Denham C. J. Hardiman J. O’Donnell J. McKechnie J. Laffoy J.
The Commissioner of An Garda Síochána,
The Minister for Justice, Equality and Law Reform,
The Director of Public Prosecutions, Ireland and
The Attorney General
Ireland, The Attorney General, and
The Commissioner of An Garda Síochána
Judgment delivered on the 3rd of November 2015, by O’Donnell J.
1. These cases are appeals against orders of the High Court in L.M. v. The Commissioner for An Garda Síochána, The Minister for Justice, Equality and Law Reform, The Director of Public Prosecutions, Ireland and The Attorney General  1 I.L.R.M. 132 (“LM”), (Hedigan J., 20th January, 2011) and Belinda Lockwood v. Ireland, The Attorney General and The Commissioner of An Garda Síochána 1 IR 374 (“BL”), (Kearns P., 10th December, 2010) dismissing the respective claims either in whole or in that part relating to a claim in negligence, after a determination in each case of a preliminary issue that the defendants in each case did not owe a duty of care to the plaintiffs. (No issue was raised in this Court as to the title of the proceedings and accordingly I have followed the approach taken in the High Court in the respective cases). It follows from the fact that the cases were determined after the trial of a preliminary issue that no evidence was heard or facts found. Accordingly, it should be understood that the facts hereafter set out are the matters alleged by the plaintiffs which the Court must assume to be true or to be capable of being established at trial in order for the determination of the preliminary issue .
LM v The Commissioner of An Garda Síochána & Ors
2. The plaintiff was born in 1978. In May, 1990, at the age of twelve, she made a formal complaint of rape against her father to An Garda Síochána. A statement was taken from her in May, 1990, and a short statement was taken from her mother in December of that year. It is said that the plaintiff’s complaint was validated by examination at Crumlin Children’s Hospital. (I understand this to mean that on medical examination, the plaintiff’s condition was found to be consistent with her complaints of sexual abuse). However, shamefully, nothing further occurred in this jurisdiction for six years. In 1996, the English Child Protection Authority (“CPA”) contacted An Garda Síochána. The plaintiff’s father was at this stage living in England. It appears that following contact from the plaintiff’s mother, the CPA had been alerted to the complaint about the plaintiff’s father by a social worker in Tallaght. An investigation was commenced by the gardaí and eventually the plaintiff’s father’s surrender was sought, he returned voluntarily from England, and was tried and convicted in the Central Criminal Court. However, on appeal, the Court of Criminal Appeal quashed the conviction and ordered a retrial. Thereafter, the accused man sought judicial review prohibiting the retrial on grounds of blameworthy prosecutorial delay. The application succeeded in the High Court and that decision was not appealed. Accordingly, the prosecution came to an end.
3. The High Court Judge had regard to the fact, described by him as extraordinary, that records had been mislaid and that there was no system to track and date records. The explanation, such as it was, offered for the delay was that papers had been lost somewhere in the prosecution service and subsequently located in archives. It was also said that the two officers in the investigation/prosecution service had died. The plaintiff had been assessed in 1999 for the purposes of a victim impact report in respect of the first trial and had been described then as relaxed, pleasant and good humoured. However, it is contended that following the quashing of the conviction and/or the prohibition of the retrial, her condition had deteriorated significantly. It is alleged that she suffers from post traumatic stress disorder. She feels she has been denied justice.
4. In 2004, the plaintiff issued proceedings claiming damages for negligence, breach of her constitutional rights and her rights protected by the European Convention of Human Rights (“the Convention”). By 2009, the case had only reached the stage where discovery was being sought, which was resisted by the defendants. The application for discovery precipitated a motion by the defendants seeking the trial of preliminary issues of law. On consent of the plaintiff, a preliminary issue was fixed, and subsequently heard, in the High Court. On the 20th January, 2011, the High Court held that on the facts alleged, An Garda Síochána owed no private law duty to the plaintiff to avoid causing her distress. In doing so, the High Court followed a decision of the High Court in W v. Ireland, The Attorney General and The Government of Ireland (No 2) 2 IR 141 (“W v. Ireland”) and persuasive UK authority Hill v. The Chief Constable for West Yorkshire AC 53 (“Hill”). Furthermore, judgement had been delivered in the case of BL in the period between the hearing of the preliminary issue and the judgment in this case, and the judge considered that the two decisions were consistent. Accordingly, the order of the High Court recorded that the claim was dismissed in its entirety.
5. It is useful at this point to mention briefly something of the facts of W v. Ireland and Hill, both of which are striking and which illuminate the legal issue to be determined. W v. Ireland involved a claim for damages by alleged victims of the late and notorious Fr. Brendan Smyth, who contended that they had suffered psychological injury by reason of the much publicised delay in securing his extradition to Northern Ireland. The High Court held, on the trial of a preliminary issue, that the defendants in that case (the Attorney General and the Government) did not owe a private law duty of care to individuals, including alleged victims of crime, when performing executive and statutory functions, and accordingly dismissed the claim. In Hill, the family of the last victim of the notorious Yorkshire Ripper, Peter Sutcliffe, brought proceedings arguing that if it had not been for a woefully inadequate police investigation, Sutcliffe would have been apprehended long before the incident in which she was attacked. The United Kingdom (“UK”) House of Lords held that the police did not owe a general duty of care to individual members of the public to identify and apprehend a violent criminal and, indeed, that the imposition of such a private law duty of care in all cases could be incompatible with the public duty owed to the general public.
