I thank the Chairman and the committee for the opportunity to come here today to give our evidence on the ongoing homicide review in An Garda Síochána.
I commenced my current role as deputy head of the Garda Síochána analysis service, GSAS, on 1 August 2007. As part of the GSAS management team, I have been integral to the introduction and establishment of the analysis service in An Garda Síochána. Not only has this involved recruitment, structuring, training, the development of systems and processes, progression planning and budgeting, but also gaining acceptance for the use of analysis in the wider organisation. I have responsibility for crime and policing analysts within headquarters units and special crime operations. My remit is very broad, ranging from oversight of national assessments of volume crime, such as burglary and assault, through serious and organised crime, for example, homicide, drugs and immigration, to matters concerning national security. I have insights into the organisation across the spectrum of policing functions and have experience of interacting with personnel at all grades and ranks.
Prior to August 2007, I worked as an analyst in the Police Service of Northern Ireland, PSNI, for nearly six years. During this time, I supported a multitude of investigations into serious and organised crime and delivered many reports containing operational recommendations, which were used to set organisational priorities and guide significant strategic decisions by senior officers.
In July 2016, I received a direct request for analytical assistance from the Garda national protective services bureau, GNPSB. The request was for the Garda Síochána analysis service to conduct a ten-year review of domestic homicide from 2007 to 2016. The chief superintendent who made this request did so to satisfy particular obligations the GNPSB has with respect to: the national strategy on domestic, sexual and gender-based violence from 2016 to 2021; An Garda Síochána’s domestic abuse intervention policy of 2016; the development of a risk assessment tool in An Garda Síochána; the development of divisional protective services units in An Garda Síochána to meet commitments made in the 2016 policing plan and made to the Policing Authority; the EU victims directive of 2015; the Criminal Justice (Victims of Crime) Act 2017; and the Istanbul convention, which is the Council of Europe convention on preventing and combating violence against women and domestic violence. I personally tasked senior crime and policing analyst, Ms Laura Galligan, with this review.
Ms Galligan had significant, relevant experience from her previous role as senior scientist with the Office of the State Pathologist. I believed this expertise would add great value to the review of domestic homicide.
Having been deputy head of analysis with An Garda Síochána for nearly 11 years, I have been uniquely placed to understand the organisation and identify areas of concern or weakness which require prioritisation, particularly if the transformation agenda is to succeed. Throughout my tenure with An Garda Síochána, recurrent areas of concern identified by the analysis service have been PULSE recording and data quality. Although there is no clearly defined process for GSAS to raise issues identified with regard to data quality, we have always escalated our concerns, whether to crime, policy and administration - latterly known as the policy development, implementation and monitoring unit - or the Garda information services centre in Castlebar and frequently to senior Garda management in analytical reports and at management meetings and conferences.
Data quality is of fundamental importance to analysis. A dearth of good data seriously limits the ability of an analyst to conduct complete and robust analysis in any sector. I voiced my concerns about data quality to Ms Galligan as she embarked on the ten-year review of domestic homicide. Within a short time, it became clear there was a deficit of information on the PULSE system to enable Ms Galligan to determine whether a homicide was of a domestic nature or otherwise. Ms Galligan therefore requested permission to liaise with the Office of the State Pathologist in order that she would have independent and verifiable supplementary information to ensure data quality. I take this opportunity to thank Professor Marie Cassidy and her team for engaging in this collaboration with the analysis service.
When Ms Galligan analysed and cross-checked the files held in the Office of the State Pathologist, information held on PULSE and homicide spreadsheets previously produced internally by the analysis service on a monthly basis, she identified inconsistencies which she deemed to be significant. When she brought this to my notice, I concurred with her views. I immediately recognised the gravity of the matters raised and the many possible ramifications. In my view, the issues raised went right to the heart of policing and the ability of An Garda Síochána to protect and serve the public.
