – Touts and the Long War

The year that Direct Rule was implemented was also the bloodiest of the Troubles. More than 400 people were killed as a direct result of conflict-related violence.^[28] As it became clear that the conflict had no immediate chance of abating, the services settled in for what became known as the “long war.” One of the key strategies for the intelligence services during this phase was the use of informants, also known as touts. From a strategic standpoint, the value and importance of touts in collecting operational intelligence and in reducing violence are indisputable. As Mark Urban highlights:

The effects of the informer war are profound: the level of violence is reduced; the republican community is rendered increasingly paranoid and must eliminate a proportion of its own membership in an attempt to retain its integrity…[it is] key to the containment of terrorist violence.^[29]

The role of touts became increasingly critical as the provisional Irish Republican Army (IRA) moved underground and adopted a more sophisticated cell structure consisting of Active Service Units (ASU).^[30]The tout system, in this respect, allowed the intelligence services two advantages: information and an increase in adversary paranoia. Touts could infiltrate into the deepest parts of the enemy organization, which provided the services vital operational intelligence and the ability to use that information to eliminate key players through arrest. On the latter point, Bew et al highlight that, “such was the rate of attrition that by some estimates the average ‘life span’ of the…Officer Commanding of the IRA’s Belfast Brigade was reduced to just six weeks.”^[31] Moreover, the psychological impact — the very fear of infiltration — proved to be key for causing internal disunity. For example, from 1978-1997, the IRA’s internal security unit, the Nutting Squad, killed 24 of its own members out of fear that they were informants.^[32]

The tout system presented clear moral quandaries. From an outside perspective, Northern Irish journalist and Troubles specialist Martin Dillon suggests that engagement led “to an erosion of confidence in the law and a dark suspicion that anything can be contemplated in pursuit of the defeat of terrorism.”^[33] Intelligence specialist Peter Gill framed the concern by indicating that the “deception and betrayal” inherent in running touts “may have deleterious effects on the moral characters of both handlers and sources.”^[34] It is important to understand that the conflict reached a critical stage at the adoption of the tout system – operational intelligence needed to abate the threat was scant, and the conflict began to manifest as a near-run thing. In such circumstances, the intelligence services were forced to examine any and all options which might offer a strategic advantage, however morally ambiguous those options were.

In this respect, the biggest moral question encountered by the services during this phase was one of negotiating cost: what moral boundaries could be crossed in an effort to protect valuable assets? If we examine the role of Freddie Scappaticci, codenamed Stakeknife, we see the implications of this question come to a head. A career informer, Scappaticci was undoubtedly one of the most valuable touts — the “jewel in the crown” for the intelligence services.^[35] Reportedly recruited as far back as 1978, Scappaticci held a primary role within the Nutting Squad. He led “investigations into suspected informers, inquiries into operations […] debriefings of IRA volunteers […] and the vetting of potential recruits.”^[36] Scappaticci’s role in the Nutting Squad meant that he was inevitably involved in criminal activity, including murder. Moreover, the cost of protecting assets as integral as Stakeknife was high. The assassination of Francisco Notorantonio speaks to this issue. Brian Nelson, a member of the Ulster Defence Association (UDA) tasked with identifying targets for the Ulster Freedom Fighters (UFF) assassination squad, was asked by the UFF to begin a file on Scappaticci.^[37] Nelson, however, was also a tout for the intelligence services. He was allegedly ordered by his handler to make a case to the UFF for targeting Notorantonio — a pre-Troubles republican paramilitary but who was no longer active — in an effort to save the strategically-critical Scappaticci.^[38] The UFF then subsequently targeted and killed Notorantonio.^[39] This assassination begs the question of what lengths are appropriate in maintaining the position of a key tout.

Martin Ingram, a former British Army soldier who served in the Intelligence Corps and later in the Force Research Unit, was familiar with the Stakeknife file. Ingram’s activities give an interesting perspective on the cost of running high-level touts. Ingram attests that although he now believes the moral price was too high for the intelligence Scappaticci was providing, his opinion was reversed when he was still operational within the intelligence establishment in the heat of the conflict. Ingram’s admission highlights an important issue in determining morality in intelligence: once removed from the conflict, hindsight forces practitioners to look at operational decisions through a lens of public morality rather than Omand’s proposal of an alternative morality not applicable within the functions of everyday life. This is not an attempt to justify or condemn the cost of running Scappaticci, but rather it is an effort to demonstrate the murky moral line which must sometimes be straddled in the pursuit of security.

Intelligence Surplus and the Timeline of Action

Evidence suggests that by the late 1980s, the amount of actionable intelligence coming into the services “meant that sometimes they had foreknowledge of terrorist attacks.”^[40] While this was a strategic coup for the services, it also presented new moral problems. The issues encountered were threefold: the question of when to use that intelligence, how to use it, and who to use it on. As this section will demonstrate, the services grappled with all three issues on a consistent basis throughout the latter part of the conflict.