BL v Ireland
6. In this case, the plaintiff alleges that she was the victim of a rape by a man, JW. The alleged perpetrator was arrested in July, 1999, and the trial commenced. Inculpatory statements made by JW while in garda custody were excluded from evidence on the basis that he had been unlawfully arrested. The arresting gardaí had invoked a power at common law to arrest. However, it was successfully argued that after the Criminal Law Act 1997 and the abolition of the distinction between misdemeanours and felonies, there was no longer any common law power of arrest, although there remained a statutory power of arrest under s.4 of the Criminal Justice Act 1984, but this had not been invoked. In consequence of this ruling, the inculpatory statements were excluded from evidence and the plaintiff contends that JW was acquitted. How precisely that result occurred or is alleged to have occurred is, surprisingly, still unclear at this stage. Counsel for the state parties informed this Court that on his instructions there had been a directed verdict on the exclusion of the statements. However, counsel for the plaintiff informed this Court that her understanding was that the trial had proceeded but was confined to the oral evidence of the complainant and the accused, who was acquitted by decision of the jury. In either case, the plaintiff contends there was a basic error, and therefore negligence on the part of the arresting garda, for which it is said the state parties are responsible. The outcome of the error in arrest was, the plaintiff claims, that she did not have the benefit of cogent inculpatory evidence when the criminal trial of JW proceeded, and she contends she suffered distress, loss and damages as a result.
7. Proceedings in this case were commenced in 2006 and, like those in LM, progressed at a pace which could charitably be described as leisurely. In 2010, an application was brought by the defendants for an order directing the trial of a preliminary issue. It is reasonable to infer that the fixing of the trial of a preliminary issue in LM led to similar procedural steps being taken in this case. It appears that on the hearing of the motion to fix the trial of a preliminary issue, however, both parties argued the substantive issue. On the 10th December, 2010, Kearns P. delivered judgment and held that the defendants did not owe a private law duty of care to the plaintiff giving rise to a claim for damages in respect of the erroneous invocation of the power of arrest.
8. It is clear that there are considerable similarities between these two cases (and superficial similarities with W v. Ireland and Hill) in the sense that they both involve claims for damages which are dependent upon a contention that public bodies, which carry on important public functions of investigation and prosecution of crime, owe a duty of care to individual members of the public giving rise to an obligation to compensate them in the event of any failure to perform those functions adequately. Accordingly, the cases were listed together in this Court for hearing on the same day.
However, there are also important distinctions between all of these cases. LM involves an investigation which was, in effect, not carried out at all, whereas BL involves an investigation which was carried out, though with an error. The negligence alleged in BL is claimed to have resulted in an acquittal by a jury after a trial in which evidence was given. The claim in LM is that by reason of prosecutorial negligence a trial was prevented. There are also important distinctions from cases such as Hill. That case involved claims of negligence at a level of planning and overview, and which continued over a period of time. BL is focussed on a single error made by an individual officer. LM involves errors and inaction on the part of both police and prosecution authorities at an operational level. W v. Ireland concerned a claim in respect of the performance of important executive functions relating to the surrender of an individual to another state. In all of these cases it can be said that there are three actors: a victim, an alleged perpetrator and the prosecution authorities. In Hill and cases of that nature, the claim is made in respect of the injury caused by the perpetrator for which it is said the prosecution authority is also responsible. In cases such as BL and LM, however, the claim is not in respect of damage done by a third party, which it is alleged would have been prevented by proper investigation, but rather is injury to arise directly from the distress said to follow from a failed prosecution. The damage is different, and logically distinct, from the damage caused by the original wrongdoer. One very important issue raised by these cases, and in particular LM, is that the alleged perpetrator of what might be described as the primary injury to the Plaintiff, has been the subject of criminal trial , and in the case of LM acquittal by a jury. At one level it may be said that the Plaintiffs claim involves a contention that but for the alleged negligence the accused person would or could have been found guilty. It might be thought that very substantial issues of public policy may arise in such a case, which would have to be addressed. Indeed it is conceivable that other parties would have to be heard in relation to the implications of such a claim. I do not suggest that all of these distinctions are necessarily critical. However, they do require to be noted at this stage.
Issue of Law
9. The issue of law raised in these cases is one which should be immediately recognisable to any law student. For many years it was a standard feature of university introduction to the common law legal system to require students to follow and analyse the development of the law of negligence both prior and subsequent to the landmark decision in Donoghue v. Stevenson AC 562. Decisions of the Irish courts have, of course, made a significant contribution to that development, for example through cases such as Glencar Exploration p.l.c. v. Mayo County Council (No 2) 1 IR 84 and in the area of vicarious liability in O’Keeffe v. Hickey and Others 2 IR 302. It would, however, be an absurd and narrow chauvinism not to recognise that the speech of Lord Atkin in Donoghue v. Stevenson AC 562 has been significantly influential in our jurisprudence as well as that of other common law countries. In considerably oversimplified terms, prior to 1932, the tort of negligence was little more than a loose collection of cases and areas in which it had been held that an action would lie. In Donoghue v. Stevenson, Lord Atkin, for the majority, discerned a unifying principle put in terms both elegant and memorable. Relying on the biblical injunction to love thy neighbour, he deduced a common law obligation to refrain from causing physical harm to anyone who, in law, could be one’s neighbour. This was someone who was in sufficient proximity (legal if not factual) to the alleged wrongdoer so that it was reasonably foreseeable that they would be damaged by carelessness on the part of the actor.
