Freely Available British and Irish Public Legal Information
[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
Supreme Court of Ireland Decisions
You are here: BAILII >> Databases >> Supreme Court of Ireland Decisions >> Ambrose v Shevlin [2015] IESC 10 (05 February 2015)
URL: http://www.bailii.org/ie/cases/IESC/2015/S10.html
Cite as: [2015] IESC 10
[New search] [Help]
Judgment
___________________________________________________________________________ | ||||||||||||||||||||||
THE SUPREME COURT [Appeal No: 13/2010] O'Donnell J. Clarke J. Charleton J.
Michael Ambrose Plaintiff/Respondent And
Patrick Shevlin Defendant/Appellant Judgment of Mr. Justice Clarke delivered the 5th February, 2015. 1. Introduction 1.2 It must be recorded that it is unfortunate that a dispute between neighbours, which occurred over 20 years ago, has given rise to such a level of dispute and acrimony. However, the Court must attempt to determine this appeal on the facts and the law. That being said, it is important to note that, when written submissions were exchanged between the parties, amongst the points made in response on behalf of Mr. Ambrose was a contention that a number of the legal issues sought to be raised on behalf of Mr. Shevlin in his written submissions related to matters which were not included as grounds in the notice of appeal. In that context, a motion seeking to amend the notice of appeal by the addition of some further grounds was brought by Mr. Shevlin, and was heard by the Court at the same time as the substantive appeal. The first issue which the Court has to determine, therefore, is as to whether it is appropriate to allow Mr. Shevlin to amend his notice of appeal. However, to properly understand the questions which arise in that context, it is necessary also to say something about the background facts and the trial in the High Court. 2. Background Facts 2.2 The next day was relatively dry, but it began raining the following evening. The trench which had been dug by the JCB and which led to the water subsiding the previous evening had been filled in, and the water levels rose again, flooding Mr. Ambrose's house. 2.3 It is also worth recording at this stage that the farm pass in question was constructed without planning permission. It would appear that, in the view of the relevant local authority, Monaghan County Council, planning permission was required, for that council served an Enforcement Notice on Mr. Shevlin on the basis of a contention that the construction of the farm pass amounted to unauthorised development. It was also said by Mr. Ambrose that the construction of the farm pass was defective in that it provided inadequate means of allowing water to pass under it (the pipe, of which mention has been made, was said to have been far too small, and no other means of water passage was said to have been included). 3. The Trial in the High Court 3.2 Next, it is important to note that the defence raised by Mr. Shevlin before the High Court was based entirely on a denial by him of some of the central facts asserted by Mr. Ambrose. In particular, it was denied that the existence of the farm pass had actually caused the flooding to Mr. Ambrose's property. In addition, it was denied that Mr. Shevlin had been responsible for the filling of the breach between Friday night and Saturday night. No serious legal issues appear to have been raised. Indeed, as the trial judge noted at p. 6 of her judgment "the central issue … is whether the flooding to the Ambrose home is attributable to the farm pass on the Shevlin land". Having reviewed that evidence, the trial judge came to the following conclusion, at p.11:-
3.4 Having noted the agreement of the parties on the quantum of damages, the trial judge indicated that it was not possible for her to make any distinction between the damage which had been caused on each respective occasion and, therefore, gave a decree in the agreed sum. Against that background, it is next necessary to turn to the appeal as originally brought. 4. The Original Appeal 4.2 That issue is before the Court, and it will be necessary to turn to it in due course. In substance, the case made by Mr. Shevlin is that, in circumstances where he can, as a consequence of the trial judge's findings, only be liable for damages arising from the Saturday night flooding, it is necessary to distinguish between the damage caused by the two respective events. His primary submission is that there was no evidence from which the trial judge could reach a conclusion as to any specific damage caused on the Saturday night and that, on that basis, Mr. Ambrose's claim should be dismissed. His fallback position is that there should be a retrial. While the original notice of appeal puts this point in a number of different ways, in substance the ground of appeal is as I have described it. 4.3 For his part, Mr. Ambrose argues that the agreement by the parties on the quantum of damages (which did not seek in any way to distinguish between damage caused on the respective nights) means that the case ran on the basis that, if liability was established, damages in that sum would necessarily have to be awarded. I use the term liability to include the factual question of whether the existence of the farm pass actually caused the flooding in question. 