Mitsui Sumitomo Insurance Co (Europe) Ltd & Ors v Mayor's Office for Policing and Crime [2014] EWCA Civ 682

Case No A3/2013/2834

Case No A3/2013/2840

Case No A3/2013/2843

Case No A3/2013/2845

[2014] EWCA Civ 682

IN THE COURT OF APPEAL (CIVIL DIVISION)

[2013] EWHC 2734 (Comm)

ON APPEAL FROM THE QUEEN’S BENCH DIVISION (COMMERCIAL COURT)

MR JUSTICE FLAUX

Venue Royal Courts of Justice, Strand, London, WC2A 2LL

Date 20/05/2014

Before

MASTER OF THE ROLLS

LORD JUSTICE MOORE-BICK

LORD JUSTICE LEWISON

Between

(1) MITSUI SUMITOMO INSURANCE CO (EUROPE) LTDRespondent

(2) TOKIO MARINE EUROPE INSURANCE LTDRespondent

(3) ROYAL & SUN ALLIANCE INSURANCE PLC2 nd Respondent

(4) LACE INTERNATIONAL LTD3 rd Respondent

(5) CLEAR VISION LTD3 rd Respondent

(6) ASPHYXIATION FILMS LTD3 rd Respondent

MAYOR’S OFFICE FOR POLICING AND CRIMEAppellant

Mr Sam Grodzinski QC and Mr David Pievsky (instructed by Directorate of Legal Services, Metropolitan Police) appeared for Appellant

Mr Michael Crane QC, Miss Tamara Oppenheimer and Miss Marianne Butler (instructed by DAC Beachcroft LLP) appeared for Respondents

Mr Michael Crane QC and Mr Charles Dougherty QC (instructed by Kennedys) appeared for Second Respondent

Mr Simon Pritchard (instructed by Keystone Law) appeared for Third Respondents

Hearing Hearing dates: 8 and 9 April 2014

Judgment

Master of the Rolls:

[1]This is the judgment of the court to which all its members have contributed. Introduction

[2]Mark Duggan was shot and killed by police in Tottenham on 6 August 2011. For the next four days thousands of people rioted in London and other English cities. During that period, at 23.40 on Monday 8 August, a gang of youths broke into the Sony distribution warehouse in Enfield, looted it, and burned it down with petrol bombs. If the gang were “persons riotously and tumultuously assembled”, the Mayor’s Office for Policing and Crime (“MOPC”) is liable to compensate anyone who sustained loss by reason of the looting or arson (or if their insurers have paid, their insurers).

[3]Flaux J decided that the gang were “persons riotously and tumultuously assembled”, but that MOPC’s liability did not extend to consequential loss. His judgment is at

[2013] EWHC 2734 (Comm)

;

[2014] 1 All ER 422

. MOPC appeals against the finding of liability. The insurers (as well as the owners of some uninsured stock contained in the warehouse) cross-appeal against his decision on the extent of the liability. Both the appeal and the cross-appeal are brought with the permission of the judge. The primary facts

[4]The primary facts can be taken from the judge’s judgment. The judge also had the benefit of a site view, which no doubt also informed the facts that he found. The warehouse is located on Solar Way in the Innova Business Park. There is a Premier Inn about 100 metres to the south. The business park lies on the eastern edge of a built-up residential area. It is bounded on the east by Mollison Avenue, which in turn is on the edge of the Lea Valley Regional Park. A short way to the north lies the M25. To the south east, separated from the business park by open ground, lies the Enfield Island Village, a large housing estate. As its name suggests, it is situated on an island formed by the River Lea Navigation and the Rammey Marsh Flood Relief Channel.

[5]There had been serious civil disorder in Enfield Town on the night of 7/8 August which culminated in an attack by a gang of 15 – 20 youths on a jeweller’s shop, in which the shop window was broken and items of jewellery were stolen. As 8 August wore on reports were received by the police that what had happened the previous night would happen again. Some of the concerns related to the possibility of rioting in the vicinity of the warehouse and the Premier Inn. One of the staff at the Premier Inn and one of her colleagues heard rumours via Facebook and Blackberry Messenger that the riots were going to move from Enfield Town to Enfield Island Village. In consequence the manager closed the restaurant slightly earlier than usual and locked the back doors. Furthermore, the security guard employed at the warehouse, Edward Mireku, was asked by Sony's Security Manager, John Rapley, to work overnight at the warehouse on the night of Monday 8 August 2011 as a precaution because of the riots elsewhere in London in the previous days. That reflected a concern that the Innova business park where the warehouse was situated might be attacked.

[6]At 18.14 a member of the public who lived in Aldridge Avenue, just over the river from Enfield Island Village, rang the police. She said that where she lived there was a gang who called themselves the GMG (Get Money Gang, also known as the Black Gang); she believed they were going to be attending the riots. There “was a load of them on bikes”, a reference to her having seen youths who were members of the gang riding around on bikes at that time. At 21.18 there was another call from Aldridge Avenue, in all probability from the same lady, who said she had called earlier about 20 youths being gathered, but they had all left now. She was concerned due to recent activities. At about 21.45, a resident of Enfield Island Village saw a large group of 15 to 20 youths gathered on the junction of Manton Road and Punchard Crescent. They were aged between 17 and 20, some younger. She recognised about ten of them as being from the Village; they were black, mixed race and Somali. She recognised them as members of the Black Gang. The other youths in the group were strangers to the Village whom she did not recognise. Another member of the public saw a group of 15 youths congregating on the corner of Punchard Crescent who were black, Somali or mixed race and members of the Black Gang. They were all in predominantly dark clothing, some wearing bandanas or balaclavas. Some were on bikes. They were all talking among themselves and using their phones, apparently to organise more people to attend and head over to the Innova business park where the warehouse was situated. That person overheard one of the youths saying on his phone and to others in the group: “Everyone doing stuff at their own ends” which he explained meant that local youths were grouping together to attack local premises, to “get rich” and out-do rival gangs.

[7]Police intelligence after the incident was that the attack on the warehouse was carried out by members of the GMG joined by members of at least one other gang from Chingford, the Chingford Hall gang. The judge found at [40] that that was indeed the case.

[8]At about 23.15 Daniel Thomas and his mother Marina were driving over the bridge to Enfield Island Village on Smeaton Road. Mr Thomas said he saw a group of 10 to 15 youths, all in dark hoodies with their faces covered in scarves and bandanas, running along the pavement on the brow of the bridge and into the little car park by the gate into the fields. One of them was carrying a baseball bat and one a metal pole. His mother’s evidence was slightly different. She reported a smaller group of about six walking rather than running over the bridge, all in dark clothing and hoodies, one of them carrying an iron bar and another a sledge-hammer. The judge did not regard the discrepancy as significant. Another intelligence report from a member of the public at 00.22 hours on 9 August was of having seen a group of about 30 hooded males carrying sticks congregating on the fields off Smeaton Road at 23.20. This was some ten minutes before the attack on the warehouse. The judge found at [42] that this was the same group as the group that attacked the warehouse ten minutes later.

[9]The group crossed the fields by the footpath to the stile on Mollison Avenue, then crossed the road, possibly reaching the door at the south end of the west side of the warehouse through the car park of the Premier Inn. At 23.40 the night manager of the Premier Inn made a 999 call reporting about four boys in hoodies walking around the bar area. This was at about the same time as the group were smashing their way into the warehouse and these boys were associated with those who were attacking the warehouse.

[10]The attack was perpetrated by the GMG and other gangs. They all wore gloves and masks, bandanas or hoodies. There were 20 to 25 of them aged between 12 and 20. They carried at least three petrol bombs with them, and a variety of other weapons, including sticks, iron bars, a hammer and a baseball bat.

The judge takes up the story in his own words:

“45. The attack itself is captured on CCTV from the Warehouse. Whilst that CCTV is silent, it still gives a good flavour of the nature of the attack. A number of the group attacked the glazed entrance door, kicking it and hitting it with whatever implements they had, which presumably included the baseball bats and sledgehammers described by the Thomases, as well as a shovel and garden hoe later found abandoned at the scene. Apart from one person in the group who seemed to be wearing a white or pale coloured balaclava or bandana and who stood relatively still whilst the door was being battered in, (suggesting he may have been the ringleader of the group), the others were moving around in what appears to have been an excited and agitated manner, whilst some of the group were attacking the door. Once the door was breached, the group (including the apparent ringleader) swarmed into the building instantly. The attack began at 23.37 and the group broke in just over a minute later. 46. The breaking into the door of the Warehouse caused a smashing sound which was heard by staff at the Premier Inn. Chris Cass … was outside in the bin yard, a smoking area which is on the side of the hotel premises closest to the Warehouse, when he heard the sound of forceful smashing of glass. He went back into the restaurant to tell staff to shut the blinds and lock up the building. Holly Charles was also outside in the bin yard with her colleague Rachel when she heard a crashing sound from the direction of the Warehouse. She immediately thought it could be rioters looting the Warehouse. She was worried because she wasn't sure what was going on. She went inside and told another member of staff, Thomas Thompson … who went over to have a look. He found the glazed entrance door on the west side of the Warehouse, near the truck entrance, smashed. There were seven to ten BMX and mountain bikes on the ground at the truck entrance. He saw two black youths standing outside, whom he estimated to be about twelve years old. One of them was holding a wine bottle which he thought had petrol in it, a home-made petrol bomb. When the youth saw Mr Thompson, he put the bottle down in the road. Mr Thompson said to the two youths that the police had been called and they might want to make a move. They thanked him. In the event, that petrol bomb was left at the scene. 47. The CCTV footage from cameras inside the Warehouse shows the group running down the aisles between the stacking, looting items which they can be seen carrying off. Two of the group lit and threw petrol bombs into the stacking before they all left the building. The first petrol bomb was thrown at 23.40.20, so about a minute and a half after they gained access. Mr Thompson describes at least fifteen youths running out of the Warehouse through the broken door and possibly the adjacent broken glass panel. They were all carrying items looted from the Warehouse, a couple of them were holding boxes and one had a handful of game controllers. Another had a handful of goods which he dropped as he tried to get on a bike. The exit of the group from the Warehouse is also caught on the external CCTV camera. It is not possible to see all the youths carrying goods as described by Mr Thompson, but it may be that they were hiding items under their clothing. 48. Mr Thompson describes the group as between fifteen and twenty five in number, with an average age of about fifteen, the oldest about seventeen. They were all wearing coverings such as bandanas over their faces, which they pulled up when they saw him. One of them, whom he describes as the only one of the group who was not black, being of mixed race or Turkish, said to him in an aggressive manner: “What the fuck are you looking at?”

