Matthew & Ors v Sedman & Ors [2019] EWCA Civ 475

[2019] EWCA Civ 475Case No A3/2018/0285 IN THE COURT OF APPEAL (CIVIL DIVISION)Venue ON APPEAL FROM THE BUSINESS AND PROPERTY COURTS IN MANCHESTER,BUSINESS LIST (ChD)His Honour Judge Hodge QC[2017] EWHC 3527 (Ch)Venue Royal Courts of Justice Strand, London, WC2A 2LLDate Date: 20/03/2019
Before
LORD JUSTICE UNDERHILLVice President of the Court of Appeal (Civil Division)LORD JUSTICE IRWIN
Between
PETER MATTHEWSCOTT NIXON (as Trustees of the Will Trusts ofEVELYN HAMMOND)
Between
DIANA ROSE COOKSALLY ANN EVELYN SELBYCOLIN RICHARD HENRY CARTLEDGEPHILIP CARTLEDGE AppellantsBARRIE SEDMANTHOMAS WILLIAM HALLAMPETER JAMES ROBERTS Respondents
Jeremy Cousins QC and Christopher McNall (instructed by Steele & Son with Bagot Heyes Solicitors) for AppellantsClare Dixon and Nicholas Broomfield (instructed by Mills & Reeve LLP) for theRespondentsHearing Hearing date: 15 January 2019
Approved Judgment

Introduction

[1]This is an appeal from the judgment of HHJ Hodge QC sitting as a Judge of the High Court in the Manchester Civil Justice Centre on 27 November 2017. The case concerns the calculation of the limitation period for a claim against trustees, which itself turns on a failure to claim under a court-sanctioned scheme of arrangement.[2]Originally, three issues arose in this appeal. Firstly, as Ground 1 the Appellants argued that the learned judge erred as to the “Bar Date” under the material scheme of arrangement, when he concluded that 2 June 2011 was the Bar Date and as a consequence concluded that the period for a legal claim in respect of the failure to claim under the scheme began too soon. At the opening of the appeal the Appellants conceded this point.[3]Secondly, under Ground 2 the Appellants argue that the judge was in error as to the limitation date for legal action. This was the central issue before the judge and he formulated it as follows: “The question is this: when a cause of action is completely constituted at the very first moment of a particular day, does that day fall to be included when calculating the applicable six years' limitation period or does it fall to be excluded? More pertinently for present purposes, if a cause of action accrued at the very first moment of Friday 3rd June 2011, is a claim issued after Friday 2nd June 2017 brought after the expiration of six years from the date on which the cause of action first accrued?” (paragraph 2)[4]Thirdly, the Respondents sought to support the judge’s conclusion on an alternative ground, namely that the legal claim in question:
“… had been issued out of time because the Appellants still suffered actionable damage by reason of the Welcome claim not being submitted on or before 2 June 2011.”
However, Ms Dixon made clear that this argument no longer arose, given the Appellants’ abandonment of Ground 1.[5]It follows that we are now concerned only with Ground 2.