10. The ease with which the neighbour principle can be stated and understood has been both its strength and its weakness in the years following 1932. The conceptual issue which came to be debated was whether the tide created by the authoritative identification of the unifying principle underlying the law of negligence would overrun those areas and claims where it had hitherto been believed, and perhaps decided, that no action in negligence arose. As attempts were made to apply the neighbour principle outside the area of physical injury, it began to be speculated that a widely applicable tort of negligence based on the neighbour principle would also overrun and render redundant existing torts, such as nuisance, the rule in Rylands v. Fletcher, and misfeasance of public office, and might also threaten the limits of the law of contract. As a result, generations of law students became familiar with a number of areas in which it is said the traditional rule of non-actionability was under threat from a widely stated neighbour principle.
11. Among those areas were cases where damage was caused as a result of a negligent misstatement, where the loss was purely economic and unrelated to any physical damage, where it was contended that a separate liability in negligence could arise from a professional relationship created by contract, where damage was caused by nonfeasance rather than misfeasance, the liability in negligence of advocates, and the negligent liability of public authorities in the exercise of public functions. The consideration of the expansion of the conceptual field of the tort of negligence (of interest perhaps only to appellate judges, academic lawyers, and students) occurred in parallel with a significant increase in industrialisation and prosperity which meant that accidents, both industrial and road traffic, which are determined largely by the rules of negligence, came increasingly to dominate the business of the civil courts.
12. Again, and at the risk of oversimplification, the story for some time following Donoghue v. Stevenson was of steady expansion as apparently anomalous areas of non-liability fell before the elegant logic of Lord Atkin’s neighbour principle. Hedley Byrne v. Heller AC 465, and Dorset Yacht Company v. Home Office AC 1004 were important landmarks in the apparently linear progress of the principle. In 1978, the UK House of Lords appeared to have made a further advance, this time in the field of liability for the exercise of public law functions by a public authority. In Anns v. Merton London Borough Council AC 728 (“Anns”), it was stated that the question of the existence of a duty of care could be approached in two stages: first, was there a sufficient relationship of proximity between the claimant and defendant so that it was reasonably foreseeable that carelessness by the defendant could cause injury to the plaintiff? If so, there was a prima facie duty of care, and the court should then consider if there were countervailing policy considerations which ought negative, reduce or restrict any duty of care.
13. However, the Anns formulation attracted criticism. The expansion of the boundaries of the tort, coupled with a considerable rise in tort litigation, led to an increasing awareness that with the expansion in the area of the field of negligence came significantly increased costs. One significant development was when the High Court of Australia in Sutherland Shire Council v. Heyman (1985) 157 C.L.R. 424 refused to follow the approach of the UK House of Lords in Anns and took a more incremental approach, recognising that different areas could have different factors at play. Although at first sight the three step test in Sutherland Shire might not appear significantly different from the two step test in Anns, it was a very important development which in due course influenced the law in other common law countries, and in particular the UK and Ireland.
14. For present purposes, the significance of these developments is that the question of the existence of a duty of care now involves a much greater focus on the distinctive features of the area in which a duty of care is sought to be imposed. The questions involved in permitting a private law action for carelessness in the performance (or non-performance) of public law functions in general are complex. On the one hand, Lord Atkin’s generation could not have anticipated the dramatic expansion in the fields of activities carried out by the State generally, certainly in the years up to the 1980s. Nor could they have anticipated the extent to which, particularly in later years, functions hitherto considered to be core functions of the State have been privatised. It is sometimes said, and was said on this appeal, that the question which arises is whether a public authority should be immune from liability for negligence which causes foreseeable harm when a private person or body would not. However, it is also argued that this is not a helpful way to frame the question. No private person is, for example, permitted (or required) to exercise police powers. Public authorities do not carry out those functions for private gain or out of personal choice, and are not permitted to avoid their responsibilities because of the risks involved. As Lord Toulson put it in Michael v. Chief Constable for Wales 2 WLR 343 at para. 114, echoing earlier observations of Lord Hoffman in Stovin v Wise:
“It does not follow from the setting up of a protective system from public resources that if it fails to achieve its purpose, through organisational defects or fault on the part of an individual, the public at large should bear the additional burden of compensating a victim for harm caused by the actions of a third party for whose behaviour the state is not responsible.”
The conceptual issues can be particularly acute in the field of policing, crime prevention and crime prosecution. On the one hand, the public policy objectives in pursuing criminality are important and anything that could interfere with that task, and the resources necessary to perform it, is to be avoided. On the other hand, the public, and private, harm caused by police failure is very substantial.