While there were concerns regarding the accurate provision of crime data to the Central Statistics Office, the primary concern was with regard to the victims, their families and loved ones. I could immediately see the very serious organisational risks for An Garda Síochána in terms of reputational damage and public confidence. I asked Ms Galligan to specifically consider a sample period from 2013 to 2015 to expeditiously highlight the main concerns. I believed it was critical that we establish an evidence base to clearly illustrate the various issues arising. Ms Galligan undertook this work on the sample period as I had directed. On 24 November 2016, she emailed me a draft of her report entitled, Comparative Analysis of the Recording and Reporting of Homicide Incidents in the PULSE Database and the Office of the State Pathologist. Following a specially convened meeting on 28 November 2016, I recommended to management that the Garda executive be briefed immediately. I am aware that this briefing occurred on 29 November 2016. We understood at this time that the professional standards unit would be asked to conduct reviews of the cases highlighted. I had this impression from communications with my management. The role of the professional standards unit is to examine and review, as directed by the Garda Commissioner, the operational, administrative and management performance of An Garda Síochána at all levels; propose measures to the Commissioner to improve that performance; and promote the highest standards of practice, as measured by reference to the best standards of comparable police services, in operational, administrative and management matters relating to An Garda Síochána.
Subsequent to the briefing of the Garda executive on 29 November 2016, I asked on several occasions whether anyone had been in contact to address the matters raised. I was concerned that there did not appear to be a sense of urgency to initiate a full review on the basis of Ms Galligan's report. In mid-January 2017, officers from policy development, implementation and monitoring, or PDIM, commenced a review of 41 cases which had been referenced in Ms Galligan's draft report. The report had been escalated to management in draft format due to the urgency with which I perceived the need to raise the issues identified. It had remained in draft format as priority was given to beginning to look at other years. I did not deem it a time for worrying about perfection in report writing. For me, the emphasis was on identifying the issues and raising them as quickly as possible. Issues identified by Ms Galligan ranged from potential misclassification to data quality issues of varying degrees of severity. Ms Galligan will shortly go through the findings of her analysis.
Ms Galligan and I attended meetings with representatives from PDIM on the following dates in 2017: 17, 19 and 24 January; 7, 13, 20 and 21 February; and 2 and 30 March. We specifically requested the meeting of 30 March. After the first meeting on 17 January 2017, all relevant reports were forwarded by me to PDIM electronically with the following email:
... As a result of reviewing case files held within OSP a number of potentially serious anomalies were identified which raised the question of whether PULSE had a complete record of all homicides (not exclusively domestic). As the numbers recorded as homicide in OSP were higher, the systems had to be manually cross-referenced. This has raised the possibility that certain cases on PULSE may be incorrectly classified. In addition, a series of other data quality issues were identified.
The documents above outline the variety of issues identified. It is recommended that a review is carried out of each case noted to determine whether there is a consensus that there are matters of concern. GSAS will happily assist in any way we can. We would appreciate being kept abreast of developments as this will determine how we proceed with the review of additional years. 2012 is nearly complete and will be disseminated to you shortly. It is then intended to research 2016 as we now have a complete year of data. This will give 5 years which have been reviewed. A decision will then have to be taken as to whether we continue to work back from 2012. The original domestic homicide review was intended to cover a period of 10 years ...
During the nine aforementioned meetings, cases highlighted between the years 2013 and 2015 were discussed. These were very robust discussions, not only about each individual case but also high-level issues such as consistency in recording practice across crime types. We entered into these meetings in good faith and genuinely believed the intention was to fully and independently review each of the cases. However, as it transpired, there was an apparent reluctance to countenance many of the issues we tried to raise. We made many attempts to highlight the potential organisational risks, the inconsistencies we were identifying and the absolute need to review the approach to death classification and investigation, not only within An Garda Síochána but also in collaboration with partner agencies. We made it very clear at these review team meetings that agreement had not been reached between the parties involved on many of the key matters raised.
Although a review of 2012 had already been carried out, we were instructed not to send the related executive summary to PDIM. We were not permitted to reference cases from outside the sample review period - 2013 to 2015 - even though they were pertinent to highlighting crucial facts. Every case will have particular nuances to be referenced, some of which may ultimately impact on policy decisions.
It was my full expectation that, as Ms Galligan and I had attended all meetings of the review team, we would, in spite of the often tense and quarrelsome debates, be involved in or at least consulted on the drafting of any report concluding the findings of the meetings and putting forward any recommendations. It was our understanding that our input was necessary given our central role in raising the issue and highlighting the problems. However, instead of being consulted or involved, Ms. Galligan and I were excluded from the preparation of any report or review document. We were not consulted regarding our views on how to ensure good data quality that would enable the work of GSAS.
