The Worshipful Company of Grocers v Keltbray Group Holdings Ltd & Anor

Introduction

The only issue

At about 4:30 pm on Wednesday, 10 August 2005, a collapse occurred at the site of the former London Stock Exchange building at 125 Old Broad Street and 60 Threadneedle Street EC2, which was being demolished by the First Defendant (“Keltbray”). A large amount of demolition rubble had been stacked on an unsupported concrete floor slab at ground floor level. That slab gave way, and about 300 tonnes of rubble and slab fell about 8.4 m down to the lowest basement level.

The Hall of the Worshipful Company of Grocers (“the Grocers”) in Princes Street EC2 is about 235-240m away from the site of that collapse. Nine days after the collapse occurred, at about 5pm on Friday, 19 August 2005, Mark Hogwood, the Hall Technician, checked Grocers’ Hall and locked it up for the weekend. All seemed in order. However, when Mr Hogwood came to re-open Grocers’ Hall at about 8am on Monday, 22 August 2005, he found water pouring through the ceiling into the reception area. At some point over the weekend, a cistern in a third-floor lavatory had fractured, so that water was able to run freely out of the base of the cistern. Much damage had been caused to the Grocers’ Hall as a result.

The Grocers (who are the Claimants in this action) allege that the cause of the fracture of the cistern and the consequent escape of water over the weekend between 19 and 22 August was the ground-borne vibrations caused by the collapse at the London Stock Exchange building on 10 August 2005. Keltbray admits that the collapse was the consequence of its negligence in overloading the floor slab. It denies, however, that the events at Grocers’ Hall were anything to do with that collapse. The amount of the Grocers’ loss is agreed at £250,000 (exclusive of interest), and it is common ground that that loss (if caused by the collapse) is not too remote in law to be recoverable. The sole remaining issue in this action is therefore causation: did the collapse at the London Stock Exchange cause the cistern at Grocers’ Hall to fracture?

The claim against the Second Defendant was discontinued (together with the Second Defendant’s claim over against Keltbray) on 13 October 2011.

The hearing

At the hearing, the Grocers were represented by Mr Lawrence Jones and Mr Justin Meiland. Keltbray was represented by Mr Doré Green. The evidence in this case was heard over five court days, between 8 and 16 December 2015. Written closing submissions were lodged (in accordance with a direction which I gave at the request of the parties) on 15 January 2016. However, because of difficulties with the availability of counsel, oral closing submissions did not take place until 10 March 2016. In those circumstances I have been greatly assisted by the provision by the parties of a full transcript of the evidence, to supplement my own note.

The evidence

Factual evidence

The Grocers called evidence from 5 witnesses. Brigadier Peregrine Rawlins (who was Clerk to the Grocers’ Company in August 2005) and Rear-Admiral Alastair Ross (who was, and still is, the Clerk of the Drapers’ Company) were both called to confirm the evidence in their witness statements concerning their experience of the collapse. No questions were asked of either of them in cross-examination. The evidence of Robert Wood (who was, and still is, Head of Compliance and Risk at the London branch of Türkiye is Bankasi AS) about his experience of the collapse was also not seriously challenged.

The evidence of Mark Hogwood and of Michael Griffin (the Finance Director and Deputy Clerk to the Grocers) about what they saw when they arrived at Grocers’ Hall on Monday, 22 August 2005 and about other topics including, in particular, the timing, location and extent of cracking at Grocers’ Hall was, however, the subject of challenge. I shall have to rule later in this Judgment on some of the points of dispute.

It was my impression that both Mr Hogwood and Mr Griffin were genuinely seeking to assist the Court to the best of their present recollection and ability. However, the events with which this trial is principally concerned occurred more than 10 years ago. As is inevitable, memories have faded or been changed by the passage of years (Footnote: ¹). The evidence of both Mr Hogwood and Mr Griffin (particularly that of Mr Griffin (Footnote: ²)) was in some respects inconsistent with more contemporary records of events. Some of those more contemporary records are themselves double hearsay - reports by a person not called as a witness of what others (sometimes not named) had said to that person. Some, however, are direct records by contractors or experienced surveyors employed for the purpose by Grocers (or their insurers) of what they saw at the time. It is therefore inevitable that I must test the evidence of Mr Hogwood and Mr Griffin, on those points where its reliability is challenged, by reference to the objective facts, the contemporary (or, at least, more contemporary) documents, and the overall probabilities (Footnote: ³).

Keltbray called no factual evidence.