4.4 It is further said that no argument was put forward on behalf of Mr. Shevlin at the trial which suggested that it might be necessary to seek to separate the damage resulting from the respective incidents. On that basis, it is argued that the failure of Mr. Shevlin to raise such an argument at the trial in effect deprived Mr. Ambrose of the opportunity to lead such evidence as he could have directed to the question of segregating the damages. On that basis, it is said that it would now be unfair either to visit Mr. Ambrose with a dismiss or a retrial on an issue which was not raised by Mr. Shevlin in the High Court. 4.5 However, it is first necessary to address the question of whether Mr. Shevlin should be given leave to amend his notice of appeal. In substance, two substantive grounds are sought to be added to the notice of appeal. The first derives from a line of case law including Gerrard v. Crowe [1921] AC 395, from which it is said that a principle of law emerges to the effect that a landowner is entitled to protect his lands against flooding even if it is foreseeable that the consequence of his protection may be to increase flooding on some neighbouring lands. To take a simple example, by way of illustration, one might consider landowners on opposite banks of a river which is liable to flooding after heavy rain. It may be fairly obvious from the lie of the land that the building by one of the relevant landowners of a barrier on his side of the river (for the purposes of protecting his lands from such flooding) is likely to lead to even greater flooding on the far side of the river given that the water has to go somewhere. Mr. Shevlin now wishes to argue that he was entitled, as a matter of law, to have his lands protected from flooding, and that Mr. Ambrose's claim should not have succeeded even though it was found by the trial judge that the building of the farm pass had actually caused the flooding on Mr. Ambrose's land. It was accepted that there was no evidence to suggest that the farm pass had been built for the purposes of protecting Mr. Shevlin's lands from flooding. However, it was argued that this did not matter, for the principle was said to be such as would protect Mr. Shevlin from liability if his construction of the farm pass had the effect, even if not the intended effect, of protecting his lands. 4.6 The second ground sought to be argued concerned a contention that the trial judge had failed to consider whether the building of the relevant land pass was or was not a reasonable use of Mr. Shevlin's lands. This was sought to be argued as a fallback position in the event that the earlier argument sought to be raised, relying on the entitlement to defend against floods, was unsuccessful. In that eventuality, it was argued that the authorities both in this jurisdiction (for example, Fitzpatrick v. O'Connor (unreported High Court, Costello J., 11th March, 1988)) and in the UK (Home Brewery v. Davis & Company [1987] Q.B. 339) make clear that a landowner is entitled to make reasonable use of his own lands so that adverse consequences for a neighbour of such reasonable use is not actionable. Indeed, the trial judge noted those principles by reference to the cases just cited in her judgment. It was sought to argue that the trial judge should have gone on to reach findings concerning the reasonableness or otherwise of the building of the farm pass. 4.7 It was argued on behalf of Mr. Shevlin that, if he was correct on either or both of those points, it would transpire that judgment had been given against him on a wrong legal basis. In those circumstances, it was said that the interests of justice required that he be allowed to argue those grounds on appeal notwithstanding the fact that they were not included in his original notice of appeal. 4.8 In response, counsel for Mr. Ambrose made the following points:-
(b) In particular it was suggested that, even if the relevant issues had been included in the original notice of appeal, it would be inappropriate for this Court to allow Mr. Shevlin to rely on those grounds, for, it was said, they were issues not raised in the High Court; (c) It was argued that, even if Mr. Shevlin persuaded the Court that there was any potential merit in either or both of the points sought to be raised (which was very much not accepted by counsel) the only fair consequence would be a retrial. It was said that there would be no reality to a retrial which would require the court to consider events which would, by the time any retrial came on for hearing, be more than 21 years in the past.