On Monday, 8 May 2017, I was handed a 59-page, hard-copy document entitled, Review of Specific Sudden Death/Homicide Incidents recorded during the period 2013 to 2015. This was a report produced by PDIM officers from the review team. I had inquired as to the status of such a report on 28 and 30 March and I am aware Dr. Singh, the head of GSAS, had requested on several occasions during April 2017 to be furnished with any such report. When this report was provided to me by Dr. Singh at 12.20 on the afternoon of 8 May 2017, I was informed that views on it were required by the close of business that day. This was not possible.
In this report, Ms Galligan's methodology was deemed to be "inherently weak", "confined" and "restricted" and there were repeated attempts to undermine and erode confidence in the findings of the initial sample review. This was of huge concern to me. I felt that my integrity and the integrity of my colleague, for whom I have the utmost respect, was under attack. Our concern was of a professional nature. We had been tasked with generating a report, as required by a chief superintendent.
In order to undertake the task and provide the report requested of us, we had to adhere to professional standards and norms. The fundamental and key component part of this, as any data analyst will confirm in any industry, is the underlying data quality as it underpins the analysis. Bad data will lead to bad analysis. We were concerned about the accuracy and quality of the data, and to have ignored our concerns would have been to ignore our ethics and professional standards. No engagement with the review team gave us any assurance regarding the data quality.
Over the course of 8, 9 and 10 May 2017, I believe that very significant pressures were brought to bear on Ms Galligan and me, to persuade us to sign off on the PDIM report. I felt very pressurised in heated meetings which occurred on 9 and 10 May 2017. I also received a series of telephone calls on the afternoon of 9 May during which significant pressures were brought to bear. Professionally, I could not sign off on the PDIM report and made it clear to all concerned that I would not bow to pressure. On 11 May 2017, I felt compelled personally to transmit my views about it in a five-page letter to several members of senior Garda management. Within this correspondence I noted:
My position has been very clearly stated on multiple occasions throughout this process, and remains unchanged. During the meetings between PDIM, Ms Galligan and I, no agreement was reached, either with regard to the classification of individual cases discussed, or in regard to the more strategic issues relating to homicide investigation and death classification.
To make sure there is 100% clarity, the report entitled, Review of specific Sudden Death/Homicide Incidents recorded during the period 2013 to 2015 is, in my opinion, a one sided view of what occurred within the meetings between PDIM, Ms Galligan and I. I do not agree with many aspects of the report and I therefore cannot and will not be signing off on recommendations contained therein.
At 8.50 a.m. on 12 May 2017, as a direct consequence of the letter I had sent internally the day before, I received a telephone call. In line with the requirement of the committee that persons are not named or referenced in such a way as to be identifiable, I am unable to say who made this call. During this call, I was made aware that a report, which I had submitted on 11 April 2017 for onward transmission to the Policing Authority - request 210 relating to the homicide review, had never been sent to it. I was informed that this report had been discussed among senior Garda personnel prior to a private Policing Authority meeting on 13 April 2017. If this was the case, then it was known at that time that agreement had not been reached during the meetings of the review team. This makes it difficult to understand how certain comments were made and assurances given during the public Policing Authority meeting on 27 April 2017. Several other things were said to me in this conversation, for which I have contemporaneous notes and which I can make available if the committee requires.
On the afternoon of 12 May 2017, Ms Galligan and I met with members of senior Garda management. We were very honest about our views during this meeting. It was made clear that there had been difficulties during the review team meetings and that we felt we had not been listened to or well treated. Following this discussion, it was agreed that there still needed to be a firm decision taken on the classification of a number of cases highlighted in Ms Galligan’s original report. During this meeting, and in a subsequent email I sent on 13 May 2017, there was agreement that Ms Galligan and I would be given sufficient time to go through the PDIM report, review the conclusions of this report vis-à-vis our own conclusions relating to the 41 referenced cases and provide a response to the report. There was also to be a meeting convened immediately to address the difficulties which had arisen at the review team meetings. This meeting eventually took place on 26 June 2017 but unfortunately resolved nothing.