The Facts

[6]The learned

judge summarised the facts and the procedural history succinctly as follows:

“3. The background to the application presently before the court is as follows: On Monday 5th June 2017 the claimants issued a claim form in the Manchester District Registry of the Chancery Division claiming equitable compensation, damages for breach of trust, an order for the reconstitution of the trust estate, an account, and further or other relief, with interest and costs. The first two claimants, Mr Peter Matthew and Mr Scott Nixon, sue as the trustees of the trusts of the will of the late Evelyn Hammond. Mrs Hammond died as long ago as 27th December 1952 leaving a will dated 21st August 1948. The beneficiaries of the trusts arising under that will are the third claimant, Diana Rose Cook (as life tenant) and the fourth, fifth and sixth defendants (as the remaindermen of the will trusts). The three defendants – Mr Barrie Sedman, Mr Thomas William Hallam and Mr Peter James Roberts – were the trustees of the will trusts until their retirement on 1st August 2014 when they were replaced by the first and second claimants. At all material times, the defendants were professional trustees, being partners or employees of Forrester Boyd Chartered Accountants. 4. The principal asset of the trust comprised shares in Cattles plc. In 1994, Cattles acquired Welcome Financial Services Limited. By April 2004 the trust owned almost 162,000 shares in Cattles which, as of 5th April 2008, were valued at some £393,000. In April 2009, trading in Cattles' shares was suspended and in December 2010 both Cattles and Welcome commenced court proceedings for court sanctioned schemes of arrangement. In the case of the Welcome scheme, the supervisor was KPMG. As a result of an order made by Mr Justice Newey on 28th February 2011 the schemes of arrangement were sanctioned. That order was registered at Companies House on 2nd March 2011. The terms of the scheme of arrangement included provision for claims to be submitted by shareholders. 5. By clause 3.6, and subject to an exception not material hereto, in order to be entitled to any scheme payment, scheme creditors were required to submit a claim form on or prior to the 'Bar Date'. Clause 3.5.3 provided in terms that claim forms must be sent to the scheme supervisors to arrive on or before the Bar Date. By clause 3.9 (headed "Variation of Time Limits") the scheme supervisors had a power in their absolute discretion to extend time limits in exceptional circumstances outside the control of a submitted scheme claimant, but that power expressly did not apply to a failure to comply with the Bar Date. The Bar Date was defined as meaning the first business day falling three months after the 'Effective Date'. The Effective Date was the date on which the scheme became effective in accordance with clause 1.5.1. That was the date on which an office copy of the final court order was delivered to the Registrar of Companies for registration. As I have indicated, that was 2nd March 2011.”
[7]It is common ground that the “Effective Date” under the Welcome scheme was 2 March 2011, that being the date when Newey J’s Order was registered at Companies House. The scheme defined the “Bar Date” as “the first Business Day falling three months after the Effective Date”. It was common ground that applications could be made up until midnight on the Bar Date. 2 June 2011 was a Thursday and hence a “Business Day” under the scheme, and it is now agreed that the judge was correct in finding that the “Bar Date” was 2 June 2011.[8]If the day following the Bar Date - the day when the action accrued - falls to be counted, then limitation expired on Friday 2 June 2017. If not, then the six year period expired on Saturday 3 June 2017. Since the necessary act on the part of the Respondents was the issue of the claim form in the legal action, something which can only be done when the Court office is open, then it was common ground, following Pritam Kaur v S Russell and Sons Ltd. [1973] QB 336 (CA), that the final day for issue would be Monday 5 June 2017. That was the date of the issue of proceedings. Hence, issue would be in time.[9]As I have indicated, this was the central issue before the judge. On this issue the Respondents relied (successfully below) on the decision of the High Court in Gelmini v Moriggia [1913] 2 KB 549. That case turned on the limitation of an action relating to the time for payment under a promissory note. The time for payment of the promissory note expired on 22 September 1906. Proceedings were issued on 23 September 1912. Channell J found that the claim was out of time because the cause of action was complete at the beginning of 23 September 1906 and so limitation expired six years later at the end of 22 September 1912.[10]The critical passage relied on by the Respondents from the judgment of Channell J is as follows:
“… The other point is, what is the true rule as to the computation of the six years under that statute? An action cannot be brought until the cause of action is complete, and in all eases of contract the person who has to pay has the whole of the day upon which payment is due in which to pay; therefore until the expiration of that day an action cannot be brought because until then there is no complete cause of action. The result is that an action cannot be brought until the next day; but it can be brought on that day because the cause of action is complete at the commencement of that day. If the cause of action is not complete, the action cannot be brought. It therefore follows that that day is one of the days upon which the action can be brought. The words of the statute are “within six years next after the cause of such action or suit.”