15. Since the decision in Hill, there has been a large number of cases in the UK on the general area of the private law duty of care of public authorities and also in the specific context of crime prevention, prosecution and investigation. Some cases have sought to explore whether the general principle of non-liability in Hill is subject to exceptions, qualification or distinction if the error is said to arise due to operational risk rather than policy choices, or where it is alleged the negligence passes some higher threshold of gross negligence or default, or where it is suggested that the particular fact situation gives rise to a limited and targeted duty to an identifiable person. See by way of example Rigby v. Chief Constable of North Hamptonshire  1 W.L.R. 1242, Brooks v. The Metropolitan Police Commissioner 1 WLR 1495 and, Chief Constable of Hertfordshire Police v. Van Colle and Smith v. Chief Constable for Sussex 1 AC 225.
16. These developments have occurred in parallel with, and have been influenced by, cases exploring the impact of the European Convention on Human Rights (“the Convention”) on domestic law. This arises both in the traditional manner of a complaint to the Court in Strasbourg about the legal protection available in a member state, and indirectly from the incorporation of the Convention in domestic law; in the UK by the Human Rights Act 1998 and in Ireland by the European Convention on Human Rights Act 2003. While there are subtle and important differences between these pieces of legislation, both impose an obligation on public authorities to respect the Convention and contemplate a claim for damages for breach of that duty.
17. In an early case, Osman v. The United Kingdom(2000) 29 EHRR 245, the European Court of Human Rights (“ECtHR”) treated the case law as establishing an immunity for public authorities which it considered was disproportionate and in breach of Article 6.1 of the Convention. This was met with polite but forceful criticism by the UK domestic courts, and in Z v. The UK(2002) 34 EHRR 3, the ECtHR acknowledged that the case law establishing that there was no duty of care in negligence in certain situations was, at least in principle, compatible with the Convention. However, it might be said that the outcome of these cases meant, and still means, that domestic tort jurisprudence in common law countries is capable of being tested against the Convention.
18. These developments have particular significance in the present context, since the ECtHR found, somewhat controversially, in O’Keeffe v. Ireland(2014) 59 EHRR 15, that there had been a violation of Article 3 of the Convention in that Irish law did not provide an effective remedy in respect of the plaintiff’s complaints of sexual abuse at the hands of a teacher. The controversial aspect of the judgment, articulated in the dissenting judgments, arose in part from the fact that the court appears to have assumed that no action lay in Irish domestic law either under the existing torts or by way of constitutional tort, and that it was therefore not necessary to exhaust national remedies in that regard. This was so even though that issue had not been squarely addressed in litigation in this Court. Cases involving a claim of liability of the State might involve a consideration of the assumptions made by the ECtHR in O’Keeffe v. Ireland and the decision in that case and its implications in Irish law.
19. In a different, though related, development, it appears to be accepted that an action will lie for damages under the Act of 2003 for breach by the State of its Article 2 obligation to protect life. This will be applicable only in those cases where it is alleged that police failures have led to the death of an individual, and thus, is not directly relevant here. However, in a relatively recent judgment, the High Court of England and Wales (Greene J.) has also held that an action for damages under the Human Rights Act 1998 will lie for breach of the State’s obligation under Article 3 to investigate torture and degrading treatment, which can arise in the case of investigation of crimes such as rape. In DSD & NBV v. Commissioner for Police of the Metropolis EWHC 436 (QB) (“DSD”), the Court had to consider claims brought by the victims of a notorious serial rapist who alleged that failure to apprehend the perpetrator was a breach of the State’s obligations under Article 3 of the Convention. Greene J. held that an action did lie in the circumstances of that case. This decision was appealed in Commissioner for Police of the Metropolis v. DSD & NBV EWCA Civ 646, wherein the Court dismissed the appeal. The decision of the Court of Appeal of England and Wales was not available when these cases were argued in this Court.
20. In an important recent Northern Ireland case, C (A Person Under A Disability) v. The Chief Constable of the Police Service of Northern Ireland NIQB 63, Gillen J. considered an application to strike out pleadings alleging that culpable failure to investigate a rape allegation caused extreme upset, distress and psychiatric injury to the Victim. In the light of the decision in DSD, Gillen J. refused to strike out a claim for damages under the Human Rights Act 1998. However, he also refused to strike out the claim in negligence because it was possible that the development of the law in negligence may be affected by developments in the field of the Convention, observing “I suspect that the last drop of ink has not been spilt on this issue”. This decision has some resonance in this jurisdiction, since the procedure and practice of the Courts of Northern Ireland is closely related to that which applies in this jurisdiction, coming as they do from a common source.
21. The developments of the law in the UK and elsewhere have been carefully reviewed and analysed in an important recent judgment of the UK Supreme Court, Michael & Others (FC) v. The Chief Constable of South Wales Police and Another 2 WLR 343 (“Michael”). This case involved a claim on behalf of the parents and children of a victim of a shocking murder. The victim, Ms. Michael, had contacted the emergency 999 number because her violent and aggressive ex boyfriend had assaulted her and had threatened to return to kill her. Because of inadequacies in the handling of the call, the relevant police service was not mobilised with sufficient urgency, and Ms. Michael was brutally attacked and stabbed by her ex boyfriend. He was apprehended, pleaded guilty to murder and was sentenced to life imprisonment.