4.11 The reason why O'Donnell J. identified a spectrum of cases in that regard is that there are two competing considerations with which the Court must be concerned. On the one hand, so far as the interests of the parties to the case in question are concerned, there is the desirability that a person who is entitled to win a case should not be faced with the prospect of losing it because a valid and decisive point was not made at trial. That consideration would, obviously, point in favour of allowing new grounds to be raised on appeal. 4.12 However, on the other side of the equation, there are very real dangers in adopting a practise which is generous in permitting new grounds to be raised. First, there is the overall desirability that parties be required to make their full case at trial. An overly generous approach to permitting new grounds to be raised for the first time on appeal can only encourage either sloppiness or imprecision in the way in which cases are run, or, indeed, attempts to take tactical advantage by only bringing forward a part of a claimant's true case in the knowledge that there will be a good chance that, if it does not work at trial, a different tack can be adopted on appeal. A culture of tolerance of parties who fail to bring forward their full case can only encourage such practises, leading to a significant risk of injustice across a whole range of cases and, indeed, causing difficulties for the administration of justice generally by a proliferation of unnecessary appeals and, in many case, retrials. It must be emphasised that the possibility of an injustice to an individual party in a specific case may be more easily identifiable than the real and equally important risk of injustice to a whole range of parties to many cases, to which a policy of excessive tolerance might give rise. 4.13 In addition, there can be a very real risk of injustice to the opposing party if new grounds are permitted to be raised for the first time on appeal. Precisely how a case was run at trial may well have been influenced, in ways that may not be completely obvious, by the issues which seemed to be before the court. All litigation (or at least almost all litigation) involves some degree of tactical decision by the advisers of both parties. Such tactical decisions often involve weighing up the pros and cons of a particular course of action. Those decisions are taken on the basis of the case as it reasonably appears to stand at the time the decision is taken. The more easily a party can reinvent their case on appeal, the greater the risk that real prejudice will be caused. Sometimes a party will be readily able to point to such prejudice. For example, witnesses may not have been called precisely because an issue was not before the court. Areas of evidence or lines of cross examination may not have been pursued. Legal issues may not have been raised. Many other examples could be given. But it is, in my view, important for a court to keep in mind that the sort of prejudice of which I speak can be subtle and difficult to demonstrate but nonetheless real. To all of that needs to be added the real risk that, in many cases, the only way of avoiding a risk of prejudice will be to direct a retrial. Such a course of action inevitably delays the completion of the relevant litigation and exposes the parties to much greater cost. 4.14 For all of those reasons, there are significant countervailing factors which lean heavily against allowing a point to be raised for the first time on appeal. The reason why there is a spectrum of cases, as identified by O'Donnell J. in Lough Swilly, is because the balance between those two competing factors may give different results depending on the type of case concerned. Where, for example, there is "a new legal argument" … "closely related to arguments already made in the High Court"... "and which is not in any way dependent upon the evidence adduced", then prejudice may be non-existent or slight, a retrial unnecessary, and the strength of the argument which concerns encouraging people to put their whole case before the court may be reduced. In such a circumstance, the justice of the case may permit an argument of the type identified to be raised for the first time on appeal because the interests of seeking to do justice on the facts of the individual case may outweigh the other weighty considerations which I have sought to identify. On the other hand, a case which would "necessarily involve new evidence" will place much greater weight on the side of the equation which lies against permitting a new point to be raised for the first time on appeal. The risk of real prejudice will be significant. The prospects of a new trial difficult to avoid. The need to encourage a party to bring forward its full case at trial will carry more weight. Against that background, it is necessary to look at this case for the purposes of attempting to place it along the spectrum of cases identified in Lough Swilly. 4.15 In that context, Mr. Shevlin argued that there were sufficient findings of fact by the trial judge to permit this Court to assess whether the legal arguments now sought to be raised could be sustained. However, it must be noted, in that context, as counsel for Mr. Ambrose argued, that the way in which evidence is presented to a trial court and the way in which the trial judge assesses that evidence for the purposes of finding facts can frequently be quite dependent on the legal issues raised or the legal framework within which the court is invited to consider those facts. While it is true, of course, that facts are facts, and a trial judge could not find different facts just because of the legal issues raised, nonetheless the focus of both the evidence led and the judge's approach to that evidence can be quite different depending on how those facts fit into the legal framework within which the particular case is fought. 