Now the day after that on which the debtor's time for paying expires is, in my opinion, the date on which the cause of action arises, and on that day an action can be brought, and that day is the first of all the days in the six years. Therefore, assuming that the day upon which the action can be brought to be a Thursday, and the period for bringing the action to be a week, the creditor can bring it at any time up to and including the following Wednesday, but not the Thursday. And the same rule applies where the period, as under the statute, is six years. I do not think that the day on which the cause of action arises is excluded. It is the previous day which is excluded, i.e., the day at the expiration of which the cause of action becomes complete.” [emphasis added] (page 552)[11]This was the approach accepted by the judge. He said that he was “satisfied that the cause of action accrued at the first moment of 3 June 2011” (para 26). The judge accepted the distinction between such a case as the instant case (and that in Gelmini) as opposed to others. He concluded as follows:
“31. In my judgment, where it is absolutely clear that the cause of action arises at the very beginning of a particular day, that day should not be excluded from the calculation for Limitation Act purposes. At any moment during that day the claimant can bring a claim; and to exclude that day from the calculation for Limitation Act purposes would have the effect of giving him an extra day over and above the statutory limitation period for bringing the claim. I therefore accept Miss Dixon's argument that where the cause of action is complete at the very beginning of a particular day, you exclude that day for the purposes of calculating the limitation period. On that footing, the limitation period in the present case began on 3rd June 2011 and expired at the very end of 2nd June 2017. On that basis, the last day for issuing the claim form was Friday 2nd June 2017, and this claim is out of time.”

The Submissions in Summary

[12]Mr Cousins QC for the Appellants essentially makes the following points. First, there is long-standing authority for the proposition that the day on which an action accrues is not counted for limitation. Second, that authority encompasses cases of various kinds. Third, even in such a claim as this, a “midnight deadline” case, the claim does not accrue until after midnight and thus (however shortly) into the day following midnight. Fourth, it is established that, unless there are specific provisions requiring it to do so, the law does not take cognisance of parts of a day. Fifth, that established rule carries the benefits of simplicity and clarity. Sixth, the decision in Gelmini has been disapproved, and disapproved on this point, both at first instance in Marren v Dawson Bentley & Co Ltd [1961] 2QB 135 and by the Court of Appeal in Pritam Kaur.[13]In reply, Ms Dixon, making her points with economy and clarity, submits as follows. The “midnight deadline” case is simply different from other claims. The judge below was right to make the distinction between such claims and others. In such a case, the potential Claimant had the whole of the day of accrual in which to make a claim, removing the rationale for excluding the day of accrual from computation. Ms Dixon submits that such cases are readily identifiable. The effect of excluding the day of accrual is to create a limitation period of (in this instance) six years and a day, in place of the period of six years prescribed by statute. As to previous authority, Ms Dixon submits that it is not in fact clear that the decision in Gelmini was disapproved on this point by the Court of Appeal in Pritam Kaur.

Analysis

[14]I confess that I have not found this an easy case to decide, principally because of the potentially fundamental question: did the relevant cause of action arise at midnight, or just after the midnight deadline expired and therefore during the next day? A claim under such a scheme made literally on the point of midnight (if such a thing could be proved) might arguably be in time. If it is accepted that such a claim accrues after midnight, even by a very short time (described in the hearing before us as a “nanomoment”), then is there a logical reason why it should be distinguished from other categories of claim, some of which will accrue very early in the relevant day?[15]However, in Dodds v Walker [1981] 1 WLR 1027, a case cited by neither counsel, Lord Diplock addressed this problem. The case involved limitation periods under the Landlord and Tenant Act 1954. The relevant passage from his speech reads:
“My Lords, I do not personally derive assistance from pursuing metaphysical arguments about attributing to the one day or the other the punctum temporis between 24.00 hours on September 30 and 0.00 hours on October 1 at which time began to run against the tenant. These seem to me quite inappropriate to the determination of the meaning of a statute which regulates the mutual rights of landlords and tenants of all business premises and is intended to be understood and acted on by them.”