Another one of the group running out of the Warehouse ran beside Mr Thompson and punched him on the side of the head, slightly dazing him. His evidence is that when the larger group of youths came out of the Warehouse he felt intimidated and when the youth spoke to him aggressively he was scared. He also thought more of them were going to hit him but they ran off. He saw them heading south down Solar Way towards the roundabout. 50. In his statement, Mr Mireku, the security guard, describes how, at the time the incident occurred, he was engaged in a patrol of the building which had commenced at about 23.30 when the other staff and security officers had left the Warehouse. As he was checking a rest room right at the other end of the building from where the looters broke in, he heard loud banging noises like hammering. Shortly thereafter, the intruder alarm went off inside the building. Mr Mireku says he was instantly very scared, as he knew someone must have entered the building. In order to escape, he went to the nearest exit in the opposite direction to where the noises were coming from, which was a fire escape door next to loading bay 10. That door was locked with a break-glass bolt lock but could also be opened with a key. To save time he broke the glass rather than searching through his keys. 51. Having left the building Mr Mireku made his way to the gatehouse (which was on the south side of the Warehouse) intending to look at the CCTV screens there. As he came round the south east corner of the building, he saw five or six people in the distance running away from the Warehouse, down the pavement on Solar Way. He did not want to be seen by them in case they attacked him, so he hid in some bushes along the perimeter of the yard and took off his high visibility jacket. He called the police on his mobile phone, that call being timed at 23.44 hours. He then went over to the main gate which he opened with a key in order to reach the drivers of two G&D tractor units which were parked in the yard ready to pick up trailers first thing next morning. He had noticed these units when he locked up the gatehouse at 23.20. Since the curtains were drawn on the cabins of both he had assumed the drivers were asleep inside. When he approached the units after the incident he banged on the sides of the cabins to alert the drivers.” 53. The departure of the group from the scene was also observed by staff at the Premier Inn. Holly Charles describes how, very soon after she had heard the crashing sound from the Warehouse, which was evidently the youths breaking into the Warehouse, she heard another crashing sound and she went outside through the bin yard into the car park to see what was going on. She saw about ten teenagers, some wearing hoodies, running away from the Warehouse (albeit she says they did not seem to be in a hurry). They ran in front of the Premier Inn along Solar Way towards Ordnance Road. Some of them were carrying boxes and dropping things everywhere. 54. Mr Cass describes how, after he had warned staff to shut the blinds and lock the restaurant, he went out to the gate to see what was happening. He saw a group of about fifteen youths running towards him from the direction of the Warehouse carrying white boxes. Two of the group were on BMX bikes. One of the youths at the back of the group slipped on the grass area near where Mr Cass was standing and Mr Cass abused him roundly, from which it appears that he at least did not feel intimidated. The youth got up and ran off. The evidence of these witnesses indicates how short a period of time the whole incident took and this is borne out by the CCTV footage, which demonstrates that it lasted around three minutes. 55. The CCTV footage from inside the Warehouse also shows the fire taking hold within the building within a matter of minutes of the petrol bombs being lit and thrown. That is also borne out by Ms Charles' evidence, which is that within a few minutes of the youths running away, Mr Thompson returned to the hotel. They could see that the Warehouse was on fire and there was a lot of smoke. They went inside the restaurant for safety reasons and she called the fire brigade. Within ten minutes of them first seeing the fire, the whole building went up in flames. The first police officers on the scene were in fact a mobile patrol from the Hertfordshire Police who had seen the flames from the other side of the M25. The Metropolitan Police arrived at 23.59. When the police and fire brigade arrived, the hotel was evacuated because of concerns the fire might spread to the hotel. There were explosions from the Warehouse. In the event the fire did not spread and the guests were safely back in the hotel by 5 in the morning of 9 August 2011. Sergeant Andrew Davison was one of the Metropolitan Police officers first on the scene. He gave oral evidence at the trial. His unchallenged evidence was that when he arrived there was no evidence of disorder in the surrounding area or of debris on the roads or footpaths.”

[12]One of the questions raised at trial was why the gang had thrown petrol bombs in the warehouse. Detective Superintendent Hibberd, who was in charge of the investigation, thought that they had done so in order to cover up forensic evidence (such as DNA), but the judge did not accept that theory. The petrol bombs were thrown in the middle of the warehouse, whereas the forensic evidence (mainly in the shape of bloodstains resulting from cuts caused by breaking glass) lay at its perimeter. The group carried petrol bombs with them although they would not have known in advance that any of them would cut themselves. One petrol bomb (with a T-shirt serving as its wick) was in any event left outside the warehouse. The judge concluded:

“63. In my judgment, it is much more likely that the group took petrol bombs with them not to cover their tracks forensically, as Mr Hibberd believed, but to use them to set fire to the Warehouse and destroy or damage it and its contents. Even if Mr Hibberd is right in his belief that the group took the petrol bombs to cover their tracks forensically, their intention was still to set fire to the building and in one sense, their reason or motive for doing so is irrelevant. However, as I say, I consider it much more likely that their motive was one of wanton destruction or damage and thus an animus towards the property.”

[13]In the aftermath of the fire the police made a number of arrests, but the CPS decided that there was insufficient evidence for a prosecution, so no one was prosecuted.

[14]The result of the petrol bombs was that the warehouse and its contents were completely destroyed. The fire is said to have been the largest arson in Europe. The claimed losses

[15]Both Sony (which occupied the warehouse) and Cresta Estates Ltd (which owned it) were insured. Each claimed against its insurers. In Sony’s case it was insured not only against damage to the contents of the warehouse but also against business interruption losses. Insurers paid out under the policy. Their claim in this action includes £9.8 million which they paid out for loss of profit as a result of the destruction of Sony’s distribution equipment, including the costs of mitigation. Cresta Estates Ltd were insured not only against physical damage to the warehouse, but also against loss of rent. Their insurers paid out around £1.5 million in respect of loss of rent, which they claim in this action. The third disputed claim is made by Lace International Ltd (and others) each of whom had stock in the warehouse. They claim approximately £3 million for loss of profit as a result of the destruction of trading stock which caused a dislocation of their respective businesses. None of them was insured against that loss. The Riot (Damages) Act 1886

[16]The claims in this case are all made under the Riot (Damages) Act 1886. The basic entitlement to compensation is conferred by section 2(1) which provides:

“Where a house, shop, or building in a police area has been injured or destroyed, or the property therein has been injured, stolen, or destroyed, by any persons riotously and tumultuously assembled together, such compensation as hereinafter mentioned shall be paid out of the police fund of the area to any person who has sustained loss by such injury, stealing, or destruction; but in fixing the amount of such compensation regard shall be had to the conduct of the said person, whether as respects the precautions taken by him or as respects his being a party or accessory to such riotous or tumultuous assembly, or as regards any provocation offered to the persons assembled or otherwise.”

[17]By section 2(2) of the Act, where insurers have indemnified an insured against losses falling within section 2(1) they are entitled to claim compensation in place of the insured. Section 3 provides:

“(1) Claims for compensation under this Act shall be made to the compensation authority of the police area where the injury, stealing or destruction took place, and such compensation authority shall inquire into the truth thereof, and shall, if satisfied, fix such compensation as appears to them to be just. (2) A Secretary of State may from time to time make, and when made, revoke and vary regulations respecting the time, manner, and conditions within, in, and under which claims for compensation under this Act are to be made, and all such claims not made in accordance with such regulations may be excluded. Such regulations may also provide for the particulars to be stated in any claim, and for the verification of any claim, and of any facts incidental thereto, by statutory declarations, production of books, vouchers, and documents, entry of premises, and otherwise, and may also provide for any matter which under this act can be prescribed, and for the compensation authority obtaining information and assistance for determining the said claims.”

[18]In July 1886, July 1894 and again in October 1921, the Secretary of State made regulations prescribing the form in which a claim under the Act should be made, but these requirements were abolished by the Riot (Damages) (Amendment No 2) Regulations 2011. There is now no officially prescribed form. The relevant provisions of the Act (including the original preamble repealed by the Statute Law Revision Act 1898) are set out more fully in the Appendix to this judgment.

[19]Both the issue of liability and the issue about the extent of the recoverable compensation turn principally on the interpretation of section 2(1). In the case of the first issue the question is whether the perpetrators of the damage caused on 8 August fall within the description “persons riotously and tumultuously assembled together.” In the case of the second issue the question turns on whether compensation is recoverable only in respect of the physical destruction or damage, or whether it also extends to consequential losses caused by that destruction or damage.

[20]Section 10(1) of the Public Order Act 1986 provides that “riotous” and “riotously” in the 1886 Act are to be construed in accordance with section 1 of that Act. Section 1 defines “riot” as follows:

“(1) Where 12 or more persons who are present together use or threaten unlawful violence for a common purpose and the conduct of them (taken together) is such as would cause a person of reasonable firmness present at the scene to fear for his personal safety, each of the persons using unlawful violence for the common purpose is guilty of riot. (2) It is immaterial whether or not the 12 or more use or threaten unlawful violence simultaneously. (3) The common purpose may be inferred from conduct. (4) No person of reasonable firmness need actually be, or be likely to be, present at the scene. (5) Riot may be committed in private as well as in public places.”

[21]“Violence” includes violent conduct towards property as well as towards persons: Public Order Act 1986, section 8. The legal and historical background

[22]The legal and historical background to the 1886 Act is of importance to both of the issues which arise on this appeal. In Field v Receiver of Metropolitan Police

[1907] 2 KB 853

the Divisional Court discussed the ingredients of riot at common law. They pointed out that at common law riot was a misdemeanour (rather than a felony): [1907] 2 KB at 859. After discussion of many texts and cases they concluded that there were five necessary ingredients of a riot:(1) at least three persons;

(2) a common purpose;

(3) the execution or inception of the common purpose;

(4) an intention to help one another, by force if necessary, against any person who might oppose them in the execution of the common purpose; and

(5) force or violence displayed in such a manner as to alarm at least one person of reasonable firmness and courage: [1907] 2 KB at 860.

[23]However, the modern history of riot begins with the Riot Act 1714. It was passed shortly after the accession of George I, whose coronation had itself been the occasion for widespread rioting. Section 1 of that Act imposed the death penalty on persons to the number of 12 or more “unlawfully, riotously and tumultuously assembled” who failed to disperse within one hour after a proclamation in a statutory form had been made. The offence under section 1 was expressly stated to be a “felony without benefit of clergy”. Benefit of clergy was a means of escaping capital punishment by reading (or reciting) the first two verses of Psalm 51 (known as “the neck-verse”), but in practice it was often excluded by statute. The statutory form of proclamation was given in section 2, from which we get the phrase “reading the Riot Act”. Section 6 of the Act provided for compensation if “persons … unlawfully, riotously and tumultuously assembled” demolished or pulled down, wholly or in part, a church, chapel, building for religious worship, dwelling house, barn, stable or outbuilding. In the case of property outside a town or city, the compensation was payable by the inhabitants of the hundred in which the building was situated. In the case of property within a city or town, it was to be paid by the inhabitants of the city or town. In either case there was power to levy a rate to pay the compensation.

[24]In Ratcliffe v Eden (1776) 2 Cowp 485 the plaintiff claimed damages in respect of the partial demolition of his house by a riotous and tumultuous assembly and the destruction of the furniture which was in the house at the time. The issue for the court was whether the plaintiff was entitled to recover damages for the destruction of the furniture under Section 6. It was argued on behalf of the defendants that there was no such entitlement since no mention was made in the statute of the destruction of furniture and this was not in the contemplation of the legislature at the time.

[25]Lord Mansfield explained the policy underlying the Riot Act 1714 as follows:

“If the Act had never been made, the trespassers would have been liable to answer for the whole injury in damages. To encourage people to resist persons thus riotously assembled, and to reward those, who, by doing their duty, shall have incurred their resentment, the same law has made a further provision, that as the trespassers are to be hanged, the country shall pay the damages: and this, by way of inducement to the inhabitants to be active in suppressing such riots, which it is their duty to do; and which being thus made their interest too, they are more likely to execute. This is the great principle of the law, that the inhabitants shall be in the nature of sureties for one another. It is a very ancient principle; as old as the institution of the decennaries by Alfred, whereby the whole neighbourhood or tithing of freemen were mutually pledges for each other's good behaviour. The same principle obtains in the Statutes of Hue and Cry. It is the principle here.”

[26]In the same case Ashton J said that:

“The object and principle of this Act was to transfer the damages occasioned by the trespass from the rioters to the hundred; to make it felony in the offenders themselves, and to put the party injured in the same state as before. It is a remedial law, and ought to be extended.”

[27]The underlying principle that the inhabitants are sureties for each other’s good behaviour is the dominant theme. In so far as it is necessary to found the action on any wrong, it is the wrong of the rioters, not that of the inhabitants of the hundred, who stand surety for the good behaviour of their fellow citizens.