4.16 A judge may give a general description of background facts precisely because those facts may not appear to be of any great relevance to the issues in controversy in the case and are only necessary for the purposes of putting the case in context. On the other hand, the judge may set out facts in much greater detail, and may be much more precise as to the exact findings in areas where it is clear, on the case made by the parties, that those findings and the way in which the facts are characterised may matter so far as the result of the case is concerned. A description of a man walking along a country road without paying too much attention to his surroundings may be entirely anodyne in one case. However, if the man is knocked down by a car and there is a controversy as to whether he inadvertently wandered out onto the road, thus causing or contributing to the accident, then a very precise description of what exactly he was doing in the moments before the accident may be vital to the decision in the case. If, in the light of the way the case runs, the trial judge has a particular controversy in mind, then findings of fact central to that controversy are likely to be detailed and precise. If the description of the man walking up the road is purely introductory or background information, having no possible bearing on the outcome of the case, the description will be general, and may well lack precision precisely because precision was unnecessary. 4.17 Therefore, while it is true to state that facts are facts, it is also true to state that the way in which the facts are presented in a judgment will, quite properly, be very dependent on the issues which the parties choose to put before the court. There is the potential for significant danger, and thus injustice, if facts are plucked from a judgment which is delivered in the context of a dispute, which has been characterised in a particular way by the parties, for the purposes of making a very different argument based on legal issues which were not addressed to the trial court at all. 4.18 It is also, in the circumstances of this case, of some importance to recognise that the legal issues which Mr. Shevlin now seeks to raise are potentially important but complex ones. How do we treat, in the modern age, questions of flooding? Does planning have any bearing on the proper result of civil proceedings such as those with which this appeal is concerned? Does it matter, for example, that the farm pass in question was not built as a flood defence but was, arguably, one which had that unintended effect? Indeed, to what extent might it be said that it is true that the farm pass had that effect? To what extent might it have been possible to design the farm pass in a way which both protected Mr. Shevlin's lands from flooding but at the same time did not expose Mr. Ambrose's house to any greater risk? Would it matter if it were possible to so design the farm pass? What margin of appreciation ought be left to a landowner in such circumstances, and, in particular, is that margin of appreciation different in circumstances where a landowner has to take urgent action to protect his lands against a flood, on the one hand, or carries out a development which was not designed to provide flood protection at all, on the other? 4.19 These, and doubtless many other, questions may well arise in an appropriate case. However, it seems to me that it would be wholly inappropriate for this Court to attempt to make a definitive ruling on these very important legal questions in proceedings where those points were not argued at all in the High Court and where, therefore, the facts as found by the High Court judge were not in any way directed towards those issues. 4.20 Likewise, it seems to me that there is a very real likelihood that, had those issues been raised by Mr. Shevlin in the High Court, further evidence would have been likely to have been called on behalf of Mr. Ambrose to attempt to deal with those issues. It seems to me that this case, therefore, falls very much at the end of the spectrum identified by O'Donnell J. in Lough Swilly where a court should not allow an argument to be raised for the first time on appeal. 4.21 It must be strongly emphasised that a trial in the High Court is not a dress rehearsal. It is at the trial that the rights, obligations and liabilities of parties are to be definitively determined. There may be cases where, as O'Donnell J. pointed out in Lough Swilly, it may strictly speaking be the case that a particular point was not raised in the High Court, but where the relevant point does not involve any different facts to those which were relevant to the issues which were raised, and where the point may be regarded as a refinement of, or analogous to, one made at trial. In such cases, the justice of the case may require that a party should be allowed to adjust their case on appeal. To take an over-technical approach to the issues being raised for the first time on appeal would bring a disproportionate risk of injustice in the individual case. But to allow a party to reinvent their case on appeal in circumstances where major legal issues are sought to be raised for the first time, and where the facts which might be relevant to those legal issues were not necessarily fully explored in the High Court, would in itself be an injustice. 