[16]It seems to me that the better view therefore is that in the case of a “midnight” deadline, it is wrong to attribute the accrual of an action such as this to the day after the relevant midnight, and the analysis must proceed from there.[17]The principle of excluding the day of accrual of an action is of long-standing. It was followed in Radcliffe v Bartholomew [1891] 1 QB 161. The Divisional Court was considering computation of a limitation period of one month in criminal proceedings. In the course of his leading judgment, Wills J concluded that the principle was wellestablished in authority, as affecting civil and criminal cases:
“I am of opinion that the decision of the justices was right. Unless there is something in the suggested distinction to the effect that the same words are to be construed differently in civil and criminal cases, I think that the case is really governed by authority, and that of a very cogent kind. In Williams v. Burgess, the language of the section was “within twenty-one days after the execution,” and it was held that the day of execution was there to be reckoned exclusively. In Hardy v. Ryle, which was an action against a justice for trespass and false imprisonment, the plaintiff had been discharged from custody on December 14, and brought his action on June 14 following; s. 8 of 24 Geo. 2, c. 44, prescribed that “no action shall be brought against any justice of the peace for anything done in the execution of his office unless commenced within six calendar months after the act committed.”
Except as to the difference between one month and six months, those words are the same as those which we are now interpreting, and it is impossible to draw any distinction between the language of the two enactments. In that case the Court held that an action brought on June 14 in respect of a cause of action which had arisen on December 14 was in time, because the day on which the cause of action arose was excluded from the computation of the six months, while that on which the action was brought was included in it. It is true that the reasoning by which this decision was supported was criticised by Parke, B., in Young v. Higgon, but that learned judge suggests no doubt as to the correctness of the conclusion. At the end of his judgment occur the cogent remarks:
“Apply the criterion which has been before suggested - reduce the time to one day, and then see what hardship and inconvenience must ensue if the principle I have stated is not to be adopted;” and these remarks are entirely applicable to the present case. The result of reducing the time to one day would be that an offence might be committed a few minutes before midnight, and there would only be those few minutes in which to lay the complaint, which would be to reduce the matter to an absurdity. Therefore, unless a distinction can be established between a statute dealing with criminal, and one dealing with civil, procedure, this case is concluded by authority. It is to be remarked that in Hardy v. Ryle the enactment was not passed for the purpose of giving the plaintiff a right to sue, but in order to impose a limitation on his unrestricted right, and the authority of that and similar cases is therefore most direct and cogent.”
[18]Wills J went on to observe that “Great mischief might, and almost certainly would, arise from altering the canon of construction of the same words, according to the subjectmatter with which they dealt”, and there was no justification for distinguishing between civil and criminal cases affected by the same language of limitation.[19]I note that neither counsel before us considered anything useful was to be derived from the cases cited by Wills J in his judgment.[20]I have already cited the relevant passage from the judgment of Channell J in Gelmini. It is worth noting that neither Radcliffe v Bartholomew nor the cases cited to Wills J were cited to Channell J.[21]This point next arose before Havers J in Marren v Dawson Bentley. This case concerned an accident at work. Counsel for the defendants argued that the accident had happened at 1:30pm, that “at common law part of a day counts as a day”, the day of the accident was therefore “one of the days upon which the action can be brought”, and must be counted. He relied on the decision in Gelmini.[22]Counsel for the plaintiff cited Radcliffe v Bartholomew, and indeed some earlier authority to the same effect. However, it is of some interest that he went on to argue that Gelmini might be distinguished on the ground that it “was a case in contract and there the cause of action was complete at the beginning of the first day and an action could have been brought on that day”. Havers J declined to distinguish Gelmini. He rather disapproved it in the following terms:
“There is a rather remarkable lack of authority on this point. Indeed, there has been no case which has directly decided the point under the Limitation Act 1939, but there has been a number of authorities dealing with other Acts in which Parliament has made a statutory provision of a somewhat similar character. I refer to a passage in Halsbury's Laws of England, 2nd ed., vol. 32, p. 142: "207. The general rule in cases in which a period is fixed within which a person must act or take the consequences is that the day of the act or event from which the period runs should not be counted against him. This rule is especially reasonable in the case in which that person is not necessarily cognisant of the act or event; and further in support of it there is the consideration that, in case the period allowed was one day only, the consequence of including that day would be to reduce to a few hours or minutes the time within which the person affected should take action. 208. In view of these considerations the general rule is that, as well in cases where the limitation of time is imposed by the act of a party as in those where it is imposed by statute, the day from which the time begins to run is excluded; thus, where a period is fixed within which a criminal prosecution or a civil action may be commenced, the day on which the offence is committed or the cause of action arises is excluded in the computation. ..." I was referred to a number of older authorities. The one to which I attach great importance is Radcliffe v. Bartholomew. This was a decision of the Divisional Court.”
