Smithwick Tribunal failed to establish collusion based on balance of probabilities

Major flaws in the tribunal process should lead to Government rejecting the collusion finding

As retired Garda Chief Superintendents we acknowledge that the murders of RUC Chief Superintendent Harry Breen and Superintendent Bob Buchanan on the March 20th 1989 were a brutal and callous act committed by the Provisional IRA.

We unreservedly affirm our view that any member of the Garda Síochána who colluded in this crime deserves to be dealt with in the severest manner provided by law and should suffer commensurate public opprobrium.

However, the Smithwick Tribunal has singularly failed to establish collusion based on the balance of probabilities.
The report runs to 1,652 pages that are available online or in hard copy of 434 pages with part II of the report contained on a CD disc. It is an extremely arcane document both in terms of layout and narrative. At the outset Judge Smithwick clearly says that;

There is no record of a phone call, no traceable payment, no smoking gun…(23.1.2)


He did not make a collusion finding against sergeants Finbarr Hickey, Leo Colton or Owen Corrigan. He makes this finding against other unnamed gardaí present in Dundalk Station on March 20th, 1989. We have searched for the logic and the reasoning behind this finding and it obviously should have two manifestations, legal and factual. In legal terms the assumption is that it is on the “balance of probabilities”. However, the balance of probabilities make sense provided there are individual facts, which collectively allows one to make a global conclusion. There are no such individual facts.

The tribunal commenced its work with a number of inherent disadvantages. In the first instance it was fully 17 years after the crimes before it got down to doing “investigatory” work. There are obvious consequences which flow from this passage of time – particularly flawed memory. Well ventilated discussions alleging collusion against a named individual during this time certainly could have contaminated recollections.

The second major problem was the fact that it only had powers to compel individuals in the Republic. All relevant parties elsewhere were voluntary participants. This was like asking an investigator to investigate a murder in a two-storey building, where the crime was committed on the second floor. The investigator was empowered to use a full range of investigatory powers for persons and things on the ground floor but when it came to investigating the crime scene and those most intimately connected with it, this had to be achieved on a voluntary basis.

Thirdly, the tribunal did not have a team of professional crime investigators to assist with the process. Such a team would have advised that all criminal investigations start at the crime scene and radiate out from there. Similarly professionals would have cautioned against pursuing a preferred suspect or suspects to the effective exclusion of all others. This bias can be either a conscious act or unconscious approach influenced by prior events.

Fourthly, it failed to conduct a crime reconstruction or have a detailed analysis carried out of Dundalk Garda Station. This was major omission of rudimentary fact-finding.

The tribunal did not forensically examine this proposition. Any professional investigator would have mapped the physical environs of the station, providing a detailed floor plan, the positions of all phones within the building, the office occupied and used by whom and the identity and positioning of all gardaí. A professional investigator would have reminded the tribunal that all crimes have three key factors; motive, opportunity and means. This reconstruction would have been a formidable task but, after all, the tribunal took eight years to arrive at its unsubstantiated conclusion.

Failure to follow this forensic path was a major flaw.
Basis for findings

The tribunal based its collusion findings on four pillars. These were:




three strands of intelligence received by the Garda Síochána;


2012 information from the PSNI, and;


the evidence of double agent Kevin Fulton/

Peter Keeley


Smithwick says that the circumstances alone are sufficient to allow a finding of collusion. It is not clear what specific circumstances are referred to or on what grounds a particular set of circumstances is preferred over another.

Indisputably the PIRA had the capacity to mount this operation independently. Over 300 security force personnel were killed by PIRA in Co Armagh between 1969 and 2001. Many civilians also died. PIRA needed no help in the killing game. The tribunal also discarded the notion of ongoing PIRA surveillance operations, which had identified the car of Supt Bob Buchanan for many months.

It is important to note that it was the car and not the individual that was repeatedly recognised in attendance at Dundalk Garda station.

The tribunal erred in asserting that Chief Supt Harry Breen was the target of the operation. The former PIRA commander informed the tribunal; “Harry Breen being spotted in Meigh (on an earlier occasion) added impetus to the operation but was not the tipping point. At that time Buchanan’s identity was unknown. It could have been anyone in the car that day. It could have been Bob Buchanan and someone other than Breen.

“If he had got a new car we would have lost the operation. It was the car. We did not know Bob Buchanan. We knew the car. I have no hesitation in saying that a new car would have lost the operation for us.”

Had the PIRA been aware that Mr Breen was in the ambushed car that day they had every opportunity to capture him. It is perfectly clear from the crime scene and witness information that the objective was to kill the occupants and retrieve information from their bodies.

In our professional opinion the PIRA operation was planned on the basis of ongoing surveillance conducted over a much longer timeframe than that accepted by the tribunal. This conclusion is based on the balance of probabilities and our many years of experience observing the activities of the PIRA in an operational context, particularly in South Armagh.