Expert evidence

The expert witness for the Grocers was Mr Robert Jessep, of William J Marshall & Partners. The expert witness for Keltbray was Mr Stuart Allen, of Richard Jackson Limited. Both of these experts are highly qualified and highly experienced civil and structural engineers. As one would expect, on many points they were in agreement, and they produced a very helpful Joint Statement dated 27 January 2012 and Supplementary Joint Statement dated 16 July 2015. However, on the central issue, they had come to diametrically opposite conclusions. I shall have to resolve that conflict later in this Judgment, since their opposing views cannot both be right.

It was nevertheless my impression that Mr Jessep and Mr Allen were both (like Mr Hogwood and Mr Griffin) genuinely seeking to assist the Court to the best of their abilities. Both gave clear, reasoned explanations for their views, both in their Reports and in answer to questions in cross-examination, and both founded their explanations upon statements in British Standards and other technical literature which it was agreed were (to a greater or lesser extent) authoritative.

In his opening submissions for the Grocers, Mr Jones drew my attention to certain observations which Edwards-Stuart J had made about the evidence given by Mr Allen in the case of Rendelsham Estates plc v Barr Ltd (Footnote: ⁴). In that case, Edwards-Stuart J had found (inter alia) that Mr Allen, in giving his evidence, had “lost the degree of objectivity that the court expects of an expert witness”, and consequently was “not .. a witness upon whom the court in general could rely”. I did not detect any similar loss of objectivity in Mr Allen’s evidence in this case.

Mr Allen did have a tendency to answer unexpected questions with an instinctive “gut reaction” response, and only thereafter to think through the technical and/or scientific steps necessary to justify his answer. This contrasted with Mr Jessep’s approach to answering questions, which was to think the answers through before he began to give them. However, in my judgment, both approaches are consistent with the role of an expert. It is the common experience of this Court that an experienced expert may sometimes instinctively know the right answer, before he or she has fully thought through the scientific reason why that answer is correct.

It is also right that Mr Allen, unlike Mr Jessep, had not commissioned a computer model of the effect of the collapse, and did not really engage with the detail of the computer model used by Mr Jessep, preferring instead to rely upon logical argument from physical observations and upon more straightforward calculations. However, the accuracy of the results of any exercise in computerised modelling depends upon the accuracy, not only of the model, but also of the inputs to the model: and it was agreed between the experts that the choice of those inputs, in the present case, involved expert judgment rather than measurement. It is that judgment which, in due course, I shall need to evaluate.

This is one of those cases in which it is not possible to assess the correctness of the central aspects of the factual and expert evidence without looking at the picture presented by the evidence as a whole. In the following sections of this Judgment, I therefore propose to present a summary of all of the various aspects of the evidence, both factual and expert, without making any more specific findings than are necessary for the purposes of giving a coherent account. At the end of that summary, I will then deal briefly with the law, before setting out my analysis, findings and conclusion in the final section.

Grocers’ Hall

Most of the factual matters with which I deal in this section were either common ground, or were not seriously in dispute by the end of the trial.

The Grocers are one of the “Great Twelve” Livery Companies of the City of London. The Grocers' first hall, in Old Jewry, was destroyed some 350 years ago, by the Great Fire of London of 1666. Its replacement on the same site was itself replaced in 1802. In 1893, the Company’s fourth hall was completed on the present site, which is off of a private courtyard to the west of Princes Street. The fourth hall survived the Blitz with only minor damage to its north wing, but was almost completely destroyed by a fire in 1965. The present, fifth, hall was constructed in 1970. Parts of the external walls of the fourth hall were retained and incorporated as a façade.

Grocers’ Hall is approximately 235-240m due west of the site of the collapse. Travelling west from the site of the collapse one comes first to a narrow road, Bartholomew Lane. One then comes to the Bank of England building, which lies directly between the point of collapse and the Grocers’ Hall. The Bank of England building is a very substantial structure. Its basement, at its shallowest depth, is approximately 3.8 m below the basement level of the Old Stock Exchange at the point of collapse, and approximately 6.9 m below the basement of Grocers’ Hall. The basement walls of the Bank of England building are constructed from reinforced concrete, 8 feet (2.44 m) thick over a 1 m deep raft foundation (Footnote: ⁵). A concrete service tunnel runs in a ring underneath the Bank of England basement (Footnote: ⁶). To the west of the Bank of England is Princes Street, below which the north and southbound tunnels of the Northern Line run, at a depth significantly below that of the Bank of England’s basement.