4.22 It would, in my view, be entirely impossible for this Court to deal in a proper way with either the factual or legal issues sought to be raised in the proposed amended grounds of appeal. The reason for that is that none of those issues were even touched on before the High Court, and the evidence heard by the High Court was not in any way directed to those issues. Mr. Shevlin chose to fight the High Court action on the basis of giving evidence to suggest that his farm pass had not caused the flooding to Mr. Ambrose and that, in particular, he had not personally or through anyone acting on his behalf, filled in the breach which released the flood waters on the Friday night. That was his choice. That was his evidence. It was rejected by the High Court in circumstances where there is now no appeal against those findings. But he must live with the consequences of having chosen to fight the case in that way. 4.23 Those consequences are that it would be wholly unjust to allow the case to be radically reinvented on appeal. Even if that were not so, the only way in which it would be possible to allow those points to be raised now would be to direct a retrial. To order, at this remove, that there be a full retrial of all liability issues in this case in relation to events which occurred 21 years ago would itself be a recipe for further injustice. In those circumstances I would refuse Mr. Shevlin's motion seeking to raise new issues on appeal. It follows that the only issue which the court needs to determine is the issue already averted to, which was included in the original notice of appeal. I now turn to that issue. 5. The Damage Issue 5.2 In that context Mr. Shevlin places emphasis on the principle identified in Performance Cars Limited v. Abraham [1962] 1 QB 33. The facts of that case were relatively straightforward but the legal point was, nonetheless, important. A defendant, by negligence, further damaged a car which had already been damaged in an earlier incident. Part of the plaintiff's claim against the defendant stemmed from what was said to be a need to respray the vehicle in question. The defendant said that the vehicle was already in need of respraying as a result of the first incident and that, in those circumstances, he could not be liable for any part of the claim relating to respraying. The amounts involved may well have made it worthwhile to litigate this point because the vehicle was a Rolls Royce. 5.3 Lord Evershed MR said, at p.40:-
5.4 There will, of course, be cases where the same damage is caused by two separate events involving different wrongdoers. In those circumstances, both wrongdoers are "concurrent wrongdoers" and will be liable for all of the damage caused, subject only to being able to recover from the other wrongdoer whatever proportion of that damage the court considers appropriate to apportion in the light of an assessment of the respective blameworthiness of the two wrongdoers concerned. But it is also clear that there may be cases where, even though there be two separate incidents involving closely connected damage, it may be possible to determine that certain damage was entirely attributable to one incident, and other damage entirely attributable to a second incident. 5.5 Performance Cars is a somewhat intermediate case where the later damage may be considered to be different but where, as a result of the first damage, a particular heading of loss had already been fully incurred. If, for example, the two respective incidents in Performance Cars had given rise to entirely separate damage, then there would have been no difficulty. What created the issue about respraying was the finding of fact of the trial judge that the car would have to have been fully resprayed as a result of the first incident anyway so that, whatever additional damage was caused by the second incident, it did not give rise to a new need to respray. That was so even though it might well have been the case that the damage caused by the second incident might have led to a situation where respraying was necessary if the vehicle had not already been damaged. 5.6 This case is, however, complicated by two factors. First, it is, of course, the case that the alleged wrongdoer in respect of both incidents was the same person rather than different people. However, for the reasons identified by the trial judge, Mr. Shevlin is not legally liable in respect of the first incident (and thus cannot be considered to be a wrongdoer in that respect) but is legally liable in respect of the second incident. It is true, of course, as the trial judge found, that Mr. Shevlin actually caused both instances of flooding. However, he is not legally liable in respect of one of them. This is, therefore, a situation where the court has to consider similar damage caused by two separate incidents which were, in a factual sense, the responsibility of the same person, but where that person is only legally liable in respect of one of them. At the level of principle I can see that Performance Cars might well have some application in a case such as that. 5.7 However, the second difficulty which arises in this case stems from the way in which the hearing before the High Court was approached. The parties agreed the quantum of damages. Neither party sought to suggest that it was necessary to attempt to separate or segregate out any losses attributable to the separate flooding incidents. It would appear that neither party suggested to the court that there was any factual or legal basis on which the court might find Mr. Shevlin liable in respect of one but not both of the incidents. That is, perhaps, not particularly surprising given the focus of Mr. Shevlin's defence, to which I have already referred. 5.8 On his behalf, it is now argued that Mr. Shevlin is entitled to succeed on this appeal because Mr. Ambrose has failed to discharge the onus of proof which rested upon him to demonstrate damage resulting from the second incident. To allow that point to succeed on appeal would, in my view, amount to a serious injustice. It is true that Mr. Ambrose did not seek to present evidence to the High Court which differentiated between the damage resulting from the two separate incidents. But the context in which that course of action was adopted needs to be acknowledged. The parties had agreed the quantum of damages without attempting any such division. There is nothing in the pleadings which suggests that any distinction was sought to be drawn by Mr. Shevlin (even as a fallback position) between the two incidents. It is true that Mr. Ambrose was, as a result of the way in which the case was pleaded in Mr. Shevlin's defence, put on proof of his claim and was required, therefore, to establish his damages. However, an agreement as to the quantum of damage (thus removing any necessity to produce any evidence concerning such damage) coupled with the absence of any suggestion on behalf of Mr. Shevlin at the trial that there was a potential distinction between the losses attributable to the respective incidents, led to a situation where, in my view, it was both reasonable and proper for Mr. Ambrose (and indeed the Court) to understand that, in the event that liability for nuisance was established, the Court was to award damages in the agreed amount. 