[28]The court held that Mr Ratcliffe was entitled to recover compensation for the damage to his furniture, as well as to his house, even though the Act did not mention furniture specifically. Lord Mansfield said that the destruction of the furniture was not a distinct act from the destruction of the building: “ . . . it comes therefore very near the case of pulling down the whole house and thereby crushing the furniture”.

[29]The effect of the case was incorporated in the next statutory change to the compensation scheme.

[30]The next case is Wilmot v Horton (1781), which is cited as a footnote to the report of the judgment of the Court of King’s Bench in Hyde v Cogan (1781) 2 Douglas 699. Again, one can see the theme of the community being answerable for the wrongdoing of the rioters. In that case the plaintiff brought a claim under Section 6 of the 1714 Act to recover damages for the destruction of his house, furniture and garden. Some of the damage was caused to the furniture and the garden by the destruction of the house, but some of the damage was caused by the rioters independently of the destruction of the house. The court decided that the plaintiff was entitled to recover for the whole of the destruction of the furniture and the damage to the garden, as well as for the destruction of the house. Lord Loughborough said:

“This statute, though penal in a great part of its provisions, and though, perhaps, there is something of a penal nature in transferring the action from the party committing the felony to the hundred, yet, with respect to the party injured, must be considered as remedial. Antecedent to this statute, and till the trespass was turned into a felony, there is no doubt that, against the actors and their abettors, the party injured would have been entitled to recover damages for all his loss. In lieu of that remedy, which can no longer be had, it was thought better to substitute an action against the hundred, in analogy to the ancient policy of the kingdom, by which the men of each district were bound to maintain peace and order, and to answer for the violation of them within that district. The Act does not say that damages shall be yielded for the injury done to the fabric of the house, but, by the demolishing or pulling down of the house; and it seemed to be admitted in the argument that, if the destruction of the furniture in the house were the necessary consequence of the demolition of the house itself, the plaintiff would be entitled to recover the full amount of his loss. …But will not the principle extend likewise to those damages which, though not the necessary consequence of the felony committed, yet are clearly the immediate effect of that cause? If, in order to destroy the walls, the mob break down the wainscoting and the glasses, or if by driving a beam or joist against the wall to throw that down, they break a glass fixed against the wainscot, it would be strange to argue that such destruction, not being a necessary, but only an immediate consequence of demolishing the house, should not be repaired by this action. The case might be different as to consequences that are neither necessary nor immediate: if, for instance, one set of rioters had broken in on the 7th of June and destroyed the house, leaving the goods in it, and the next day, another party had come to the house so left, and robbed it of the furniture.”

[31]In Hyde v Cogan itself the plaintiff claimed damages in respect of the destruction of goods and furniture in the house, although they were not destroyed by the act of pulling down the house. The claim was allowed. Willes J said that it would “be overturning the good sense of the statute to restrain it in the manner contended for by counsel for the defendants”. Ashurst J said:

“the purpose of [Section 6 of] the Act is remedial and therefore it ought to receive a liberal construction”

. Buller J agreed that, since the section was “remedial”, its interpretation should be “liberal” and the “most extensive sense must prevail”. He agreed with the statement of Aston J in Ratcliffe v Eden that the sense of the Act was that “the damages to which the party offending would have been liable at common law should be transferred to the hundred”.

[32]In Mason v Sainsbury (1782) 3 Dougl 61 (another Gordon Riots case) it was held that, where insurers have paid the amount of the loss occasioned by the demolition of a house by rioters, they may maintain an action in the name of the assured against the hundred under article VI of the 1714 Act. In the course of his judgment, Lord Mansfield, who had himself been a victim of the riots when his town house in Bloomsbury Square was sacked by the mob, said:

“The case is clear: the Act puts the hundred, for civil purposes, in the place of the trespassers; and upon principles of policy, as in the case of other remedies against the hundred, I am satisfied that it is to be considered as if the insurers had not paid a farthing.”

[33]In the same case Buller J said that it was rightly admitted that “the hundred is put in place of the trespassers.”

[34]Section 38 of the Seditious Meetings Act 1817 extended the compensation provisions to “any house, shop or other building whatever” and also to fixtures, furniture goods or commodities. It also extended them to any damage or injury to buildings, and in the case of goods etc to theft as well. However, section 38 used the phrase “any riotous or tumultuous assembly of persons,” whereas the Riot Act 1714 had used the phrase “riotously and tumultuously assembled.” Whether the change was deliberate we do not know.

[35]These provisions were consolidated, and to some extent amended, in the Remedies against the Hundred Act 1827. Section 6 provided that any building:

“feloniously demolished, pulled down, or destroyed, wholly or in part, by any Persons riotously and tumultuously assembled together, in every such Case the Inhabitants of the Hundred, Wapentake, Ward or other District in the Nature of a Hundred, by whatever Name it shall be denominated, in which any of the said Offences shall be committed, shall be liable to yield full Compensation to the Person or Persons damnified by the Offence, not only for the Damage so done to any of the Subjects herein enumerated, but also for any Damage which may at the same Time be done by any such Offenders to any Fixture, Furniture, or Goods whatever, in any such [building].”

[36]As can be seen, it required compensation to be paid for property “feloniously demolished … by any persons riotously and tumultuously assembled”. The word “and” has remained in the legislation ever since.

[37]The main change made by the 1886 Act was to transfer liability from the inhabitants to the rate of the police district in which the property was situated. There have been subsequent changes in nomenclature following re-organisations of policing. So far as the Metropolitan Police District is concerned, the compensation is now to be paid by The Mayor’s Office for Policing and Crime (a corporation established by section 3 of the Police Reform and Social Responsibility Act 2011, whose occupant is the Mayor of London). Thus, in marked contrast to the position under the general law, which imposes no duty of care on the police, the police have a potentially onerous obligation to make compensation in the circumstances envisaged by section 2.

(a) Riotously and tumultuously

[38]We have seen that “riotously” is to be construed in accordance with the modern definition of “riot”. What about “tumultuously”? Mr Crane submitted that since the recasting of the offence of riot, and in particular the raising of the minimum number of participants from three to twelve, there was less need to “invest” in the meaning of tumultuously. However, the meaning of that word in the context of the 1886 Act is the subject of case-law which in our judgment is binding on us. JW Dwyer Ltd v Metropolitan Police District Receiver

[1967] 2 QB 970

concerned an armed robbery of a jeweller’s shop in East London. The four robbers did not force their way in, but once inside the shop they immobilised the staff and threatened the manager’s wife with an iron bar. The robbery took place without attracting the attention of anyone outside. Lyell J held that an assembly had to be both riotous and tumultuous before liability to pay compensation under the 1886 Act arose.

[39]In DH Edmonds Ltd v East Sussex Police Authority (unreported 6 July 1988; [198] CAT 672) this court expressly approved Lyell J’s decision and did so again in Bedfordshire Police Authority v Constable

[2009] EWCA Civ 64

; [2009] 2 All ER (Comm) 200 at [12] and Yarl’s Wood Immigration Ltd v Bedfordshire Police Authority

[2009] EWCA Civ 1110

;

[2010] QB 698

at [50]. It is necessary therefore to examine with some care exactly what Lyell J said.

[40]He began the relevant part of his judgment by considering what the adverb “tumultuously” meant. He said at 979G-980B:

“I would not attempt to give a full definition of any of those words, but it seems to me that all of them bring a certain impression to one's mind. When those words are applied to an assembly of persons, the impression is that the assembly should be of considerable size; that it should be an assembly in which the persons taking part are indulging in agitated movement; an excited, emotionally aroused assembly; excitement or emotion common to the members of the assembly; and generally, though not necessarily, accompanied by noise. I agree with the submissions which have been made, that it is a question of degree whether any assembly of people can properly be said to be acting tumultuously.”

[41]He continued at 980 B-G:

“I now turn to consider both the meaning of the words and the question as to whether the words “riotously and tumultuously” from their history are to be read as cumulative requirements, differing in character. Until very recently the victims of crime had, in general, no claim to be compensated for the injury they suffered as a consequence of the crime. Compensation for loss caused by riot was a special case. This raises the question: Why was it made a special case? If a crowd of people collect in angry and threatening fashion this should become obvious to the local forces of order, and it would then become their duty to prevent the crowd from becoming a riot. This is a duty which has been recognised for centuries, and which until the 19th century was put upon the local administrative area …; and there was a duty upon them to compensate for damage which was done by persons assembled riotously and tumultuously. The Act of 1886, in fact, did no more than modernise the mode of obtaining compensation and transferred the burden from the inhabitants … to the local police authority. There is nothing secret or furtive about a crowd of people who are acting riotously and tumultuously. In my judgment the word “tumultuously” was added to “riotously” for the specific purpose that it was intended to limit the liability of compensation to cases where the rioters were in such numbers and in such a state of agitated commotion, and were generally so acting, that the forces of law and order should have been well aware of the threat which existed, and, if they had done their duty, should have taken steps to prevent the rioters from causing damage.”

[42]The claim for compensation thus failed.

[43]It is clear from the second passage that the rationale for the obligation to compensate was that the assembly in question was one of which the forces of law and order ought to have been aware and ought to have prevented. Lyell J was not, so far as one can tell from the report, referred to the eighteenth century cases where the underlying principle of the Riot Act 1714 was said to be the transfer of liability from the rioters to the community on a principle analogous to suretyship. Lyell J also seemed to suggest that the test was whether the forces of law and order would have failed in their duty in not preventing the assembly or the ensuing damage. He used the language of fault (“if they had done their duty”), not that of notional fault. It is common ground, however, that an actual failure by, or fault on the part of, the forces of law and order is not a necessary ingredient of a successful claim. This was decided by Swift J in Pitchers v Surrey CC

[1922] 2 KB 57

(affirmed on appeal) in which he said at 64-5:

“The right to compensation does not seem to me to be in the least dependent on any action or inaction on the part of the police … It is a right which is given to the individual who is hurt by a riot to have his damages shared by the whole of the community, and the community for the purpose of this sharing is the police district within which the building injured or destroyed is situated.”

[44]Lyell J was referred to that case in argument but did not mention it in his judgment.

[45]It is true that Lyell J began the second quoted passage at 980B with the words “I now turn to consider both the meaning of the words and the question as to whether the words “riotously and tumultuously” from their history are to be read as cumulative requirements, differing in character.”, but the subsequent discussion does not in fact return to the meaning of the words. Rather, as we have said, it deals with the rationale underlying the presence of the word “tumultuously” in the statute and the purpose of its inclusion. Even his historical explanation differs to some extent from the historical explanation given by Lord Mansfield and Lord Loughborough.

[46]As both parties rightly accepted, the right to compensation is not fault-based. The liability to compensate is strict. Rix LJ made the same point in Yarl’s Wood Immigration at [54]:

“I can see no useful distinction for present purposes whether the 1886 Act is described as providing for no-fault compensation or as providing for liability premised on the notional responsibility of the police to maintain law and order. The police undoubtedly have a real and not merely notional responsibility for law and order (and within that responsibility a duty to deal with and if they can prevent riot) and there is no need for the 1886 Act to acknowledge that. What it seems to me is more significant for present purposes is that such a responsibility in terms of the 1886 Act is a strict one, however that is described: strict, notional, or not dependent on fault, blame or breach. What is the rationalisation of such a liability, obligation or duty? It seems to me that what Lord Mansfield had to say about that question, so much closer to the origin of the first Riot Act 1714, still retains pertinence, expressing as it does the common sense of the matter. It is for the sake of the party whose property has been damaged, it is to encourage the inhabitants (now the police force) of the locality, but including the party injured himself, all to assist in the preservation of the peace, it is to share the burden both of keeping the peace and of the misfortune of loss or injury. Moreover, as is so often the case with strict liability, it is because those who are liable to compensate are also regarded by the law as standing in the shoes of the wrongdoers themselves (as, for instance, in the case of the vicariously liable), in part because their obligation, their strict obligation, is to prevent what has happened happening.”