22. Perhaps the central issue in Michael was whether there was an exception to the Hill principle which rejected a general duty of care on the basis of a limited and targeted duty to an identifiable individual arising on the facts of the case. This broadly followed the observations of Lord Bingham of Cornhill in Chief Constable of the Hertfordshire Police v. Van Colle and Smith v Chief Constable of Sussex Police 1 AC 225. It was also argued that the circumstances gave rise to an assumption of liability on the part of the police services.
23. In Michael the UK Supreme Court reviewed the law of a number of common law jurisdictions. Briefly, it appears that the Australian courts and the State Courts in the United States took broadly the same approach as that in Hill. In New Zealand, however, the Court of Appeal overturned a decision striking out such a claim. In Canada and South Africa it was possible to establish liability at common law for negligent police actions. The UK Supreme Court noted the two judgments under appeal, together with another decision of the High Court, as indicating that the Irish Courts had “consistently held that the police owed no private law duty of care in respect of the investigation and prosecutorial functions”. By a majority, the Supreme Court in Michael reaffirmed the Hill principle. However, Lord Kerr and Lady Hale dissented, and would have permitted a negligence claim to proceed on the basis that the particular facts created a duty of care to the victim. It is noteworthy that the Court in that case was only concerned with the common law claim in negligence. It was accepted that the Article 2 claim could proceed to trial.
24. I should emphasise that this is only the briefest outline of developments in the common law world and is not intended to indicate any settled, or even tentative, view on the many issues which arise. It is sufficient, however, to observe that the preliminary issue raised in both cases is one which has involved a considerable legal debate at the highest level in other common law countries, and a debate which, it might be thought, still has some distance to go. In this regard, it is particularly noteworthy that the decision in Michael has already generated considerable comment. It has been described as “arguably the third most important case after Donoghue v Stevenson and Hedley Byrne v Heller on the English law of negligence to be decided by the UK’s highest court”. See McBride, University of Cambridge Faculty of Law Research Paper 21/2015. Another commentator has suggested that it may well mark the beginning of a new era in duty of care jurisprudence in the UK and a willingness to break with Caparo: Goudkamp, “A Revolution in Duty of Care” (2015) 131 L.Q.R. 519. I mention these commentaries not to endorse them necessarily (since neither they nor the decision in Michael were available when the submissions of most of the parties were made in this case) but rather to illustrate the significance of the issues which are to be addressed in these cases.
25. These cases appear to have been argued in the High Court on the basis that they were to be determined by reference to the general principle illustrated in W and Hill. However as already mentioned these cases may themselves involve separate and important considerations which did not arise in Hill or Michael for example. Even if it were possible and appropriate to permit a claim for negligence in respect of a failure to apprehend someone who later commits a crime, and is later convicted of that crime, does any such principle extend to a situation where the alleged perpetrator has not been charged, or if charged whose trial has not proceeded, or most significantly of all, if a trial has resulted in a verdict of not guilty? Even if it were possible in general to claim damages against the investigative or prosecution services, (itself a difficult issue as the discussion above illustrates) would it be possible to do so in circumstances where any such claim might involve a contention that without such negligence the accused, would or could, have been found guilty? It should be clear that very large questions of public policy could arise. Is such a claim permissible? What standard of proof would that require? Would the acquitted person be a necessary party? I mention these issues here in outline only and without any elaboration, because they have not been addressed hitherto, but any consideration of such issues, and any refinement of them, only illustrates the complexity and importance of these cases.
The Trial of a Preliminary Issue
26. These cases came before the High Court by applications made by the defendants to have preliminary issues of law tried to determine whether or not the defendants owed a duty of care in the circumstances of the case, and arguably, whether any actionable claim arose on the facts. In detailed and helpful written submissions on behalf of the defendants in BL, reference is made to a striking passage in the judgment of Lord Justice Jacobs in the Court of Appeal of England and Wales in M v. Commissioner of Police of the Metropolis EWCA Civ 1361. Referring to certain comments as to the caution with which a court should approach the trial of a preliminary issue, Jacobs LJ. stated at para 44:
“I would only add that the history of the tort of negligence shows that Lord Browne -Wilkinson’s words of caution should not be regarded as too prescriptive. After all many important authorities on the topic, and especially about whether or not there was duty of care, have been decided in strike out or preliminary point circumstances. Mr Beer provided a list, obviously not comprehensive (it is an interesting game to think of others), of House of Lords cases so decided:
The defendant’s submissions in BL argued that “these observations are equally applicable to this case.”