[23]The judge went on to analyse the decision in Radcliffe v Bartholomew, noting (see p. 140) that the decision carried the authority of the Divisional Court, and was based in part on the decision in Hardy v Ryle, in which the like conclusion was not doubted by so distinguished a judge as Parke B.[24]Havers J then addressed the decision in Gelmini, noting the absence of authority cited to Channell J. He did not adopt the argument of the plaintiff’s counsel distinguishing Gelmini, but rather expressed his disagreement with the approach of Channell J:
“I think that the principle which was established by the Divisional Court in Radcliffe v. Bartholomew is of general application to statutes whether they deal with civil or criminal matters, and that the rule which the Divisional Court in that case laid down is of general application. I think, therefore, that I am bound by that decision. But if I am not bound by it, and it is upon me to choose between Radcliffe v. Bartholomew and the decision of Channell J., then I prefer the decision in Radcliffe v. Bartholomew and the reasons on which it is based.”
[25]In the Pritam Kaur case, this court was considering limitation in a personal injury claim. The deceased had been killed at work during the working day. At first instance, Wills J followed Marren v Dawson Bentley and Radcliffe v Bartholomew and excluded the day of the accident from computation. However, he concluded (consistently with the second limb of Gelmini) that where, as in that case, the last day of the limitation period so calculated fell on a day when the court office was closed and no proceedings could be issued, nevertheless such a dies non counted, and therefore limitation had expired. The plaintiff appealed that conclusion. The Defendants cross-appealed arguing that the judgment at first instance should be affirmed on the basis that the day of the accident should count, submitting that Marren v Dawson Bentley was wrongly decided. In the Notice of Cross Appeal the Defendants averred:
“…it is accepted that the court has in many cases disregarded the day on which the cause of action accrued. That is an act of clemency in that if a man were injured at 11.50 p.m. it would be unfair to deprive him of a day of his time; but it is quite different in actions in contract or debt.”
[26]There were thus two limbs to the appeal. It is agreed that in relation to the dies non point, Gelmini was wrongly decided, and the appeal was allowed. The other issue is that relevant for this case.[27]The point was addressed by Lord Denning MR in his leading judgment as follows:
“We are asked to decide this preliminary point of law: Was the action commenced within the period of three years allowed by the Statutes of Limitation? Or is it statute-barred? The Act of 1939, as amended by the Act of 1954, says that the action " shall not be brought after the expiration of three years from the date on which the cause of action accrued." The Act of 1846, as amended by the Act of 1954, says that it "shall be commenced within three years after the death." Nothing turns on the difference in wording. The period is the same in either case. The first thing to notice is that, in computing the three years, you do not count the first day, September 5, 1967, on which the accident occurred. It was so held by Havers J. inMarren v. Dawson Bentley & Co. Ltd. [1961] 2 Q.B. 135. The defendants here, by their cross-notice, challenged that decision: but I think it was plainly right.”
