2002 theft of intelligence files from Castlereagh police centre

The 2002 theft of intelligence files from Castlereagh police centre was a meticulously prepared raid on the covert network at the heart of security operations in Belfast. Three men entered one of the city’s most heavily guarded facilities and left with files identifying informants inside the IRA and other paramilitary groups, turning an intelligence centre’s secrets into a source of immediate danger and political dispute.

The Raid on Room 2/20

Just after ten o’clock on Sunday night, St Patrick’s Day 2002, three men in suits arrived by car at the gates of the Castlereagh complex in East Belfast. The area was quiet, and only about twenty people were on duty across the fortified compound. Trevor Campbell, who had spent decades interrogating and turning paramilitaries, was preparing to retire and was not there that night. The Royal Ulster Constabulary had been replaced four months earlier by the Police Service of Northern Ireland under the Good Friday Agreement, but Castlereagh still resembled a forward operating base, its high walls topped with barbed wire.

The men presented what appeared to be army identification badges at the gate and again at the front desk. Both times, guards let them through. Inside a facility that housed police, British Army personnel and intelligence staff, they moved directly toward a particular office. Its usual location was being renovated, and its work had been temporarily moved elsewhere; the thieves knew where to find it. They reached Room 2/20, the secret, round-the-clock centre where hundreds of informants across Belfast could contact handlers in the police, the army and MI5.

Only one Special Branch constable was in the room, manning the phones. The men struck him in the jaw, bound him to an office chair, hooded him and taped his mouth. They put headphones over his ears, playing music from a Walkman, then used keys left on the desk to open drawers and filing cabinets. Although one of the intruders periodically checked the constable’s pulse, the checks eventually stopped. When he managed to free himself and remove his blindfold, the men had gone and the shelves had been emptied.

The Intelligence Breach and Its Aftermath

The men escaped without being noticed, leaving behind only a lapel badge marked SAVE THE RUC. The files and notebooks they took contained identifying details and code names of informants working inside the Irish Republican Army and other paramilitary groups. The seizure exposed people who had supplied intelligence to the security forces and forced police to warn informants that their identities might have been compromised. More than three hundred people were ultimately relocated.

The breach provoked sharply conflicting accounts of what had happened. A security source described it as a violation of the peace process and an act of war; the Irish Republican Army, by contrast, said the operation was an inside job by the authorities. Most observers concluded that the Provisional IRA had carried it out, but the apparent ease with which the thieves reached a secret room deep inside Castlereagh raised the possibility of help from within. Investigators sought a Castlereagh cook for questioning. He was an associate of Denis Donaldson, a former Irish Republican Army leader who had become a Sinn Féin official at Stormont.

Speculation also focused on whether the files could identify one especially important informant. Press reports linked the break-in to the hunt for a supposed high-ranking spy inside the Provisional IRA, known by a code name variously rendered as Steak Knife, Stake Knife or Stakeknife. The name’s image of a blade driven into the movement captured the fear surrounding the possibility that British intelligence had penetrated its highest ranks. Stakeknife had been described in an earlier account as the crown jewel of British intelligence in Ulster, making the stolen records seem potentially valuable not merely as a list of compromised sources but as a route toward the movement’s most consequential suspected informer.

The Files and the Limits of What They Could Prove

The significance of the stolen files lay partly in the questions their reported target raised, not in any confirmed answer the raid supplied. The suspected high-level informant known as Stakeknife had become a symbol of possible British penetration at the heart of the republican movement. Yet the account of the break-in establishes neither that the thieves found Stakeknife’s identity nor that the stolen material resolved the rumours surrounding him. The files exposed a vulnerability in the security forces’ intelligence network; the identity and reach of the suspected spy remained contested.

That distinction matters in the book’s wider treatment of evidence about informers. In the case of Jean McConville, the children’s recollections placed her interrogation and beating in early December 1972, shortly before her abduction. A police log later located by Nuala O’Loan recorded a beaten woman, named as Mary McConville and identified by her address at St Jude’s Walk, who said she had been warned to stop giving information to the military. But the log dated that encounter to 29 November, seven nights earlier than the date the children believed their mother had been taken to be questioned. If the log and the family’s abduction date were both accurate, more than one night separated the warning from her disappearance, a chronology more consistent with Brendan Hughes’s account than the family’s remembered sequence. These records complicate the story; they do not by themselves establish that McConville was an informer.

The competing archives surrounding the case also show why a stolen intelligence file and an oral account carry different kinds of risk. In fifteen hours of recorded interviews with Anthony McIntyre, Dolours Price never mentioned McConville. In separate recordings made by Ed Moloney, however, she discussed the disappearance in detail, including being called back to Dundalk after McConville had been held there for several days. When Ivor Bell’s Belfast Project interview was later considered in court, a voice analyst judged that the speaker known as Z was likely Bell, while the defence and a PSNI detective disagreed over whether the recording implicated him in aiding the murder. The break-in’s haul of names and code names made intelligence records immediately dangerous to those exposed; the later disputes over interviews show a different problem: evidence could disclose a speaker or a contradiction without settling what had happened or who bore responsibility.

A Breach in an Unfinished Peace

The raid took place in a society where the end of large-scale violence had not ended the struggle over who could speak for the past. The Good Friday Agreement had brought power-sharing and a reformed police service, but no formal process for deciding how the conflict should be understood or commemorated. Gerry Adams continued to deny that he had ever belonged to the IRA, while Sinn Féin’s leadership, wary of internal dissent, was described as maintaining a monopoly over the memory of republican armed struggle. The theft of files from a covert intelligence centre belonged to this unsettled landscape: the peace was in force, yet the identities of informants, the reach of intelligence networks and the authority to tell the conflict’s history remained charged matters.

The wider setting resisted any simple distinction between a finished war and a settled peace. Belfast’s centre had acquired the familiar stores and tourist attractions of a prosperous city, and former combatants guided visitors around the Troubles’ murals and flashpoints. Yet more than 90 per cent of children in Northern Ireland still attended segregated primary schools, and peace walls continued to divide neighbourhoods. A later report argued that republican and loyalist organizations, through the authority of their surviving hierarchies, could influence and restrain members and thereby help prevent a return to extreme violence. That paradox gives the Castlereagh theft its wider resonance: the clandestine structures implicated in the conflict had not simply vanished, and the fragile order after it depended in part on organizations whose internal discipline and hidden knowledge remained subjects of suspicion.