27. However, in my view, a consideration of the observations of Lord Justice Jacob reveals, if anything, a difference in the approach of the Irish courts and those of the UK. The same observation simply cannot be made about the development of the Irish law of negligence. The major recent cases in the Supreme Court, such as Ward v. McMaster  I.R. 337, Siney v. Dublin Corporation  I.R. 400, Glencar Plc v. Mayo County Council (No 2) 1 IR 84, and O’Keeffe v. Hickey 2 IR 302, are examples of cases which went to trial and were not determined on the trial of a preliminary issue. Indeed, the difference in outcome and approach can be illustrated by the fact that one very substantial piece of litigation in the recent past involved separate claims against the auditors of the failed Insurance Corporation of Ireland, brought by the liquidators of the company, and the purchasers of the company. That raised, of course, the question of the liability of auditors of a company to purchasers of a company who relied upon the audited accounts in purchasing the shares in the company. An application for a trial of a preliminary issue was, however, rejected. Subsequently, the House of Lords of the UK decided the same issue in Caparo Industries plc v. Dickman 2 AC 605, and did so on a preliminary issue
28. It is possible to speculate on reasons why there is such a significant divergence in practice between the two jurisdictions in an area where both the law and procedures have such marked similarities, but it is impossible to deny that the difference exists. Some of this may be traced to cultural differences, and a preference in Ireland for the full oral hearing before a court. It is also the case that until relatively recently, Irish courts did not experience many lengthy and elaborate claims. Some of the differences, I suspect, may also be traced to the existence, during the period most of the cases referred to were decided, of a generous civil legal aid system in the UK. On the other hand, plaintiffs’ actions for personal injuries and other claims in Ireland have often been dependent upon the willingness of a legal team to advance a claim on the hazard that if the claim failed they will receive no payment. The imbalance in resources that can sometimes exist may lead a court, sensitive to the risk that a claim with some merit may founder in an interlocutory war of attrition, to a disposition to permit a claim at least to go to trial, where the merits of the claim may be more apparent than at an earlier stage. It is also, perhaps, the case that pleadings in Ireland have tended to be more general, not to say vague, than in other jurisdictions. If a preliminary issue is to be useful in disposing of the case, there must be a degree of precision about what the case involves, which at least, as a matter of history, has sometimes been lacking in this jurisdiction. Without such precision, preliminary issues can be mired in a fog of speculation and become rather futile exercises in legal ingenuity in hypothesising possible claims, rather than an identification and determination of the claim.
29. Nothing in these observations is intended to suggest that one approach is preferable. Indeed, much of the recent reform of civil procedure in the England and Wales appears to have sprung from dissatisfaction with a concentration on over-refined pleadings and a proliferation of costly interlocutory procedures. By the same token, developments in this jurisdiction have tended towards a welcome tightening up of pleadings and procedures, and an increasing recognition that large and unwieldy litigation must be brought under control. It is possible, therefore, that while practice in both jurisdictions may not have converged, the difference may be less than was the case in previous years.
30. I should emphasise that I do not start from any a priori hostility or scepticism to the trial of preliminary issues. There are, of course, some cases where it will be apparent that little, if anything, will be gained by trying an issue before a hearing which may dispose of all issues after a full hearing, and in a relatively similar time scale. On the other hand, there are some cases, particularly with substantial claims, where there will be considerable savings in time and therefore cost, in adopting an approach which seeks to isolate issues capable of being determined in advance, and which may have a substantial impact on the scope of the proceedings. This will particularly be the case where it is apparent that any trial will be lengthy and costly, and where the preparation necessary may be protracted and expensive. There are also circumstances where a novel claim is brought by a number of persons in a similar situation, and where the trial of a single issue may be dispositive.
31. There are, therefore, in my view, circumstances in which justice may well require the trial of a preliminary issue. It is unrealistic not to recognise that much, if not all, substantial litigation is brought against parties which have the resources to meet any award of damages. It is also not uncommon for plaintiffs in such claims to be unable to discharge any award of costs from their own resources. Discovery, although available to all parties, will often bear more heavily on defendants against whom allegations are made than on the party making the allegation. Where a claim is extensive, and is brought by a plaintiff not able to satisfy any award of costs, and where discovery and consequent preparation of evidence is extensive and costly, the economic and commercial logic of settling such claims may become pressing. Discovery in particular is intrusive, and litigants who are subject to the process are understandably doubtful that any information disclosed can be limited to its immediate recipients and used only for the purpose of the litigation. They may also anticipate a hearing where sensitive matters will be focused on and ventilated. Again, this increases the pressure on a defendant, in particular, to compromise the proceedings. In some cases, particularly where there is some possible substance to a plaintiff’s claim, this may be a form of justice, even if rough at times, but in other cases the pressure to settle caused simply by the length and scope of the proceedings, the breadth of the claim, the cost (financial and otherwise) of making discovery and the absence of any possibility of early determination may be perceived as something less than justice. In such cases, the trial of a preliminary issue may be entirely appropriate and indeed required. In other cases, the trial of a preliminary issue may simply be a sensible course for all parties.
32. It is, as a general matter, important that the point sought to be tried as a preliminary issue should have the possibility of either terminating the claim altogether or at least resulting in an obvious saving in both costs and time consequent on a reduction of the issues to be tried. A point should also raise a clear issue to which it is possible to give a clear answer. The more qualified and contingent the possible answers, the less likely that the court will be able to provide a clear and decisive disposition of the case and a clarification of the law. The decision to direct a trial of a preliminary issue is therefore one which requires careful consideration by trial judges. It is important that judges do not too readily accept a respondent’s protestations of complexity, impossibility or inconvenience in trying a preliminary issue, while at the same time interrogating with some scepticism a moving party’s claim that the point is clear and potentially dispositive of the litigation or some significant portion of it.