5.9 As the trial judge noted, she was unable to distinguish between the damages caused by the respective incidents. But the reason why she was placed in that position was that the quantum of damages had been agreed without addressing that distinction, and no one at the trial suggested that there was any basis for distinguishing between the two incidents. What inference could be drawn from the agreement of the quantum of damages other than that the damages were not to be attributed in any distinguishing way as and between the two incidents? If any other position was to be adopted, then an agreement on the damage attributable to both incidents separately would need to have been reached. 5.10 Mr. Shevlin did not, of course, have to agree damages at all. In those circumstances, Mr. Ambrose would have to have established his damages. But if, having led evidence as to the overall damage, no questions were raised in cross examination to suggest that a distinction could or should have been made between the damage attributable to the respective incidents, and if no argument were raised at the trial which would have put Mr. Ambrose on notice that he should attempt to make such a distinction himself, then it would be manifestly unfair, on the run of the case in that way, to now allow, for the first time on appeal, a Performance Cars type point to be raised. 5.11 But the situation is not, in my view, any different where, as here, evidence of damage did not have to be called at all by reason of the agreement on quantum and where, again, nothing in the way in which the case was approached on behalf of Mr. Shevlin gave any suggestion that there would be a need to make a distinction in the damages attributable to the two incidents. 5.12 Furthermore, even if an appeal on the Performance Cars point were to be allowed, the only result which could fairly be put in place would be a retrial, for the running of the case would, in that context, have deprived Mr. Ambrose of the opportunity to seek to put forward whatever evidence he might wish to identify the damage attributable to the second incident. 5.13 There might also, had the issue been raised, have been questions such as those which arose in some of the asbestos cases, as to what the court should do when it is not possible to say with any great precision as to the way in which two or more incidents caused a particular item of damage. For example, at least so far as some flood damage is concerned, it may be that particular items might be capable of salvage or less expensive remediation after a first flood, but would have had to be replaced, or to be the subject of more significant remediation after a second. We just do not know. But the reason we do not know is because no evidence on this issue was called. The reason no evidence was called was that Mr. Shevlin chose not to put the question of any distinction between the damage attributable to the two incidents in issue before the court. He must now live with that choice. It is, indeed, like the situation which arose in relation to the legal issues, in respect of which he sought to amend his notice of appeal, in that it is perhaps entirely understandable that these points were not run at trial in circumstances where Mr. Shevlin chose to fully defend the case on the basis of asserting facts which the court ultimately rejected. 5.14 In those circumstances I am not satisfied that the appeal on the Performance Cars point is properly maintainable in circumstances where it was not argued before the High Court and where there would be significant prejudice to Mr. Ambrose deriving from that fact. Likewise, it is clear that the only way in which this matter could be dealt with, if the appeal were successful, would be by a retrial. For the reasons already addressed in the context of the application to amend the notice of appeal, I feel a retrial would be unreal at this remove. 5.15 It should also be noted that much of the difficulty which arose on this appeal can be traced back to the way in which the case was fought and lost by Mr. Shevlin in the High Court. It is somewhat incongruous to fight a trial on the basis that the farm pass did not cause the flooding, and that he did not fill in the breach, and then to turn around on appeal and say that the pass and the filled-in breach amount to a flood defence. It is permissible to defend cases on alternative, and even sometimes potentially contradictory, bases. Because of that, it is all the more important that radically different alternative bases be clearly signalled at trial. 6. Conclusions 6.2 There remains the original ground of appeal. I would dismiss the appeal on that ground on the basis, elaborated on earlier in this judgment, that it would now be an injustice to permit the question of a segregation of the damages attributable to the two flooding incidents to be engaged in, in circumstances where Mr. Shevlin agreed the amount of damages (without, of course, any admission of liability) without alerting either Mr. Ambrose or the Court to any possible argument which might have required a segregation of those damages. Furthermore, no such issue was raised at the trial. In addition, the only conceivable consequence of a successful appeal under that heading would be a retrial which, in the context of an incident which occurred 21 years ago, would bring its own serious risk of injustice. 6.3 I would, therefore, dismiss the appeal. |