[28]Karminski LJ concurred. The third member of the court was Megarry J who addressed the point as follows:
“The case arises upon two similar statutory provisions. There is a claim under the Fatal Accidents Acts 1846 to 1959; and by section 3 of the Act of 1846, as amended by the Law Reform (Limitation of Actions etc.) Act 1954, section 3, "every such action shall be commenced within three years after the death of such deceased person." There is also a claim in tort for negligence and breach of statutory duty which falls within section 2 (1) of the Limitation Act 1939, as amended by section 2 (1) of the Act of 1954; and this provides that the action "shall not be brought after the expiration of three years from the date on which the cause of action accrued." No point, I may say, has been taken in argument on the difference in wording between "within three years after the death" and "after the expiration of three years from the date." At one time there was some argument on whether or not the period was to be reckoned by excluding the date on which the accident occurred, but in the end the point was not pressed. The decision of Havers J. in Marren v. Dawson Bentley & Co. Ltd. [1961] 2 Q.B. 135, based on section 2 (1) of the Limitation Act 1939, was that the day of the accident was to be excluded in the computation of the time; and in the present case the judge applied that decision. The language of section 2 (1), with the phrase "after the expiration of three years from the date," plainly supports that view. If the wording of the Fatal Accidents Acts, with the phrase "within three years after the death," is less apt, it would nevertheless be regrettable to introduce any fine distinctions, especially as the period of three years was inserted into each statute by the same Act, that of 1954. I would therefore agree with the judge in excluding the day of the accident from the computation under both heads.”
[29]Ms Dixon has made the point that the Pritam Kaur case was not a “midnight deadline” case, and she submits must not be regarded as direct authority on such a case. This is really to advance the distinction between such cases and all others which had not attracted Havers J in Marren v Dawson Bentley.[30]In addition to reliance on Gelmini, Ms Dixon relies on commentary in McGee on Limitation Periods (7th Edition, paragraphs 2-005/007) where the Editors wrote:
“The general rule in calculating the expiry of a limitation period is usually expressed as being that parts of a day are ignored. This formulation is ambiguous, and needs to be clarified by example. In Gelmini v Moriggia the defendant had given a promissory note. The time for payment of this expired on 22 September 1906. The claimant’s writ on the note was issued on 23 September 1912. Channell J held that the cause of action was complete at the beginning of 23 September 1906, since that was the earliest moment at which proceedings could have been commenced, notwithstanding that the court office obviously would not have been open at midnight. Consequently the sixyear limitation period expired at the end of 22 September 1912, and the writ issued on the following day was out of time. This is the simplest possible example, since the cause of action was held to accrue at the very beginning of a day. … Perhaps the most satisfactory of the authorities on this point is Marren v Dawson Bentley & Co. The claimant was injured in an accident at 13.30 on 8 November 1954, and the writ was issued on 8 November 1957. The question was whether time had expired at the end of 7 November 1957, and Havers J held that it had not. The day on which the cause of action accrues is to be disregarded in calculating the running of time. It therefore followed that time began to run at the first moment of 9 November 1954 and expired at the end of 8 November 1957. Havers J expressly declined to follow Gelmini v Moriggia, but it is not clear whether his decision is inconsistent with that in Gelmini. The latter case deals with one very specific situation, namely where the cause of action must accrue on the stroke of midnight. It is arguable that here there is no question of disregarding any part of a day; the cause of action was in existence throughout 23 September 1906. Consequently, it may be argued that on those very special facts the decision is still good law. The alternative is to say that time did not begin to run until the start of 24 September, which seems a very odd conclusion, given that the time for payment expired at the end of 22 September. It is submitted that the cases are reconcilable and that both are correct on this point. The rule is that any part of a day (but not a whole day) happening after the cause of action accrues is excluded from the calculation of the limitation period. Strictly speaking this will normally lead to the extension of the limitation period by a few hours but it could equally be argued that the contrary rule would lead to the shortening of that period.”