[47]Thus Rix LJ seems to have accepted that the rationale of the liability to compensate was not anything to do with fault (either real or notional) but to encourage the preservation of the peace. Importantly, he also treated the liability as a form of vicarious liability for the wrongdoers themselves. This too suggests the analogy with suretyship.

[48]MOPC placed some reliance on the decision of this court in Edmonds, so it is necessary to look at it in a little more detail. The case arose out of an armed robbery by three or four men of a jeweller’s shop in the Lanes in Brighton. Kenneth Jones J held that the shop owner was not entitled to compensation under the 1886 Act and this court upheld his decision. The robbers drove up in a van and stormed out, making a lot of noise. They attacked and smashed the shop window with a sledge-hammer and terrified those inside. Summarising the facts Kerr LJ said:

“One can see how the scene began and what it involved until the men ran away. It obviously began with the men stealthily driving up in a van and getting out. From that moment onwards they were undoubtedly shouting; there was a reference to war-cries and noise intended to frighten the bystanders. There was then more noise when they used the sledge-hammers on the window. But the main noise undoubtedly came from the terrified spectators. So it was, as the judge said, a scene of great commotion; I think he used the word “chaos.””

[49]The two points that were argued on the appeal were: i) whether the word “tumultuously” connoted an assembly of some considerable size; and ii) in so far as the word involved noise, whether it was sufficient if the assembly caused bystanders to behave tumultuously, in the sense of making a lot of noise, or whether it was necessary that the persons assembled together should themselves be the source of the noise.

[50]The argument on the first point was that “tumultuously” added some requirement of noise, but did not add any requirement about the number of persons assembled (riot in those days being capable of being committed by three or more persons). Kerr LJ was sceptical about approaching the phrase “riotously and tumultuously” sequentially, rather than interpreting it as a whole, but he nevertheless engaged with the argument. In addition to approving the decision of Lyell J, Kerr LJ said:

“The word comes from “tumulus,” a mound in one sense, or a swelling in another sense, and in the etymological dictionary it is defined as “excitement, uproar, agitation.”

But when one comes to the Oxford English Dictionary one finds that “tumultuous” is defined as “Full of tumult or commotion; marked by confusion and uproar; disorderly and noisy; violent and clamorous; turbulent.” If one then goes to the definition of the word “tumult,” since the first meaning of “tumultuous” is “Full of tumult or commotion,” one finds that the first definition of “tumult” is the following:

“Commotion of a multitude, usually with confused speech or uproar; public disturbance; disorderly or riotous proceeding”

. Rightly or wrongly – apparently etymologically wrongly, but I would say rightly as a matter of common usage nowadays – the connotation of “multitude” or “crowd” or “mob,” or of a large number of people, has been introduced into the meaning of “tumultuous” and “tumult.” That is certainly how I would construe it as a matter of first impression.”

[51]On the second point Kerr LJ regarded it as self-evident that whatever conduct or effect was involved in the adverb “tumultuously” it must relate to the persons who are assembled together, both “riotously” and “tumultuously,” and not to the bystanders.

[52]Commenting on the facts he said:

“The other point which Mr Poulton took was that Mr Justice Lyell’s suggestion that what matters is whether or not the police should have been alerted was satisfied in the present case, since there was a great deal of noise. He said that this should have attracted the attention of the police. I do not accept that. It is certainly not the kind of picture which Mr Justice Lyell had in mind when he gave his impression of a “riotous and tumultuous assembly.”

After all, this was not a case of a crowd which had assembled and which should have attracted the attention of the police by its very presence. On the contrary. These robbers drew up in a van quietly and then stormed out, no doubt making a good deal of noise, but in the minimum time possible; and they then vanished as quickly as they could. The whole nature of the raid was one of furtiveness at the beginning, and then surprise and speed of departure thereafter. In my view that is far from a “riotous and tumultuous assembly”.

[53]We do not regard these observations of Kerr LJ as laying down the legal proposition, binding on us, that the test is whether or not the police should have been alerted, whether actually or notionally. In the first place the point was not argued; in the second place Kerr LJ described Lyell J’s observation as a “suggestion” and no more; and in the third place the argument failed on the facts. Thus consideration of the legal test was unnecessary to the decision, because even if Lyell J’s “suggestion” provided the legal test, the appeal still failed.

[54]The issue in Bedfordshire Police Authority v Constable was whether a liability to pay compensation under the 1886 Act fell within the terms of an insurance policy which covered any sum which the assured was “legally liable to pay as damages”. Thus the issue was one of contractual interpretation. Although Longmore LJ referred to (and quoted from) Dwyer at [12], he did so in the context of asking the question:

“… why victims of riot who had suffered injury or damage to their property were made a special case at a time when victims of crime were not generally compensated for their injuries.”

[55]The eighteenth century cases were not before the court. Nevertheless Longmore LJ went on to consider the nature of liability under the 1886 Act. At [24] he said:

“If I am asked whether the liability under the 1886 Act is a liability which falls to be paid “by reason of some breach of duty or obligation” … I can only say that it is. It is a liability that arises from the fact that the police are responsible for the preservation of law and order in the area of their police authority.”

[56]The key word here is “responsible” as Longmore LJ went on to explain at [26] and [27]:

“[26] Once one appreciates that the reason for the 1886 Act placing the burden of paying compensation to the victims of riot damage on the police authority is that the police are responsible for law and order and that they are (notionally) in breach of that responsibility, it seems to me, as an English lawyer, that compensation payable is a sum which the police authority is ‘liable to pay as damages’….. [27] It may be objected that the concept of responsibility in the previous paragraph is different from and an extension to the concept of ‘some breach of duty or obligation’ as used by Sir Wilfrid Greene MR in the Hall Brothers case. But it is not. Torts of strict liability are based on the concept of responsibility. As a matter of language it is somewhat unnatural to say that the land owner in Rylands v Fletcher, the common carrier or the innkeeper are in breach of duty when their obligation is to be strictly answerable because of the responsibility inherent in their respective positions. But it cannot be doubted that their liability is a liability in damages and that sums paid by them are sums which they are ‘legally liable to pay as damages’. The same applies to statutes which give rise to strict liability such as the 1920 Act and the Animals Act 1971. The 1886 Act is no different.”

[57]Longmore LJ thus held that liability did not depend on some breach of duty (real or notional) but on the concept of responsibility. In our view, this is also akin to the explanation given by Lord Mansfield and Lord Loughborough to the effect that the community is responsible for the wrongdoing of rioters within their midst.

[58]In Yarl’s Wood Immigration Rix LJ referred to Dwyer at [49]. He said that the claim failed in Dwyer because Lyell J had held that “tumultuously”:

“…had to be given its own separate weight, connoting an assembly of considerable size and agitated movement, and generally, even if not necessarily, accompanied by noise.”

[59]This is a summary of the first of the quoted passages from Dwyer. Rix LJ went on in the same paragraph to explain the second of the quoted passages as “the historical underpinnings of the 1886 Act”, but the theme running through his judgment is that of the community being answerable for the wrongdoing of the rioters, and not on breach of any notional duty. Thus at [39], referring to Mason v Sainsbury, he said:

“In any event, the hundred's liability is strict, and does not turn on blame or negligence. As Buller J put it, they answer for the trespassers, not for their own fault. Or as Lord Mansfield put it, the hundred is put “in the place of the trespassers”.”

[60]At [42] approving the decision of Swift J in Pitchers he said:

“However, the right of compensation did not depend on what the police did or did not do or on any principle of blame.”

[61]His conclusion is at [54] in a passage that we have already quoted at [46] above. What is important is that Rix LJ endorsed the reasoning of Lord Mansfield in the very early days of the 1714 Act and applied it to the 1886 Act.

[62]In our judgment these statements of the underlying rationale explain why the statute uses the expression “riotously and tumultuously” and why the liability is strict. They do not (and Lyell J’s statement of the underlying rationale does not) explain what that expression means. Lyell J did that (in terms approved in Edmonds) in the passage at 979G-980B which we have already quoted. If and in so far as he meant to suggest that the test is whether the police should, even notionally, have prevented the damage, we do not consider that to be correct and we do not consider that there is any authority that binds us to hold that it is. However, even if it is the right test, the judge found that it was satisfied on the facts. We make that point because whether persons are “riotously and tumultuously” assembled, which is a factual assessment, should not be allowed to turn into a counter-factual inquiry into whether the (or a hypothetical) police force could have prevented the damage. What assumptions are to be made about the police? Are they notionally equipped with unlimited manpower? What intelligence do they have? Are they to be assumed to be shadowing the assembly before it has turned into a riotous and tumultuous assembly, or from the point of time at which it becomes tumultuous without being riotous? Or are they to be assumed to act instantaneously once the assembly becomes riotous and tumultuous? Where is the notional police officer to be found? On the facts of our case does he follow the group as they set off from the playground armed with petrol bombs and other weapons, or is he waiting at the warehouse?

[63]Nor would such an inquiry in any event be very helpful. If the police had been tipped off about the planned armed robbery of the jeweller’s shop in Brighton in Edmonds and had been present at the scene when the robbers stormed out of the van, no doubt they could (and perhaps should) have done something to stop them. In addition a counter-factual inquiry of that kind comes very close to asking whether the police were in some way at fault, but clearly that is not the test.

[64]In Dwyer Lyell J had observed that:

“There is nothing secret or furtive about a crowd of people who are acting riotously and tumultuously.” and in the course of his judgment in Edmonds, commenting on the facts, Kerr LJ said that “ the nature of the raid was one of furtiveness at the beginning, and then surprise and speed of departure thereafter.”

That, he said was “far from” a riotous and tumultuous assembly, but it must be borne in mind that he was simply drawing a contrast on the facts of that case. A lack of furtiveness is not part of the statutory definition or, indeed, the judicial exegesis of the meaning of “tumultuously”.

[65]Mr Grodzinski submitted that the hallmark of a riotous and tumultuous assembly was that it was (or was potentially) confrontational. The confrontation might be confrontation with the authorities (e.g. the police) or with a rival or opposing group (e.g. a rival gang) or with an idea (e.g. anti-capitalist riots). However, there is no trace of this test (or hallmark) in the authorities. Given that riot may be committed where the only violence is violence towards property and in a private place, the supposed hallmark cannot be found in the word “riotous”. Is it part of the meaning of “tumultuous”? It did not form any part of Lyell J’s explanation of the word and we were shown no other material to support the argument. If a frenzied mob attacks and destroys an empty building we do not see why it should fall outside the description “riotous and tumultuous”. We would therefore reject this submission. In addition, in the present case the warehouse was guarded by the security guard, Mr Mireku, who ran away when he heard the break-in and hid in the bushes. Had he remained or gone to investigate there would potentially have been a confrontation. Mr Thompson, who saw the group leaving the warehouse, was in fact confronted by one of them and punched on the side of the head. The result was that he felt intimidated and scared.

[66]There is one final point that needs to be made in view of some of the submissions that Mr Grodzinski advanced. As section 1(5) of the Public Order Act 1986 makes clear, a riot may take place in a private place. The Yarl’s Wood case (where the activities took place in an immigration detention centre) stands for the proposition that persons may also be “riotously and tumultuously” assembled in a private place, albeit that the detention centre was a “public institution”. If there is noise, we do not consider that it is necessary that the noise be audible outside the place where the riotous and tumultuous assembly takes place.

[67]In short, the focus of the inquiry is whether property has been damaged or destroyed as a result of mob violence. As Lyell J rightly said, whether an assembly is “riotous and tumultuous” is a question of degree. It is for the trial judge to carry out an evaluative exercise to decide that question in the light of the primary facts found.