Grocers’ Hall is a relatively narrow five-story building, comprising a single level of basement and four floors above ground. At the northern end of the building, the Livery Hall at First Floor level is essentially double height. The Third and Fourth floors are situated in a mansard roof. There are further plant rooms on the roof above the main staircase and lift in the south-west of the Fourth Floor level.

The building is founded on shallow pad and strip footings at the southern end and on deep piled foundations at the northern end (except for the retained façade, which is founded on corbelled strip footings) (Footnote: ⁷). In substance, it is a concrete-encased steel-framed building with cast in-situ reinforced concrete floors (Footnote: ⁸).

The design of the upper floors of the building had some unusual features: In his oral evidence (Footnote: ⁹), Mr Jessep described the design of the third floor at Grocers’ Hall as:

.. very unusual, in that the floor slab is cast between the bottoms of the beams [that span across the width of the building] .. and .. there's nearly a metre difference in height [between that floor slab and the actual wooden floor that people walk on] ..

Moreover the building, as built, did not always conform to the specification which the designer had laid down. In particular:

The result in summary, as stated by the experts in their Joint Statement, is that Grocers’ Hall was in 2005:

.. a poorly detailed steel framed building, less robust than a properly detailed steel framed building, [and so] would be prone to movement and distortion and consequent damage

In about 2002 or 2003, the Grocers moved their offices from the Ground Floor to the Fourth Floor. That was where (in Mr Griffin’s words (Footnote: ¹⁵)) the “operations centre” of the Grocers Company was situated in 2005, and where (in the south-east corner) Brigadier Rawlins and his secretary had their offices (Footnote: ¹⁶).

The Third Floor consists, in the main, of a number of bedrooms with ensuite bathrooms. Those bedrooms are used relatively infrequently, on perhaps 15 to 20 nights in any year. The last night on which the bedrooms were used prior to the events with which this case is principally concerned was 14 July 2005 (Footnote: ¹⁷). Also on the Third Floor is the valet’s room, where she does the internal washing and ironing (Footnote: ¹⁸), and the Master’s Flat.

In the south-east corner of the Third Floor, immediately under Brigadier Rawlins’ office, there is a Committee Room and adjacent Sitting Room (which doubles as an occasional additional bedroom). According to Mr Griffin (Footnote: ¹⁹):

The committee room is where we hold our meetings and where clients can also use that particular accommodation. The sitting room is also known as the “Haven”. It is a room where guests who are coming into the committee to help us as experts will sit there quietly until they are called in .. It is occasionally used as a further bedroom

The cistern from which the water escaped was situated in the bathroom adjacent to that Sitting Room. As Mr Hogwood accepted, that bathroom (and the lavatory to which the cistern was attached) would have been likely to be used by anyone using the Committee Room and/or the Sitting Room (Footnote: ²⁰). In my judgment, its location means that it would probably also have been used by anyone else coming to or using the Third Floor who did not have their own allocated bedroom and bathroom there.

The month of August is in the middle of the Grocers’ summer recess, when no events are scheduled at Grocers’ Hall, but redecoration and refurbishment is carried out. As Mr Griffin accepted, at the relevant time there would have been a number of regular contractors either coming in and doing work on site, or coming in to collect items to work on elsewhere. August is a relatively busy period in terms of works (Footnote: ²¹).

The collapse

I now turn to set out the evidence which was presented to me about the collapse itself. Again, most of the matters which I deal with in this section were either common ground, or were not seriously in dispute by the end of the hearing.

The collapse took place at about 4:30 pm on Wednesday, 10 August 2005. In their Joint Statement, the experts described what took place as follows:

During demolition of the former Stock Exchange building, which was located at 125 Old Broad Street/60 Threadneedle Street at about 4:30 PM on 10 August 2005 there was a collapse with about 300 Tonnes of demolition rubble and floor slabs falling from the ground floor level at about +15.6m Ordinance Datum (OD) to the lowest basement level at around +7.2m OD, that is, about 8.4m.

It was also common ground that what in fact happened was that the amount of demolition rubble which had been stacked on an unsupported cantilever floor slab exceeded the structural strength of that slab, which snapped. That floor slab and the rubble on it fell onto the floor slab below, which also snapped. The rubble and the two slabs then fell onto the 1.4m thick concrete basement floor slab of the building. This was supported upon piled foundations, and so transmitted the energy of the collapse into the ground (Footnote: ²²).