Three strands of intelligence received by the Garda Síochána
Certain bits of information came to gardaí through a Garda agent handler and from one defined source some time after the murders but not before there had been much newspaper speculation on collusion. This information names no individual and names no station, it is hearsay information. It is impossible to evaluate that information at this remove. The tribunal speaks approvingly of its contact with the Garda handler and the hearsay information. This halo effect is of course unfortunate and is no criticism of the handler but all good practice suggests that the handler should not be the evaluator of the information. There in an internationally recognised system for grading information. This is basically an alphanumeric grid by which the source of the information and the quality of the information is separately evaluated. A top source would be rated A, top quality information would be rated 1. So the best possible rating would be A1 and the least compelling would be rated D4. The tribunal was not provided with a rating for this information but on the disclosed information it would be rated low in the quality of the information.

In our professional opinion the tribunal should not have placed probative value on these strands of information. The information is at best speculative and was largely double hearsay.

Live and of the moment intelligence
This also is "hearsay" information which was unverified by the tribunal. The so-called strands of information are a mixture of innuendo and suggestion. It is also possible that dissident republicans who are opposed to the line taken by the PIRA in the peace process made the allegations. Again the tribunal placed great confidence in Assistant Chief Constable Drew Harris's credibility as a witness while disregarding the fact that he was the conduit of the information, not the handler and not the source. This intervention also had a major impact on the findings. It introduced the suggestion of other colluders but not gardaí publicly associated with the tribunal.

The tribunal should have insisted on a process whereby both services agreed the evaluation of the information. Failing that agreement the tribunal should have invoked the appropriate dispute mechanism as set out in their terms of reference. This intervention was referred to as “nonsense on stilts” by counsel for the Garda commissioner.

This situation has to be taken alongside the stated position of the RUC in 2000 where they confirmed that they had no information on collusion with regard to the murders. The PSNI departed from earlier practice in that no grading or no substance was provided to the tribunal.

In our professional opinion the information was more prejudicial than probative and should have been disregarded by the tribunal. We agree with the comments of counsel for the Garda Commissioner when he described this information as “nonsense on stilts”. Acceptance of this information also definitively ruled out any garda named before the tribunal.

The evidence of Kevin Fulton/
Peter Keeley
This is an area where the tribunal strayed into the underworld of black operations. For the record it would be naive in the extreme not to understand that most parties to conflict engage in deniable black operations. This involves agents like Fulton who infiltrate the opposition or it involves the recruitment of informants. Motivation is an important part of this scene; people do it for money, some people are coerced into playing the role and a number may simply do it for philosophical reasons. Fulton was a paid agent. He operated on both sides of the Border and indeed a good question is how many colleagues he had in this profile. Fulton was basically a low-level operative but it is clear that he was accepted into PIRA circles. He admits to being a driver for the PIRA internal discipline unit, the so-called "nutting squad" who abducted, tortured and killed, and who were also highly infiltrated by the British agent "Stakeknife".

He admits to lying to his handlers in the past and, paradoxically, the tribunal awards him credit for this admission. He also alleges that RUC officers were passing information to the PIRA (Day 67, page 93, tribunal record). However the preponderance of Fulton’s evidence relates to Sgt Corrigan.

Corrigan has not been found to be the source of the alleged collusion. Accordingly, Fulton’s contribution is of no probative value in relation to the findings. Yet the tribunal finds him credible.

Also it should be noted that a greater portion of the tribunal’s time was spent attempting to find evidence of collusion by Sgt Corrigan and the main potential source of this information was Fulton. When no finding was made against Mr Corrigan a big gap opened up and, very conveniently, the unnamed gardaí were parachuted into this space without a scintilla of evidence being adduced.

In our professional opinion, the acceptance of Fulton’s information is both confusing and contradictory and does not support the finding of collusion against unnamed gardaí.

Were the Provos telling the truth to Smithwick?
It may seem to be morally repugnant to accept the information provided to the tribunal by former PIRA members. Nevertheless their information requires professional scrutiny. They are a primary source of information. These are the people who committed the murders. On the balance of probabilities there is much to consider in their account.


We note with deep concern that the tribunal’s finding:

1Impugns the good name and reputation of gardaí who have had no opportunity to be represented before the tribunal or to hear and address the allegation, which has been made against them;

2Adversely affects the relationship between the Garda Síochána and the PSNI; and

3Undermines the capacity of this State to actively pursue the many substantiated acts of collusion committed in Northern Ireland in the course of the conflict.

Our initial reaction as retired gardaí was one of intense shame that members of the Garda Síochána had allegedly colluded in sending two RUC officers to their deaths in 1989. Having served in Dundalk and recognising the bond of professional respect and confidence which existed between Dundalk gardaí and Newry and Armagh RUC this was an appalling finding. Indeed one was very much aware that RUC Newry had suffered the greatest police casualties during the years of conflict.

The tribunal has singularly failed to establish collusion based on the balance of probabilities and we ask the Government to reject the collusion finding.