(b) The judge’s reasoning on liability

The judge summarised his conclusions on the legal test at [28] in which he said:

“(1) There must be a riot within the meaning of section 1 of the Public Order Act 1986…. (2) The assembly must be of some size, certainly more than three or four persons. [This criterion] was satisfied in the present case, given that there were in the region of twenty to twenty five people. (3) … the persons assembled must be acting in an agitated, excited, volatile manner, usually … also making a noise, rather than acting stealthily, so that it can be said that their riotous behaviour could, at least notionally, have been prevented by the police. However, Edmonds is not authority for the proposition that unless the crowd is making a tremendous noise, they cannot be riotously and tumultuously assembled together if other indicia of a riotous and tumultuous assembly are present. The real touchstone is that there must be some “public” element to the behaviour, what Mr Crane QC for the insurer claimants described as a perceived or palpable threat of a riot to which the police could, notionally, have responded. (4) Those cases where the courts have held that the victims of riot are entitled to compensation under the 1886 Act are ones which involve the rioters engaging in wanton damage to property or, as Mr Crane QC put it, where the rioters exhibit an animus towards the property in question and are not simply looting in order to steal.”

[69]Having made the findings of primary fact summarised above, the judge reached his conclusions at [69] to [77]. Those which are no longer in dispute can be stated very briefly, while those that remain in dispute deserve to be quoted in full.

[70]First, there was a riot within the meaning of the Public Order Act 1986. This finding is not challenged on this appeal.

[71]Second:

“71. … the behaviour of the group of youths was agitated and volatile when they were gathering on the playground, as they were moving towards the business park and when they were attacking the Warehouse. The incident itself, as revealed on the CCTV footage, was a frenetic, agitated, chaotic process, with some youths almost dancing around outside whilst others were smashing their way into the building with a variety of implements, then a large group swarming into the building once entry was achieved. They then looted some of the contents but only so much as they could carry (in some cases more than they could carry since there is evidence of items dropped outside), not returning to loot more, since rather than simply empty the building of its contents, they set fire to it.”

It was a planned attack but not a systematic or sophisticated burglary. He continued:

“72. … the behaviour of the group throughout the period when they were gathering on the playground, when they were moving towards the Warehouse, when they were smashing down the door, when they were looting and setting fire to the Warehouse and when they were making their escape, some of them dropping what they had looted, can properly and objectively be described as the behaviour of an agitated, excited and volatile group, not the behaviour of a gang of professional thieves.”

[72]Third:

“73. As for Mr Grodzinski QC’s seventh point about there having been insufficient noise to amount to a tumult, there was undoubtedly noise from the smashing down of the door, heard by the Premier Inn staff and by Mr Mireku. Whilst there is no direct evidence of any loud talking or shouting and the CCTV has no soundtrack, it seems to me, looking at the footage of the group moving around frenetically and in an agitated manner outside the building as they are forcing an entry, that it is unreal to suppose that there was not some talking or shouting. In any event, even if there was no other noise than the smashing down of the door, as I have already said at [28(3)] above, Edmonds is not authority for the proposition that, unless the crowd or mob is making a tremendous noise, they cannot be “persons riotously and tumultuously assembled together”, if other indicia of riotous and tumultuous assembly are present, as they are in the present case.”

[73]Fourth:

“74. … on 8 August 2011, there was undoubtedly a perceived or palpable threat of rioting in Enfield and specifically in the vicinity of Enfield Island Village and the Innova business park, both in terms of what local people feared and the rumours flying about and in terms of the conduct of the gang or gangs of youths who came together that night. In contrast to the armed robbers in both Dwyer and Edmonds, who appeared by stealth as if from nowhere and committed the crimes in a matter of minutes, the evidence is that these youths were openly gathering on the playground up to two hours before the incident, some of them recognised by local people as both residents of the Village and members of the Get Money Gang or Black Gang (see [38] to [40] above). They were making no attempt to hide what they were doing. They then moved quite openly through the Village and over the bridge before turning right into the fields to go to the business park. I reject Mr Grodzinski QC’s suggestion that they were acting furtively at any stage: the route over the fields was simply the quickest way to the Warehouse.”

[74]The judge amplified his fourth reason in the succeeding parts of his judgment. He noted that the group was gathering quite openly on Enfield Island Village some two hours before the incident and was then seen moving towards the business park fifteen minutes or more before the incident. That, he said, demonstrated the palpable threat to which the police could have responded. The behaviour of the group from the time they were gathering on the playground onwards, was such that the police ought to have been aware of their presence and could have prevented the incident which eventually occurred, even if that incident took place very quickly in just over three minutes.

[75]Fifth:

“77. … whatever the true motive for the use of petrol bombs to set fire to the Warehouse, objectively, the use of petrol bombs evidences wanton violence towards the property damaged or destroyed. This wanton violence towards property is a hallmark of riotous and tumultuous behaviour and is a paradigm example of the situation where the victims of the damage or destruction should qualify for compensation under the 1886 Act…”

The role of the appeal court on questions of fact

[76]As CPR Part 52.11(1) makes clear, the general rule is that an appeal is limited to a review of the decision of the lower court. In addition appeal courts have been repeatedly warned in recent cases at the highest level not to interfere with findings of fact by trial judges, unless compelled to do so. This applies not only to findings of primary fact, but also to the evaluation of those facts and to inferences to be drawn from them. The best known of these cases are: Biogen Inc v Medeva plc

[1977] RPC1

; Piglowska v Piglowski

[1999] 1 WLR 1360

; Designer Guild Ltd v Russell Williams (Textiles) Ltd

[2000] UKHL 58

, [2000]1 WLR 2416; Re B (A Child) (Care Proceedings: Threshold Criteria)

[2013] UKSC 33

[2013] 1 WLR 1911

and most recently and comprehensively McGraddie v McGraddie

[2013] UKSC 58

[2013] 1 WLR 2477

. These are all decisions either of the House of Lords or of the Supreme Court.

[77]That is the approach to be applied in the present case.

(c) MOPC’s case on appeal

[78]Basing himself principally on the authorities we have mentioned, Mr Grodzinski submitted that a “riotous and tumultuous” assembly is one which: i) is of a considerable size; ii) is excited and emotionally aroused; and iii) is behaving in such an agitated, volatile, noisy, angry and threatening manner that it should be obvious to the police that something needs to be done. That is to be contrasted with a group acting secretly or furtively.

[79]For the reasons we have given we do not consider that the test is whether the group’s behaviour is such that it should have been obvious to the police that something needs to be done. That said, as we see it the difference between Mr Grodzinski’s submission and the judge’s formulation of the principle at [28] (3) is that Mr Grodzinski’s submission entails that the assembly is actually behaving in an agitated etc manner. The mere threat of a riot is not enough. In our judgment there is considerable force in that submission. However, in the first part of [28] (3) the judge had expressly directed himself that the persons assembled must “be acting in an agitated [etc] manner”, so we are not convinced that his reference to the perceived threat of a riot makes any real difference. Moreover, it is difficult to see how this affects the judge’s findings of fact, to which we shall come.

[80]Second, Mr Grodzinski submitted that the judge was wrong to place weight on what the gang were doing some hours earlier at the playground and some distance away. At that point they were not “riotously and tumultuously assembled” and the police would not have had the power to disperse them, because there was no immediate threat of a breach of the peace: Laporte v Chief Constable of Gloucestershire

[2007] 2 AC 105

. Even if they were “riotously and tumultuously assembled” at the playground, that did not bear on the question whether they were “riotously and tumultuously assembled” at the warehouse.

[81]Strictly speaking, all that matters for the purposes of the Act is whether the group which attacked the warehouse was riotously and tumultuously assembled at the time they caused the damage, a question on which the CCTV recording provides the best evidence. Nonetheless, if, as we think the judge found, there was in effect one continuous incident which began at the playground and ended at the warehouse, he was, in our judgment, entitled to consider evidence of how the group were behaving at the playground. Their behaviour there and on the way to the warehouse was relevant to the questions(a) whether were they tumultuously assembled and

(b) (to the extent that it is relevant) whether were they acting furtively. Moreover, by the time the group were on the way to the warehouse they must have already armed themselves with the petrol bombs and with the weapons that they would use to break-in. As the judge found, they were seen with the weapons by the Thomases and by at least one other member of the public. Even if the group’s behaviour at the playground ought to be left out of account, the manner in which the group approached the warehouse seems to us to have been a relevant feature of their behaviour. Mr Grodzinski submitted that the temporal factor was very important. He stressed the fact that the activities at the warehouse lasted for no more than about three and a half minutes and they were nowhere near any other similar incident. However, the attack on the warehouse did not come out of the blue. It was an attack by an armed mob which had travelled some distance to reach it, took fifteen minutes to do so, and attracted the attention of at least three members of the public. That is another reason why, in our judgment, the judge was entitled to have regard to the group’s behaviour on the way to the warehouse. But even if the judge was wrong to have considered the group’s behaviour either at the playground or on the way to the warehouse, he recognised that the real question was whether they were “riotously and tumultuously assembled” at the warehouse itself and that is the question that he answered. We do not think that it matters whether the police could lawfully have dispersed the group when they were at the playground. In the first place, as Mr Grodzinski submitted, dispersing the group at that point would not (or not necessarily) have prevented the attack on the warehouse. Second, this submission is in danger of requiring a counter-factual inquiry into what the police could (or should) have done, which we do not think is appropriate. Third, even if such an inquiry were warranted, the notional preventative measures that the police could have taken would have included protecting the warehouse by a police presence there. Fourth, the presence of an armed group would surely have attracted the attention of the notional police force, as they attracted the attention of the Thomases and the other member of the public.

[82]In addition, it is impossible to see how Mr Grodzinski’s submissions, even if correct, undermine the judge’s critical findings of fact about the attack on the warehouse itself. They were: i) The behaviour of the group of youths was agitated and volatile when they were attacking the warehouse. ii) The incident itself was a frenetic, agitated, chaotic process. iii) The behaviour of the group when they were smashing down the door, when they were looting and setting fire to the warehouse and when they were making their escape can properly and objectively be described as the behaviour of an agitated, excited and volatile group. iv) They were making no attempt to hide what they were doing. They were not acting furtively at any stage. v) The use of petrol bombs evidenced wanton violence towards the property damaged or destroyed.

[83]Mr Grodzinski next submitted that the judge was wrong to find that the group’s behaviour in the warehouse was sufficiently agitated and volatile to amount to a “riotous and tumultuous” assembly. He says that it was no more agitated or volatile than the activities of the robbers in Dwyer or Edmonds. This submission overlooks a number of the judge’s other findings about the group’s behaviour at the warehouse. Unlike the robbers in Dwyer and Edmonds the group’s behaviour was also frenetic and chaotic. Mr Grodzinski invited us to view the CCTV footage and form our own conclusion since, he said, we were in as good a position to do that as the judge. We have all viewed the CCTV footage, but, as we have said, this court does not without special reason rehear an appeal. It conducts a review of the judge’s findings. The judge’s findings are, in our judgment, findings that were clearly open to him in the light of the CCTV evidence. The judge also found that the group were wantonly violent towards the building itself, having petrol bombed it within a few minutes of their entry. Mr Grodzinski criticised this latter finding on the ground that the judge should have accepted the evidence of Detective Superintendent Hibberd that the petrol bombing was not wanton violence but was an attempt to cover up forensic evidence. However, the judge heard him give his evidence and did not accept it. This court could not possibly reverse the judge’s evaluation of the live evidence.

[84]In our judgment the judge’s findings of primary fact at [45] (which are borne out by the CCTV evidence) amply justify his conclusions on this point. In short, it was mob violence.