33. In this case, it might be said (and was said) that these observations are merely academic since in neither case was the question of whether or not the preliminary issue should be tried determined by the High Court, and in any event, it was not the subject of an appeal. In LM, the fixing of the issue was not contested and may have been ordered on the consent of the plaintiff’s representatives. In BL, the issue was not determined at all. Instead, the substantive issue was tried and determined on the motion to fix a preliminary issue, apparently with the agreement, either express or tacit, of the plaintiff’s representatives. Nevertheless, when this matter was discussed in this Court, counsel for the appellant in LM sought leave to extend time to appeal the order which had fixed the trial of a preliminary issue (and which, on the face of the order at least, appeared to have been made by consent). I am quite satisfied that the plaintiff cannot be permitted to extend time to appeal the decision, both after this lapse of time and at this point in an appeal, particularly when the plaintiff cannot bring herself within any of the criteria in Eire Continental Trading Company Limited v. Clonmel Foods Limited  I.R. 170, did not seek to appeal at the time or even shortly thereafter, and where it appears that the order was made on consent.
34. However, I also consider that a court is entitled, on the hearing of the preliminary issue, to consider if it is an appropriate case for determination by this procedure. If, for example, the court proceeded to hear and seek to determine the preliminary issue after a full and elaborate argument, it would, as I conceive it, still be open to the court to conclude that in the light of the arguments and the matters advanced, that it was not possible to give the sort of clear and unequivocal answer to the issue which would dispose of the case or any issues in the case. Therefore, the case should proceed to trial to have issues of law determined in the concrete and precise circumstances of an individual case. Indeed, counsel for the defendants in these cases conceded that this could be done in an appropriate case, but I do not wish to rest this decision, particularly in the context of this case, on any such concession. In my view, a court retains power to refuse to determine a preliminary issue if, after careful analysis, it becomes apparent that some aspect of the issue was heavily fact dependent, or that a possible outcome would be so contingent or qualified as to require almost a form of advisory opinion.
35. The possibility of taking such a course may be particularly appropriate in cases where the rationale of seeking the determination of a preliminary issue is to produce a precedent which may determine many other cases. It could be wrong, and possibly dangerous, to offer hypothetical views in the absence of the facts of other cases and, indeed, the parties to them. If, then, it is possible to conceive of circumstances in which after full argument a court might decide that it should not, or could not, determine the preliminary issue, then in my view, the court is entitled to consider at the outset of an application whether, on the materials advanced, it appears that the case is an appropriate one to determine for disposal as a preliminary issue. This is particularly so when the point raised is a major issue of law likely to affect many other cases. Normally these issues will be teased out in the course of an application to fix the trial of a preliminary issue, but where that is not done, (and perhaps even when an order has been made after full argument) the court trying the preliminary issue can conclude that it is not possible to dispose of the issue in a definitive way in advance of trial. Here, the uncertainty surrounding the manner in which the preliminary issue was fixed is, in my view, not merely a basis for the court being required to proceed to determine the issue, but is rather one further indicator of an unsatisfactory lack of precision in these cases making the trial of a preliminary issue inappropriate.
36. Having considered the matter, there are, in my view, a number of persuasive reasons why these cases are an inadequate and unsatisfactory vehicle for the determination of the important and complex issue of law raised.
37. First, I apprehend that the major value of the determination of the preliminary issue in these cases is for its effect as a precedent, rather than because in the individual cases there would be a significant saving in time and cost. It is not apparent that discovery here would be extensive, at least by the standards of modern litigation, or that the trial would involve considerable oral evidence. Nor would there be any particular novelty to the facts alleged. One of the features of both of these cases is that the background facts have themselves been the subject of trial in open court, and in one case a contested appeal. Furthermore, in the factual circumstances of these cases, it cannot be said that there is any value of confidentiality attaching to any further details about the circumstances in which each case terminated. Much is already known about these matters, and it seems unlikely that injustice would be done if the plaintiffs in each case became aware of further details of how these cases came to founder.
38. But whether these cases are approached as cases meriting a preliminary issue because of the saving of time and cost in respect of the cases themselves, or because of their possible value as a precedent in governing future cases, the question is considerably complicated by the possibility of the Convention claims. In so much as it is, in principle, possible that a claim for breach of the Convention may be made and may proceed, then that would be a factor tending against the determination of the preliminary issue on the question of the decision of the duty of care, since there would be little, if any, saving in either evidence or time in taking that course. Indeed, as Gillen J. observed in C v. Chief Constable of the PSNI NIQB 63, it is at least possible that the common law position may be influenced to by developments at the level of the Convention.