On 26 May 2017, a draft response was provided to senior Garda management, which included the agreed review of the 41 cases. A 123-page final report was submitted to senior Garda management on 7 June 2017. This final version of the report included 20 pages of recommendations, made on the basis of the findings of the 2013 to 2015 review. On 30 June 2017, another member of senior Garda management was also briefed by Ms Galligan and me, as the member was to set up a multi-agency working group to deal with policy and governance issues arising. This group was to include representatives from the Central Statistics Office, the Policing Authority and the Department of Justice and Equality, among others.
In the weeks that followed, Ms Galligan continued to produce reports identifying cases of concern for 2017. These reports were escalated to senior Garda management. Again, these concerns related to potential misclassification of incidents and broader data quality issues. Ms Galligan also continued to provide assistance with rectifying issues uncovered with regard to fatal collisions and incidents of dangerous driving causing death. Ms Galligan also continued to refer cases to the protective services bureau, where we had very genuine concerns that a person may be living in a vulnerable situation. For example, we made the decision to escalate cases where we felt that a new partner may be at risk. These were identified as Ms Galligan did her utmost to piece together histories where escalation in the number and severity of incidents pointed to significant risk indicators. At times this was like stitching together a patchwork quilt, as there were incidents that were correctly classified and then there were those found languishing in inappropriate categories, such as attention and complaints. I strongly believed that we would never regret referring such instances to the protective services bureau, but we might regret not doing so.
I continued to pursue senior Garda management for responses and guidance as to how we should proceed with the review of other years. I was concerned that GSAS had been tasked with reviewing a ten-year period and only four years had been initially reviewed, with much more work required. Between May and mid-September 2017, we continued to come under pressure to come to agreement regarding the 41 cases. However, even following adjudication by a member of senior Garda management, agreement could not be reached because the stance was still to insist there was nothing wrong. In mid-September 2017, Ms Galligan and I were informed that we had indeed been correct all along and that there were cases which had been misclassified. This gave us an immense sense of relief and confirmed that we had been correct in standing our ground. However, we have yet to receive any formal guidance as to what is required of us.
In particular, the stress and pressure we had been subjected to on a continual basis from our first raising these issues have not been satisfactorily acknowledged or adequately addressed. It is my view that we were subjected to severe pressure to withdraw our concerns, to ignore our professional standards and to agree with the views of the sworn members of the review team. Our integrity, both personal and professional, was undermined and attacked. We do not know what the motivation of the review team was in disregarding our views, seeking to minimise the importance of the issues and then seeking to force its report on us and have us sign off on same, inclusive of the scant and inadequate recommendations contained therein.
On 11 December 2017, Ms Galligan and I requested, by email, a private meeting with senior Garda management due to our ongoing concerns about the lack of progress. This email seemed to set in motion the appointment of another member of senior Garda management to lead on the homicide review. Our private meeting with senior Garda management occurred on 10 January 2018. Ms Galligan and I also wrote to senior Garda management again on 12 January 2018. While this seemed to promote some further activity, it would seem that recent media coverage of the homicide review and the intervention of the justice committee have led to the most significant progress to date.
Such was our level of concern at certain points of the process that we made attempts to engage with the Policing Authority by telephone or in writing on the following dates in 2017: 3 April; 16 May; 30 June; and 19 July. As outlined earlier, a report was also prepared on 11 April 2017 for onward transmission to the Policing Authority, but I subsequently discovered on 12 May 2017 that this report had never been sent to the authority. I do not understand the rationale for this, as there was no sensitive, case-specific information included. On 1 August 2017, I was told by a member of management that the member had been informed by a member of the Policing Authority, over the telephone, about the approaches I had made to the authority. This was raised with me in the context of management trying to ascertain what my thinking was when I did this. To date, we have met no one from the Policing Authority regarding the homicide review.
Ms Galligan and I are fully committed to doing everything possible to brief and support the newly established working group, which met properly for the first time on 19 February 2018. We may be about to achieve the full and independent review of death classification and investigation that we have been consistently seeking since November 2016. However, we still have no formal written direction as to what is required of us into the future or clarity as to how this will be balanced with existing demands. Most notably, Ms Galligan’s original methodology, which was the subject of particular attack, will form the cornerstone of the new working group.
We hope that a multi-agency group will soon be established to address the policy, governance and education aspects of the review as raised in our report of 7 June 2017. We continue to have serious concerns about what will be done to address misclassification and data quality into 2018. We are unaware of any current monitoring of the issue and, to our knowledge, no process has yet been agreed in that regard.