[85]Mr Grodzinski argued that the judge wrongly found that the mere noise of the break-in combined with some talking or shouting by the group was sufficient to amount to tumult. He says that the principle underlying the authorities is that the noise should be of a type that distinguishes the group from those who are merely breaking and entering. However, the judge was entirely correct in saying that in Dwyer (approved in Edmonds) Lyell J said that a tumultuous assembly would be “generally, though not necessarily, accompanied by noise.” Although the judge found that there was some noise which was both audible (and in fact heard) outside the warehouse, that was not the determinative reason for his finding. Moreover, the group were not “merely” breaking and entering; they were, as the judge held, “attacking” the warehouse. It is also, in our judgment, relevant to consider the effect of the noise on those who were present at the scene (or in the immediate vicinity). As the judge found at [46], the noise caused Mr Cass to go back into the restaurant of the Premier Inn to tell the staff to shut the blinds and lock up the building. On hearing the noise Ms Charles immediately thought that it could be rioters looting the warehouse. Both these findings of primary fact are supportive of the judge’s conclusion that the group were “tumultuously” assembled.

[86]Mr Grodzinski said that the judge was wrong in placing any weight on the fact that the attack was unsophisticated. However, the point that the judge was making was that the nature of the attack was excited, agitated and ill-disciplined, in other words that it was tumultuous.

[87]Mr Grodzinski criticised the judge’s finding that the group did not act furtively at any stage. He pointed to the fact that(a) the warehouse was on a quiet industrial estate away from the town centre

(b) the group approached the warehouse under cover of darkness and by crossing a field rather than by road

(d) they carried out their robbery and arson as quickly as they could and

(e) the warehouse must have been targeted because of the goods inside it. The attack was not therefore a random and wanton attack.

[88]It is as well to remember that the yardstick of “furtive” behaviour is no more than a judge-made expression of a contrast between behaviour that is “tumultuous” and behaviour that is not. We do not think that the location of the warehouse is material to the question whether the group were riotously and tumultuously assembled. A group may be riotously and tumultuously assembled even in a remote location. Moreover, as it happens there was a large hotel with a bar and restaurant a short distance away from the warehouse. Nor do we think that the fact that the attack took place at night bears on the question (and in any event the exterior of the warehouse was lit). Whether the individual participants could be recognised is also irrelevant, since the focus of the inquiry is on what they did rather than on what they looked like. The speed of the incident is a relevant factor, but it is not one that the judge ignored. Lastly, although the attack on the warehouse was, no doubt, targeted, it seems unlikely that it took place simply with a view to theft, bearing in mind that the warehouse was vast and that the group in fact took very little from it. As Mr Crane put it, the crime was “expressive” rather than “acquisitive”. In addition, one of the ingredients of “riot” is a common purpose – see the Public Order Act 1986 s.1(1) and s.1(3) – so the fact that the warehouse was targeted does not detract from the judge’s conclusion. What matters more is the way in which it was attacked.

[89]The last substantive point that Mr Grodzinski made is that, whether or not the judge was right to find that the group were wantonly violent towards the warehouse, wanton violence is not the test. The test, he said, is what should (notionally) have been apparent to the forces of law and order. In our case, he said, the police cannot have been expected to predict, even notionally, that the warehouse would be set on fire, or to have taken steps to prevent it. For the reasons we have given we do not consider that the test is whether the police could notionally have prevented the incident. That is merely the rationale for the inclusion of the word “tumultuously” in the statutory test. The fire bombing of the warehouse was all of a piece with the initial attack, just as the destruction of Mr Ratcliffe’s furniture was all of a piece (“one continued act”) with the attack on his house. This point therefore fails too.

(d) Conclusion

[90]In short, the judge carried out an evaluative exercise on the basis of the primary facts that he found. He directed himself correctly in law and the result of his evaluation was one to which he was entitled to come. It is the sort of evaluation that an appeal court should be very reluctant to overset. We therefore dismiss the appeal on the question of liability. Damages

[91]Earlier in this judgment we described the historical and legal background to the 1886 Act. What emerges is that, prior to the 1714 Act, a person who had suffered property damage as a result of riotous and tumultuous behaviour would have been able to recover damages for all his loss from the trespasser. We see no reason why this should not have included his consequential losses. The effect of the Riot Act 1714 was that all losses recoverable from the trespasser were now recoverable from the hundred. It has (rightly) not been suggested that, if consequential losses were recoverable under the 1714 Act, they ceased to be recoverable by virtue of the 1827 Act.

(a) Section 2(1) of the 1886 Act

[92]Mr Crane submitted that on the plain wording of section 2(1) of the 1886 Act, the compensation recoverable includes all losses consequent on damage to property, subject to the usual common law rules of causation and remoteness. The physical damage or destruction of the property is the trigger for liability to pay compensation under section 2(1), but, on the natural reading of the words, all losses proximately caused by such damage or destruction are recoverable.

[93]We begin with a linguistic analysis of the words of section 2(1) without reference to other provisions of the 1886 Act or the antecedent history or any other considerations. Mr Grodzinski submitted that the words “loss by such injury, stealing or destruction” can only be a reference back to damage to the building (or the property inside it). He said that there is, by contrast, no reference to any wider concept of the consequences of the acts giving rise to the claim, or to the consequences of any property damage. We agree that the injury or destruction is a reference back to damage to the building, but the question is whether the consequences of the property damage are capable of being loss by such injury or destruction. In our view, they are. If a person owns a building which is let to a tenant who pays rent and as a result of its being damaged, he incurs the cost of repairing it and a loss of rent whilst it is being repaired, the loss he suffers as a result of the damage to the building is(i) the cost of the repairs and

(ii) the loss of rent. As a matter of ordinary language, both (i) and (ii) are losses that he has sustained “by such injury”. In our view, there is nothing in the wording of section 2(1) itself which supports the proposition that the loss sustained cannot include consequential losses caused by the injury or destruction of the property.

(b) The preamble and section 7

[94]Tbe preamble to the 1886 Act as originally enacted (repealed by the Statute Law Revision Act 1898) provided as follows:

“Whereas by law the inhabitants of the hundred or other area in which property is damaged by persons riotously and tumultuously assembled together are liable in certain cases to pay compensation for such damage, and it is expedient to make other provision respecting such compensation and the mode of recovering the same…”

[95]Section 7 provides:

“For the purposes of this Act— (a) where a church or chapel has been injured or destroyed, or any property therein has been injured, stolen, or destroyed, the churchwardens or chapelwardens, if any, or, if there are none, the persons having the management of such church or chapel, or the persons in whom the legal estate in the same is vested; and (b) Where a school, hospital, public institution, or public building, has been injured or destroyed, or any property therein has been injured, stolen, or destroyed, the persons having the control of such school, hospital, institution, or building, or the persons in whom the legal estate in the same is vested; shall be deemed to be the persons who have sustained loss from such injury, stealing, or destruction, and claims may be made by any one or more of such persons in relation both to the building and to the property therein, and payment to any such claimant shall discharge the liability of the compensation authority to pay compensation, but shall be without prejudice to the right of any person to recover the compensation from such payee.”

[96]Flaux J said at [113] that the fact that compensation was limited to physical damage to the premises or the property in it was made “absolutely clear” by the preamble to the 1886 Act and further supported by section 7. Although the preamble was repealed by the Statute Law Revision Act 1898, we accept that it can in principle be relied on as an aid to the true interpretation of the 1886 Act. The preamble says two things. First, it refers to the pre-existing law (“whereas by law the inhabitants of the hundred...are liable in certain cases to pay compensation for such damage”). Secondly, it states that “it is expedient to make other provision respecting such compensation and the mode of recovering the same”.

[97]The judge said that the first part of the preamble sheds light on the meaning of the 1886 Act. He was of the view that compensation “for” property damage meant compensation for the cost of repairing property damage or the diminution in value of the property caused by the property damage, but we see no reason to construe the preamble in that limited way. As a matter of ordinary language, we consider that it includes all compensation necessary to make good the loss caused by the damage to the relevant property. Let us return to the example we have given above of the rented property which is damaged in a riot. If the owner incurs the cost of repairing the property and a loss of rent whilst it is being repaired, the loss he suffers as a result of the damage to the building is the cost of repairs and the loss of rent. Compensation for both losses is compensation “for” the damage. In any event, since the recital was reflective of the pre-existing law, it should be interpreted in accordance with that law. It would appear that the wording of the preamble was taken from the language of article II of the 1827 Act which provided for compensation “for the Damage so done”.

[98]Mr Grodzinski submitted that since there were differences between the 1886 Act and its predecessor legislation we should interpret the 1886 Act restrictively. However, in Hyde v Cogan Ashurst J said that since the purpose of the Riot Act was remedial “it ought to receive a liberal construction.” So, too did Buller J in the same case, adding that “the most extensive sense must prevail”. The liberality of interpretation is borne out by the decision in Ratcliffe v Eden because Mr Ratcliffe recovered for the loss of his furniture, even though furniture was not mentioned in the Act. The 1886 Act is also remedial and we see no warrant for reversing this approach to its interpretation.For the reasons already given, there is no reason for saying that the entitlement to compensation under the 1827 or 1886 Act differed in any material way from the entitlement to compensation under the 1714 Act.

[99]As regards section 7, the judge said at [114] that the words “in relation both to the building and to the property therein” are expressly and in their context referring to damage to the building or the property in it. They do not have any wider meaning. Mr Grodzinski also made the point that there is no mention of claims for any other type of damage, such as loss of rent, or other consequential loss. In our view, section 7 does not carry the argument any further. The purpose of the section is to identify who may be a plaintiff (or claimant) for the purposes of the 1886 Act where(a) specified buildings have been injured or destroyed or

(b) any property therein has been injured, stolen or destroyed. It does not purport to prescribe what losses may be recovered. We accept that, if section 7 had been clearly drafted on the basis that losses recoverable under section 2(1) did not include consequential losses, that would be a powerful pointer in favour of the exclusion of such losses, but we do not consider that it was so drafted. Section 7 speaks of persons who “have sustained loss from [the injury to the building or property therein]” claiming compensation “in relation both to the building and the property therein”. It is a normal use of language to say of a person whose building has been damaged and has suffered a loss of rent while it is being repaired that the loss that he has sustained from the damage to the building is both the cost of repair and the loss of rent. Both heads of loss derive from the physical damage caused to the building. The claims for compensation for these losses are both “in relation to the building”.

[100]It follows that we cannot agree with the judge that the answer to what he described as a “short point of construction” is to be found in construing section 2(1) by reference to the preamble and section 7. Neither the preamble nor section 7 supports the interpretation of section 2(1) as meaning that loss sustained by damage to a building may not include consequential losses suffered as a result of the damage. (c) Construction of section 2(1) in the light of the previous history

[101]The judge said that, in construing the 1886 Act, no assistance could be derived from the legislative history to which we have referred. That may be right if the previous legislation is considered without regard to the case-law which interpreted it. In our view, however, if the case-law is taken into account (as it must be), the history does cast light on the meaning of section 2(1). It is clear that the object of the previous legislation was to secure that, where buildings and property were destroyed or damaged by the acts of trespass of rioters, those who suffered loss as a result of the destruction of or damage to their property received from the hundred the compensation which they were previously entitled to recover from the trespassers. The hundred stood as “sureties” for the trespassers (per Lord Mansfield in Ratcliffe v Eden); the liability for damages of the trespassers was “transferred” to the hundred (per Aston J in Ratcliffe v Eden and Buller J in Hyde v Cogan); and the entitlement of the injured party to recover damages for “all his loss” from the trespassers was replaced by a “substituted” right of action against the hundred (per Lord Loughborough in Wilmot v Horton). The purpose of the 1886 Act was to make “other provision respecting such compensation and the mode of recovering the same”. Save to the extent that it made express provision for the liability of the compensating authority not to be co-extensive with that of the trespassers, there is nothing in the Act to indicate that Parliament intended to change the law so as to deprive claimants, as a matter of principle and in all circumstances, of the right to compensation for loss consequent on physical damage. Finally, we should add that we agree with the statement by Rix LJ in Yarl’s Wood at [54] that what Lord Mansfield had to say about the rationale of the liability of the local inhabitants (now the police) to pay compensation “still retains pertinence”.