39. Here, it is said in LM that the Convention claim was dismissed (and the order made certainly suggests this) although there does not appear to have been a separate argument or analysis of this issue in the course of the High Court proceedings. Counsel for the appellant (who it should be said was not in the High Court) maintained, however, that the Convention issue was still live in this case. This confusion only multiplies the possible issues in the case. It can, however, be said that if the Convention issues are live, then the case will not be disposed of even by finding that no duty of care was owed, and there will be little reduction in evidence or discovery if the preliminary issue is determined against the plaintiffs. On the other hand, if the Convention issues are not live in this case, then even a positive result from the respondent’s point of view will be deprived of much value as a precedent since those issues will have to be addressed at some stage.
40. To some extent, the same may also be said of the claim for constitutional tort. It is of course the case that under the current jurisprudence of this Court, it is only necessary to consider the existence of a tort of interference with constitutional rights if the common law is inadequate. In this case, it may well be that if such an inadequacy is perceived, then the common law can be adjusted. But, once again, if the constitutional tort claim is still live, then little if anything will be gained by a determination of a preliminary issue as to the existence of a negligence claim since the same ground will have to be traversed in evidence, and much of the argument will cover similar ground. On the other hand, if the constitutional claims are no longer alive in these cases, then any determination in the defendants’ favour on the common law claims would lose much effect as a precedent, since it would not be possible to say that some claim would not lie on similar facts. Moreover, it is difficult to say, at least at this stage, that the general argument on these matters would not be better conducted when all issues could be decided at the same time.
41. Furthermore, if these cases are looked at for their value as precedents, then a number of further difficulties emerge. Reference has already been made to the surprising lack of clarity, even at this stage, as to the circumstances in which the prosecution in the BL case was terminated. This, and the confusion as to the existence of the Convention claims, is symptomatic of the significant lack of clarity in these cases. Again, and more significantly, it is at least possible that the court might conclude that the answer to the preliminary issue raised may not be a simple yes or no, but will be a qualified “yes but” or “no but”. Certainly that is part of the case made on behalf of the appellant. Even, therefore, if the Court were to endorse the determinations made in the High Court, it would be necessary to explain why a more qualified response could not be given. But the more qualified and contingent the determination of the Court, the less value it is as a precedent for disposing of other cases. Furthermore, if the Court might resolve the preliminary issue in a manner subject to qualifications, then there is at least a risk of injustice to cases not before the Court since any binding rule would not be decided in the light of those cases and the possible variants that they would pose. It is by no means apparent that these particular cases provide enough by way of factual variables or are sufficiently developed in fact to allow a comprehensive single determination to be given. It is also apparent that the pleadings in these cases lack the sort of precision that would allow legal issues of this magnitude to be determined. These cases are too simply unsteady a foundation to bear the weight of the substantial arguments sought to be constructed upon them.
42. While it would not be determinative of the question of whether a preliminary legal issue should be determined, it should also be said that the papers in this case do not facilitate the determination of an important legal issue. The parties have submitted to the Court, in addition to booklets of pleadings and motions, five lever arch files of agreed authorities organised in an entirely haphazard fashion which contained notable omissions both in terms of the decided authorities and the academic commentary. Only one learned article was referred to, although it is apparent that this is an issue which has received scrutiny at the highest level in a number of jurisdictions. At the hearing of this appeal, which was to finish in one day, a copy of a useful recent article in the Irish Jurist by Professor Dermot Walsh, “Liability for Garda Negligence in the Prevention and Investigation of Crime” (2013) Ir Jur 1, was handed into Court. This article comments on the two cases under appeal, and makes reference to the case law and academic commentary in other jurisdictions. It follows from its late delivery that neither it, nor the extensive literature referred to therein, were addressed by the parties in their submissions. Because of the timing of the delivery of submissions, there was also only limited reference to the important case of Michael v. Commissioner of Police. This Court has issued a practice direction, SC 16 - Conduct of Proceedings in the Supreme Court, in respect of cases being case managed which are certified for hearing in the Supreme Court, and which seeks to ensure that submissions are focussed, with all papers and authorities organised and presented in a fashion that facilitates analysis of the central issues of law. There is no reason why this template should not be followed in all important cases, whether at trial or appellate level.
43. In my view, it would be dangerous, and therefore wrong, to seek to address the important issues of law raised here in the context of the limited information and material before this Court. It may be possible that these cases could be gathered together with other cases raising the issues, and possible variants of them, and case managed to the point where they could be usefully determined. I do not by this intend to say that such a trial could not be conducted in modular fashion, or that it might not, at an appropriate stage, be possible to isolate legal issues and determine them.
44. The combined effect of the importance and complexity of the legal issue raised in these cases, the procedural inadequacies and confusions, the lack of factual and legal precision even after a number of years, the unlikelihood that even the bluntest answer will resolve all issues in such cases, and the possibility, at least, that the Court might not be able to offer more than a highly qualified, contingent or abstract answer, all lead me to the conclusion that the determination of the preliminary issue in these cases is an inadequate and inappropriate vehicle for the determination of the important issues raised. While the point raised is an important and absorbing one from the point of legal theory, it is also of particular importance at a human level, not least to the individual plaintiffs whose experience of the legal system in its broadest sense, cannot be said to have been positive. It is important that these cases are properly and fairly determined, which means in my view that these plaintiffs should be allowed bring their cases to trial. Accordingly, I would allow the appeals, set aside the orders of the High Court in each case, and direct that the matters proceed in the High Court.