The foreknowledge of attacks left the services with two options: to arrest individuals pre-emptively if irrefutable evidence existed, or allow the attack to proceed as planned in order to catch perpetrators in the act. The latter option — what some have called a policy of ambushing — became the favoured option in practice by 1983.^[41] This option almost inevitably resulted in a situation where terrorists (predominantly in these cases IRA volunteers) were killed rather than apprehended. This was seen by some as a deliberate “shoot-to-kill” policy on behalf of the security establishment. As Bamford argues, there is no legislative evidence that such a policy formally existed or that any such decision was taken at the highest level of the British government.^[42] However, the “large number of terrorists killed during this period, in circumstances where many believe that an arrest could have been made, is seen as evidence that such a policy existed” in practice.^[43]

Both the Loughgall and Gibraltar incidents, in 1987 and 1988 respectively, are examples of these supposed ambush and shoot-to-kill policies and are also reflective of how and when foreknowledge is used in practice. Loughgall involved the deaths of all 8 men in an IRA ASU headed by Jim Lynagh, after intelligence alerted the services to an impending attack on a Royal Ulster Constabulary base in the area.^[44] The SAS knew the time and location of the attack and waited for the ASU to plant its IED, but the ASU operatives were killed rather than apprehended.^[45] The Gibraltar incident, named Operation FLAVIUS, cuts a similar shape. After intelligence determined that an ASU consisting of Seán Savage, Danny McCann, and Mairéad Farrell was headed to Gibraltar for an attack, the SAS was again called in.^[46] All three were killed rather than apprehended. In both incidents, it is unclear who opened fire first. Moreover, both raise questions as to why the IRA volunteers were not arrested rather than terminated. Some argue that these ambush/shoot-to-kill incidences are examples of a deliberate policy between “the security service and the Thatcher government,” whereas MI-5 biographer and intelligence specialist Christopher Andrew has argued that “there is no persuasive evidence that the decision was premeditated” in either the Loughgall or Gibraltar incidences to assassinate.^[47] Regardless, the moral question here is centred on timeframe — knowing that last-minute efforts predominantly result in a perpetrator’s death, how long is it morally appropriate to wait before apprehending someone planning terrorism?

The attempted assassination of Gerry Adams by loyalist paramilitaries in March 1984 poses another interesting moral question about for whom preventative intelligence is used. Information was received from a tout within the UDA regarding a UFF plan to shoot Adams.^[48] Evidence suggests that there was no warning issued; only once the shots had been fired, missing their target, did the SAS – alleged to be present for the entire incident – apprehend the UFF members.^[49] As Urban argues, it “raised uncomfortable questions about the difference between the security forces’ response to foreknowledge of republican and loyalist attacks.”^[50] Urban’s assessment is also in line with the Stevens Inquiry’s finding that threat intelligence was not always treated equally between Protestant and Catholic groups.^[51]

Finally, the question of to whom intelligence should be shared also demonstrates some moral quandaries for the services. There were repeated allegations of collusion between the services and loyalist paramilitaries groups throughout this period. For example, many inside and outside the conflict were convinced that the services were “behind the increasing effectiveness of loyalist paramilitary organization in targeting republicans from the late 1980s.”^[52] Practitioners such as former Belfast Special Branch head Bill Lowry dismiss these claims, saying, “If we colluded as much with Loyalists as is often alleged by Republicans why did they kill so many innocent Catholics and very few Republicans?”^[53] From an independent standpoint, the Stevens Inquiry concluded that “collusion involving direct support of loyalist paramilitaries by security agencies was low level”; however, their forensic examination of contemporary classified documents found the fingerprints of eighty-one people on documents which “they had no lawful reason to possess.”^[54]

The question of morality by this latter stage in the conflict is rendered additionally complex when societal reaction is examined. Much of the public by this point seemed to “applaud the elimination of republican terrorists and [were] less than curious about the moral tightrope that must be walked to make such actions possible.”^[55] In this respect, the concept of conflict fatigue must be taken into account when moral questions are at hand. As Bamford notes, the Gibraltar incident in particular is a prime example of the public’s growing indifference to violence: it “showed the moral ambivalence of a liberal democratic society toward the use of lethal force in counterterrorism operations.”^[56] While public opinion should not be the only barometer by which intelligence agencies should measure the morality of their decisions, how the parameters of morality alter and expand as a conflict protracts is worthwhile to note.

Conclusions

The standard argument today assesses that the conduct of the British intelligence services in Northern Ireland “was ultimately very effective but at the price of employing some highly dubious methods.”^[57] The conflict, as it presented itself from its inception, was one which demonstrated the difficult position of comprehending the complexities inherent in quashing domestic terrorism. As we have seen, negotiating methods vis-à-vis critical time restraints and information scarcity in the first phase of the conflict led to some hastily-implemented and occasionally counter-productive collection realities, some of which that had questionable moral groundings. As the conflict protracted and long-term engagement was established, the murkiness of moral conduct by the services became further complicated, somewhat paradoxically, by the increasing success and improvement of their intelligence capacity in the province.

From the broadest perspective, the most significant question raised by the British experience in Northern Ireland is one of scale: how far is a state willing to expand the borders of moral conduct for the sake of security? Perhaps the best way to approach these ambiguities is to consider Omand’s recommendation. While there is not a suggestion that intelligence services should act within a moral vacuum or that they should conduct their business without the appropriate oversight and review mechanisms in place, there ought to be an acknowledgement that given the very nature of intelligence, its conduct and practice requires a “different morality” than to what is abided in civilian life. n

Ms. Kalajdziovski is an MA candidate in the Intelligence and International Security program at the Department of War Studies at King’s College London. Her research interests include the role and historical evolution of intelligence in counterterrorism practice pertaining to domestic terror threats.