[31]On that basis Ms Dixon argues for the distinction she advances.[32]Having reflected on the matter, with the dictum of Lord Diplock in Dodds v Walker in mind, it appears to me that Ms Dixon is correct. A “midnight deadline” case is different from others in the sense that the deadline provides a categorical indication that the action accrued by that point in time, rather than accruing on the day following midnight. For that reason, no fractions of a day arise.[33]Different considerations might arise, in my view, if the scheme of arrangement had stipulated a claim by 9am on the relevant day, leaving the whole of the working day later than the moment the action accrued. That would be equally certain as to timing and would provide equal opportunity for initiating legal steps. Then what if noon were stipulated, or 3pm? What then of fractions of a day?[34]The classic statement, to the effect that English law will disregard fractions of a day is in Lester v Garland (1808) 15 Ves Jun 248 at 257, 33 ER 748 at 752:
“It is not necessary to lay down any general rule upon this subject: but upon technical reasoning I rather think, it would be more easy to maintain, that the day of an act done, or an event happening, ought in all cases to be excluded, than that it should in all cases be included. Our law rejects fractions of a day more generally than the civil law does. (See the note, 14 Ves. 554, where it is admitted in bankrupty.) The effect is to render the day a sort of indivisible point; so that any act, done in the compass of it, is no more referrible to any one, than to any other, portion of it; but the act and the day are co-extensive; and therefore the act cannot properly be said to be passed, until the day is passed. This reasoning was adopted by Lord Rosslyn and Lord Thurlow in the case before mentioned of Mercer v. Ogilvie. The ground, on which the judgment of the Court of Session was affirmed by the House of Lords, is correctly stated in the fourth volume of the Dictionary of the Decisions of the Court of Session. In the present case the technical rule forbids us to consider the hour of the testator's death at the time of his death; for that would be making a fraction of a day. The day of the death must therefore be the time of the death; and that time must be past, before the six months can begin to run.”
[35]In the end, this question does not arise here. The Court of Appeal in Pritam Kaur was not addressing the “midnight deadline” case. Although they approved the decision in Marren v Dawson Bentley in general terms, the approval must be seen in the context of the facts they were addressing, and which, for these purposes, could not be distinguished from those in Marren v Dawson Bentley.[36]In the end, if the distinction advanced by Ms Dixon and accepted by the judge is correct, then the problem of the exclusion of a part of a day does not arise.[37]For those reasons, I would dismiss the appeal.[38]I agree that this appeal should be dismissed. In my view there is, as propounded at the end of the passage from McGee on Limitation Periods set out at paragraph 30 of Irwin LJ’s judgment, a clear distinction between the case where a cause of action accrues “at the stroke of midnight”, because it is based on a failure to do something by the end of a specified day, and the case where the cause of action accrues part way through a day. In the latter case it is indeed well-established that for limitation purposes you ignore the date on which the cause of action accrues: the authorities go back to the early nineteenth century and were not originally concerned with the Limitation Acts, but they culminate in Pritam Kaur, which is binding authority on their application in the context of what is now the Limitation Act 1980. But in the former type of case the cause of action arises, as Channell J put it in Gelmini, “at the commencement of [the] day”. Even without the benefit of Dodds v Walker I would not have accepted that in such a case there is even a “nano-moment” after midnight when the cause of action is not in being, but Lord Diplock’s observations quoted at paragraph 15 above confirm my view: the cause of action arises at, not after, midnight. I regard Gelmini as being, on this point, rightly decided. Havers J took a different view in Marren, as Irwin LJ shows,but his decision is not binding on us, and I agree with Irwin LJ that the general approval of Marren in Pritam Kaur is not authoritative as regards this issue.[39]As to the question raised by Irwin LJ at paras. 33-34 of his judgment, my strong provisional view is that even in a case where the cause of action arises very early in the relevant day that day falls to be excluded, because I cannot see a rational basis on which to distinguish between fractions of a day that are or are not sufficiently big to count. But, as he says at paragraph 35, the question does not arise in this case, because the essential point is that there was no part of the day during which the Appellants did not enjoy their cause of action.

Cited in 1 later judgment