(d) Other pointers

[102]The next question is whether there is anything elsewhere in the 1886 Act which shows that, despite(i) the apparently broad language of section 2(1) and

(ii) the history to which we have referred, consequential losses are nevertheless not recoverable under the Act. Mr Grodzinski pointed to a number of other factors which he said support that conclusion.

[103]He submitted that there are several difficulties with the proposition that police compensation authorities are to be regarded in all relevant respects as “stepping into the shoes” of rioters under the 1886 Act. First, Parliament described the compensation payable using the language of “discretion”:

“... shall, if satisfied, fix such compensation as appears to them just” (section 3(1)). That is to be contrasted with damages flowing from common law torts which are awarded as of right. Secondly, section 2(1) states that, in fixing the amount of compensation, “regard shall be had” to the conduct of the person whose property has been destroyed or damaged “whether as respects the precautions taken by him or as respects his being a party or accessory to such riotous or tumultuous assembly, or as regards any provocation offered to the persons assembled or otherwise”

. This is an exercise which is very different from what was permissible at common law at the time of the enactment of the 1886 Act and even that which is permissible today, despite the changes to the general law introduced by the Law Reform (Contributory Negligence) Act 1945. Thirdly, compensation under the 1886 Act is limited to the particular situations which are described in the Act itself. They do not include compensation for personal injury. Even within the more limited category of property damage, not all situations are covered. For example, the Act does not cover compensation for damage caused by rioters to vehicles. Fourthly, the conditions for claiming compensation are those that have been set out by the Secretary of State: see section 3(2). The defendant can only be sued once certain preconditions have been met: see section 4(1).

[104]We accept that the effect of the 1886 Act is that the liability of the compensating police authority is not coterminous with that of the trespassers. The object of the Act was, as the preamble stated, “to make other provision respecting such compensation and the mode of recovering the same”. There is no doubt that it did introduce some changes tothe legislative scheme. It is not profitable to discuss whether it is apt to characterise the award of such compensation as appears to be “just” as an exercise of discretion, as it was said to be in this court in Yarl’s Wood. We doubt whether it can properly be described as an exercise of discretion in the strict sense, but the important point, as Mr Grodzinski said, is that it is quite different from an assessment of damages at common law.

[105]In our view, however, Mr Grodzinski’s points do not shed light on the meaning of section 2(1). It is clear that the compensation payable under section 2(1) does not precisely replicate the damages for which the trespasser who destroys or damages a building would have been liable at common law before the 1714 Act, but it does not follow that the compensating authority does not stand in the shoes of the trespasser for any purpose. In the light of the history to which we have referred, the better interpretation is, as we have said, that in relation to loss sustained by destruction of or damage to a specified building, the authority stands in the shoes of the trespasser save to the extent that the 1886 Act otherwiseprovides. There is nothing to suggest that the fundamental “standing as sureties” principle embodied in the 1714 Act was completely swept away by the 1886 Act. Damages for personal injury were not recoverable under any of the previous legislation. Accordingly, the fact that damages for personal injury cannot be claimed under the 1886 Act says nothing about the scope of compensation that is recoverable for loss sustained by destruction of or damage to property under section 2(1). Similarly, the fact that the compensation recoverable in respect of property damage is such as “appears” to the compensating authority to be “just” says nothing about whether it may or may not include compensation for consequential loss. The same point can be made in relation to the fact that section 2(1) provides that, in fixing the amount of compensation, regard shall be had to the claimant’s conduct. The fact that this was impermissible at common law says nothing about whether compensation may be paid for consequential loss. We therefore conclude that there is nothing elsewhere in the 1886 Act which shows that consequential losses cannot be awarded under section 2(1).

(e) The 1886 Regulations

[106]Regulations prescribing the manner in which claims under the 1886 Act were to be presented (“the 1886 Regulations”) were made under section 3(2) of the 1886 Act shortly after it was passed. Regulation 2 required any claim to be submitted within 14 days of the date on which any injury, stealing or destruction took place. Regulations 3 and 4 included various formal requirements for a claim for compensation. Regulation 5 required the claim to state separately the sums claimed for(a) destruction of premises;

(b) injury to premises;

(d) theft or destruction of other property in or on the premises, distinguishing as regards (c) and (d) property belonging to the claimant from property belonging to others in his care. Regulation 5 did not mention the possibility of claiming any sums for consequential loss such as loss of profit. Regulation 6 required the claim to state the nature of any injury to premises or property therein, the cost of any repairs and/or an estimate of the cost of specified repairs required. It did not require any particulars or evidential material in relation to the economic consequences of any such injury to premises or property. Amendments were subsequently made to the Regulations but none of them was material to the issue with which we are concerned.

[107]Mr Grodzinski submitted that, if section 2(1) of the 1886 Act is ambiguous, the 1886 Regulations decisively provide a clear Parliamentary or administrative contemporanea expositio which should be taken into account in deciding what it means. The judge said at [118] that, because the wording of section 2(1) was clear, it was not necessary or appropriate to have regard to the 1886 Regulations, but he added that, if the wording was ambiguous, it would be permissible to have regard to the Regulations which were made only a short time after the Act. The Act was passed on 25 June 1886 and the Regulations were made on 28 July 1886. He said:

“118. . . . The Regulations make it quite clear that the loss which is to be compensated is only physical loss and damage, hence the limited categories of claim set out in Regulation 5. That conclusion is also reinforced by the short period of fourteen days prescribed for bringing a claim in Regulation 2, which is inconsistent with an entitlement to claim for consequential losses which would invariably take a longer period both to eventuate and to be computed. Thus, if the statute were ambiguous, the Regulations are consistent with a narrow interpretation of what damage is recoverable and, in accordance with Lord Lowry's third principle, confirm that interpretation.”

[108]His reference to Lord Lowry’s third principle was to the following passage in Hanlon v The Law Society

[1981] AC 124

at p 193H:

“A study of the cases and of the leading textbooks … appears to me to warrant the formulation of the following propositions: (1) Subordinate legislation may be used in order to construe the parent Act, but only where power is given to amend the Act by regulations or where the meaning of the Act is ambiguous. (2) Regulations made under the Act provide a Parliamentary or administrative contemporanea expositio of the Act but do not decide or control its meaning: to allow this would be to substitute the rule-making authority for the judges as interpreter and would disregard the possibility that the regulation relied upon was misconceived or ultra vires. (3) Regulations which are consistent with a certain interpretation of the Act tend to confirm that interpretation. (4) Where the Act provides a framework built on by contemporaneously prepared regulations, the latter may be a reliable guide to the meaning of the former. …”

[109]Mr Grodzinski adopted the reasoning of the judge at [118]. The 1886 Act and the 1886 Regulations were roughly contemporaneous with each other. They could all be regarded as “part of the same legislative exercise”: see per Lord Hope in R (A) v Director of Establishments of Security Service

[2010] 2 AC 1

at [42]. The Regulations do not contemplate or allow for compensation for claims for consequential loss. Regulations 5 and 6 make it clear that the loss which is to be compensated is restricted to the cost of making good physical loss and damage: see [106] above.

[110]In our judgment, the 1886 Regulations cannot be relied on to interpret section 2(1). For the reasons already given, we do not consider that the meaning of the subsection is ambiguous. That is fatal to any reliance on the Regulations, but even if there is some ambiguity, we do not consider that they can be relied on in order to resolve it. The court should be cautious when having regard to the Regulations made under the Act as an aid to construction. As Rix LJ noted in Yarl’s Wood at [63] when rejecting an invitation to have recourse to the 1886 Regulations as an aid to construction:

“the Regulations cannot drive the interpretation of the 1886 Act”

. We accept that the Regulations were made sufficiently close in time to the passing of the 1886 Act not to be disqualified from being considered as part of the same legislative exercise on temporal grounds, but the court should be particularly slow to give any weight to regulations which have not been subject to any form of Parliamentary scrutiny. How can such regulations give any insight into what Parliament intended? The 1886 Regulations were not debated in Parliament or subject to any other form of Parliamentary scrutiny. They were simply published in the London Gazette. The judge at [102] said that it was not a precondition to reliance on regulations as an aid to construction that they had been considered by Parliament. He said that this was “implicit” from Scottish & Newcastle v Raguz

[2008] 1 WLR 2494

. With respect to the judge, we do not see how Raguz assists. The question in that case was the true meaning of the provision in section 17(2) of the Landlord and Tenant (Covenants) Act 1995 that a former tenant is not liable to pay a fixed charge unless the landlord serves a notice within six months beginning with the date when the charge becomes due informing him that “the charge is now due”. Was this a reference to a sum of rent currently payable or to some unquantified sum that would become quantified and become payable at some unknown date in the future? The Lord Chancellor made regulations pursuant to the statute. It was plain from the language of the regulations that the draftsman took the view that the reference to a charge that “is now due” was to a charge which was currently payable, but had not been paid. This was the interpretation that found favour with the House of Lords. Lord Scott said at [28]:

“In reading these notes, and in reading the Regulations themselves, it must be borne in mind that they cannot add to or detract from section 17 correctly construed. They were, however, made in November 1995, more or less at the same time as the Act, and can, in my opinion, form part of the contextual background, against which section 17 should be construed: see Bennion, Statutory Interpretation, 5th ed (2008), p 706.”

[111]It is not clear whether the regulations under consideration in Raguz were scrutinised by Parliament. Certainly no argument is recorded as having been based on the fact that they were not. In any event, Lord Scott chose his words very circumspectly. He said that the regulations could not add to or detract from section 17 properly construed. It is difficult to see how reliance on the 1886 Regulations in the present case would be for any purpose other than adding to section 2(1). Whatever precisely Lord Scott meant by saying that the regulations formed part of the contextual background, it is difficult to see how that background helps to construe section 2(1).

[112]We were told by counsel that they had been unable to find any authority in which the court had decided whether it was appropriate to rely on regulations as an aid to the construction of a statute where the regulations had not been considered by Parliament. We do not find that surprising. If regulations are considered by Parliament at more or less the same time as the enactment of the statute under which they are made, one can see how it may be legitimate to have regard to them in construing the statute, since they are part of the same legislative exercise. In R (A), I said in the Court of Appeal at [47] that the Act and the Rules “were all enacted and came into force at the same time as part of a single legislative scheme” and that that pointed the way to the resolution of the question of statutory construction that arose in that case. On the other hand, where the regulations have not been considered by Parliament, the most that can be said is that they probably indicate what the person who made them believed Parliament intended. However, his belief may have been mistaken. In any event, the case for relying on the 1886 Regulations as an aid to construction assumes that the person who drafted regulations 5 and 6 believed that the 1886 Act did not allow compensation for consequential loss, but the explanation for the absence from these regulations of a reference to sums claimed for consequential losses could simply be that no thought was given to whether they fell within the compensation payable under section 2(1). It is not surprising that the focus of the 1886 Regulations was on the cost of repairs. It is noteworthy that they do not make provision for a claim for the diminution in value of property damaged or destroyed by rioters, and yet it is not disputed by Mr Grodzinski that a building owner who decides not to repair his property, but instead to sell it in its damaged state, is in principle entitled to compensation for the diminution in value under section 2(1). The diminution in value is likely to reflect the purchaser’s estimate of the cost of repair and any loss of rent. This shows the danger of relying on the Regulations to interpret the statute.

[113]We have not overlooked the fact that the second proposition enunciated by Lord Lowry states that regulations provide a “Parliamentary or administrative” (emphasis added) contemporanea expositio. He does not explain why an administrator’s exposition should carry weight. For the reasons we have given, we think it should carry little (if any) weight. In any event, we see no reason to think that Lord Lowry considered that the administrative exposition carries as much weight as that of Parliament.

[114]Finally, the judge relied on R v Secretary of State for the Home Department ex p Mehari

[1994] QB 474

to support the proposition that the 1886 Regulations could be used as an aid to the construction of the 1886 Act, although he acknowledged that the court should “exercise a degree of caution given the absence of any evidence that the Regulations were ever laid before Parliament” [104]. In Mehari, the issue was whether the Statement of Changes in Immigration Rules could be used as an aid to the construction of the Asylum and Immigration Appeals Act 1993. It was argued that they could not, because the time had not yet passed within which Parliament might disapprove them. In rejecting this argument, Laws J said at p 484A:

“[Counsel for the applicant] does not dissent from the proposition that if the time for Parliamentary disapproval had passed, I could have regard to the paragraph: and this is surely right. In my judgment, the fact that that time has not yet passed goes perhaps to the weight I should attach to the paragraph, but cannot constitute an absolute bar against my taking it into account. At present the paragraph has legal effect for the administration of immigration control, and does so in the context of the Act of 1993. There is manifestly no kind of presumption that Parliament will strike it down. Its position is analogous to, though not identical with, that of a statutory instrument which may be prayed in aid to construe main legislation, where it is clear that the two are intended to form an overall code; I do not think that in such a case the court would have to disregard entirely the statutory instrument on the ground only that it still remained open to Parliament to strike it down by negative resolution.”

[115]It is not clear how much weight the judge placed on this decision. Laws J expressed himself with considerable caution. More importantly, he regarded the position as analogous to that of a statutory instrument which is part of an overall code. Moreover, he had in mind a statutory instrument which was susceptible to Parliamentary scrutiny. That is not this case. For that reason, we do not consider that Mehari supports recourse to the 1886 Regulations as an aid to the construction of the 1886 Act.

(f) Consistency of interpretation

[116]Mr Grodzinski also submitted that it is relevant that the 1886 Act has been consistently interpreted over a long period of time as excluding compensation for consequential losses and that the 1886 Regulations, which do not make provision for claims in respect of consequential losses, have not been challenged as being ultra vires. Moreover, Parliament could have amended the 1886 Act in the knowledge of the existence of the 1886 Regulations so as to make it clear that they had wrongly excluded consequential losses from the compensation available under the statute. Parliament did not do so, even when the statute came before it for amendment on various later occasions. Mr Grodzinski relied on the powerful presumption that the settled construction of a statute should not be disturbed. The courts are reluctant to disturb a settled construction and the practice that has been based on that construction: see Bloomsbury International Ltd v Sea Fish Industry Authority

[2011] 1 WLR 1546

per Lord Phillips at [57] – [61]. The position in the present case is particularly stark since the claimants are asking the court to depart from a customary interpretation of the 1886 Act that has lasted for well over 100 years.

[117]The rationale for this reluctance on the part of the courts is by no means certain: see, for example, the discussion by Carnwath LJ in Anglesey County Council v Welsh Ministers

[2010] QB 163

at [40] – [43]. The reason given for the presumption was expressed by Carnwath LJ in these terms:

“Where an Act has been interpreted in a particular way without dissent over a long period, those interested should be able to continue to order their affairs on that basis without the risk of being upset by a novel approach”

[118]As Lord Phillips said at [59] in Bloomsbury, this “has the air of pragmatism rather than principle”. For that reason, and because the strength of the presumption is uncertain, in our view it should be treated with caution. As applied to the present case, the argument is that, since the 1886 Act and the Regulations have been consistently interpreted as excluding the right to compensation for consequential loss, the court should be slow to disturb that settled interpretation. We reject this argument. As Mr Crane pointed out, there is no evidence before the court to support it. There is no evidence that claims have consistently been made and settled on the basis that there is no entitlement in principle to compensation for consequential loss. We do not know whether claims for consequential losses have been made and settled over the years or not. Moreover, there is no evidence that police compensating authorities have consistently organised their affairs on the basis that they are not liable for consequential losses and that to impose such a liability on them now would seriously disrupt the way they have become accustomed to arranging their affairs. A case based on settled practice does not prove itself.

(g) The anomaly point

[119]The judge accepted at [124] that, in principle, loss of profits consequent on physical damage to property may be recoverable under the 1886 Act if they form part of an assessment of the market value of the damaged property. By parity of reasoning, it would follow that loss of rent consequent on physical damage may also be recoverable if it forms part of the market value of the damaged property. Thus, if the owner of a building which has been damaged by rioters decides to sell it in its damaged state, he is in principle entitled to the whole of the diminution in value of the property attributable to the physical damage that it has suffered. This includes the part of the diminution that is attributable to loss of rent (if rent ceases to be payable while remedial works are carried out) or loss of profits (if the property is sold with a business as a going concern) that the purchaser will suffer pending the completion of the remedial works. That is because the whole of the diminution in value is loss that the owner has sustained “by such injury ... or destruction” (section 2(1)). As Mr Crane pointed out, this is also apparent from the judge’s ultimate conclusion on the second preliminary issue which is couched in terms that consequential losses are not recoverable pursuant to section 2(1) “at least as a free-standing head of claim”. In other words, they may be recoverable if they do not form part of a free-standing head of claim. Where the owner decides to sell the building in its unrepaired state, any loss of rent or profits that will be suffered pending repair is capitalised into that part of the diminution in value that is attributable to the loss of rent or profits. Mr Grodzinski supported the position taken by the judge on this point.

[120]However, if the owner decides to repair the building and suffers the loss of rent or profits as a loss of income stream pending completion of the repairs, on the judge’s approach he cannot recover that loss. The reason the judge gave at [124] was that “the 1886 Act does not provide compensation for consequential losses as a free-standing head of claim such as is claimed in the present case”. With respect to him, this is not a satisfactory answer. It is merely an assertion that section 2(1) does not allow compensation for loss of rent or other consequential losses. The anomaly demands an explanation. No rational basis has been suggested for imputing to Parliament an intention to allow recovery of a capitalised sum for loss of rent or loss of profits as part of a claim for diminution of value but not to allow a free-standing claim for losses of the same character. In our view, this anomaly calls into question the judge’s conclusion that consequential losses cannot be the subject of compensation under section 2(1).

(h) The nature of the liability

[121]We heard a good deal of argument on the question whether the liability to pay compensation under the 1886 Act is in the nature of, or akin to, a strict liability in tort (as Mr Crane submitted) or is a sui generis statutory compensation provision which has nothing to do with tort (as Mr Grodzinski submitted it to be). We were referred to passages in the judgments of Rix LJ in Yarl’s Wood and of Longmore LJ in Bedfordshire Police Authority v Constable [2009] 2 All ER (Comm) 200. We do not find it helpful to debate that question. Whether compensation payable under section 2(1) of the 1886 Act includes consequential loss is one of statutory interpretation. It does not further the debate to decide the precise nature of the liability. Board of Managers of St Mary’s Kenmure

[2013] SLT 285

[122]Mr Grodzinski relied on this decision of the Outer House (Temporary Judge M Wise QC) in support of his case that compensation payable under section 2(1) cannot include consequential loss. This seems to be the only case in England or Scotland in which the issue of the recoverability of consequential losses has been decided. The judge held that compensation under section 10 of the Riotous Assemblies (Scotland) Act 1822 (as amended) does not include consequential loss. The judge gave a number of reasons for that at [47]. The first was that the entitlement is to “full compensation for the loss or injury”, not for “any loss or injury”. The use of the definite article was said to mean that “the loss or injury must relate back to the description of damage or injury done to the building”. However, even if this is a convincing reason for excluding consequential loss, it does not apply to section 2(1) of the 1886 Act where the phrase “compensation for the loss or injury” does not appear. The second reason was based at least in part on article IX of the 1714 Act (which applied only to Scotland), which seemed to permit recovery only in respect of physical damage to buildings. The third reason was that to allow recovery of consequential loss would “be bound to lead to unreasonable and objectionable results” which cannot have been intended by Parliament. The judge said that the absence of a proviso such as that contained in the English legislation “highlights the inability of the court to restrict damages in claims under the Scottish legislation to that which is ‘just’ or which takes account of conduct or failure on the part of the claimant to mitigate loss”. This merely serves to show that the Scottish legislation is sufficiently different from the English legislation to make it unhelpful to rely on the reasoning in this decision in order to decide the true meaning of section 2(1) of the 1886 Act.

(i) Conclusion

[123]For all the reasons we have given, we conclude that section 2(1) provides a right to compensation for all heads of loss proximately caused by physical damage to property for which the trespasser is liable at common law, save to the extent that they are excluded or varied by the statute. It does not matter whether consequential losses such as loss of profits were recoverable from a trespasser in tort in 1886. In principle, section 2(1) covers all heads of loss compensable under English law for damage to property caused by trespassers in the course of a riot and the heads of compensation recoverable are to be determined with reference to the English law of damages as it develops over time. We reach this conclusion without recourse to the “always speaking” principle: see Bennion on Statutory Interpretation (6th ed) at section 288.

[124]In any event, the judge was right to say at [124] that the better view, based on the judgment of Sir Francis Jeune P in The Kate

[1899] P 165

at pages 168-169, is that, by the time of the 1886 Act, loss of profits from charterparties terminated by reason of the total loss of a vessel in a collision were recoverable as damages in tort. As the President said, the general principle which governs the assessment of damages is restitutio in integrum “qualified by the condition that the damage sought to be recompensed must not be too remote”.

[125]Mr Grodzinski submitted that, if the hundred or the compensating police authority is to be strictly liable for certain consequences of riotous and tumultuous behaviour, that can only be because it is fair and appropriate for them to be notionally and collectively responsible for failing to prevent those consequences. It is one thing to say that the community must be taken to have knowledge of those facts for which they are held responsible (for example, the property about to be attacked, and as to the means of prevention), but quite another to suggest that the community should also be fixed with the knowledge of a private citizen’s economic affairs, so that it becomes responsible not only for the damage that it can see, but also for the consequential losses which it cannot see.

[126]We do not consider that this generalised appeal to fairness is particularly illuminating. It may be said that it is surprising that the community should be under a strict liability to pay compensation for the consequences of riotous and tumultuous behaviour at all, when the police are not liable in tort for such consequences even where they have been seriously at fault: see Hill v Chief Constable of West Yorkshire

[1989] AC 53

. However, this has been the law since 1714 and only Parliament can change it. The principle that the community stands in the shoes of the trespasser is long-established. So too is the liberal approach of the courts to the interpretation of the legislation. Even if this is considered to be unfair and unwarranted in the 21st century, it is for Parliament and not the courts to amend it or remove it altogether. Parliament has already mitigated the full rigour of the principle in important respects. First, it has recognised that regard should be had to the conduct of the person whose property has been damaged, stolen or destroyed. Secondly, it has provided that the compensating authority shall fix such compensation as appears to them to be “just”. Rix LJ said in Yarl’s Wood at [63] that this entitled the authority to “take a broad view of what the claimant is entitled to”. This is not the place to decide what the limits of this broad view may be, but it is further Parliamentary recognition of the need to strike a fair balance between the interests of the property owner and the community. In striking that balance, Parliament could have excluded consequential loss altogether, but it has not done so. Overall conclusion

[127]For the reasons we have given we dismiss MOPC’s appeal on the issue of liability and allow the claimants’ cross-appeals on the issue of damages.

Cited in 1 later judgment

The Mayor’s Office for Policing and Crime v Mitsui Sumitomo Insurance Co (Europe) Ltd and others 2016