In the course of his cross-examination, Mr Allen gave a slightly more detailed description of the collapse (which was not challenged in cross-examination, and which I accept as broadly accurate) (Footnote: ²³):

.. Mr Jessep and I discussed this and decided that as it’s a shear failure of a cantilever slab, it fails pretty quickly .. It snaps fairly quickly. It would then have dropped the 3½ m or thereabouts (Footnote: ²⁴)to the intermediate floor. Given the fact that the one above had snapped under the load, and this was now a travelling mass, it would have snapped the next one pretty instantaneously and carried on straight down to the bottom.

The intermediate break on the whole thing, I suppose, has a value, but we don’t think it was very much, and actually if nothing else it probably would compress the rubble slightly, so the rubble would probably be a bit more compressed as it hit the ground. So effectively it hits as one lump ..

The experts’ Joint Statement records their agreement that:

The energy imparted by the collapse at the former Stock Exchange would be between about 21 million Joules and 25 million Joules, essentially as a single shock

The Joule is a derived unit of energy. In every day terms, it can be defined as the work required to produce 1 watt of power for 1 second. A million Joules is sometimes referred to as a “Megajoule”. One Megajoule is approximately the kinetic energy of a 1 tonne vehicle moving at 160 km/h. A kilowatt hour is equivalent to 3.6 Megajoules.

The experts’ agreement that that considerable quantity of energy would have been imparted “essentially as a single shock” is, unfortunately, not the end of the matter. There is still the issue of the “loading impulse” or “loading duration” – the time taken for that energy to be transmitted into the ground. As Mr Allen went on to observe (again in a passage which was not challenged, and which I accept) (Footnote: ²⁵):

[A] .. The total load arrives on top of the slab at the same time .. Effectively it would all arrive at once.

The question then is: how stiff the ground is to stop that load sinking into the ground and recovering. Just sinking into the ground is the only figure we are interested in, because that is imparting the force, and .. how long that is depends on how long the piles are, how stiff the raft is, what the centre of the piles are in the location where the force hit - and that’s the information I don’t have.

[Q] So what we are talking about is not actually the time it takes for the mass to hit; it’s the time it takes for the force thereafter to be transmitted to the ground?

[A] It is exactly that ..

The energy thus transmitted into the ground would have spread out from the site of the collapse in waves. The extent of travel in any given direction will have depended not merely on the loading impulse, but also upon what lay between the site of the collapse and the point with which one is concerned - in particular, upon the stiffness, density and other properties of the ground at all points from (and including) the collapse site itself. It was common ground between the experts that the soil strata in that part of the City of London consist of a thin band of made ground, over a stratum of terrace deposit gravel, over various strata of London clay. The experts did not, however, agree as to the degree of “attenuation” that would have occurred as the energy from the collapse travelled through that ground.

The evidence about the extent to which the collapse was noticed

I next summarise the evidence about the extent to which the effects of the collapse were perceived, first at other locations and secondly at Grocers’ Hall itself.

At other locations

The collapse took place only a little more than a month after the events of 7 July 2005, when a series of terrorist suicide bomb attacks had targeted civilians using the London public transport system during the morning rush hour. Everyone in London was therefore at a heightened level of alert for further explosions.

The site of the collapse was in a densely built-up area of the City of London. As Mr Allen pointed out, there are a number of relatively fragile buildings within the same radius of the collapse site as Grocers’ Hall, including a number of churches and other listed buildings (Footnote: ²⁶). Mr Ballard’s report dated 1 June 2007 recorded that he had:

.. inspected the exterior of a number of buildings between the Hall and site of the incident, and [could] find no comparable evidence of movement: in particular, the buildings immediately surrounding the old Stock Exchange show no evidence of comparable or more severe damage or distress

The nearest structures to the site of the collapse which were referred to in the evidence before me were the tunnels of the Central Line. These run at a distance of about 15 m from the collapse site, and at a depth of about 11.5 m below the basement slab (Footnote: ²⁷). In an email dated 10 January 2012 sent to Mr Allen, London Underground Ltd confirmed that they had no record of any damage to London Underground assets and infrastructure in and around the Bank area in August 2005.

At a distance of about 25 m from the site of the collapse, on the opposite side of Old Broad Street, is a five-storey stone-clad office building, with shops at ground floor level. There was no evidence of any report of damage to that building (Footnote: ²⁸). Indeed, except to the extent noted below, there was no evidence before me of any report of any damage caused by the collapse to any other building (Footnote: ²⁹).

There was evidence that the effects of the collapse were noticed at Drapers’ Hall, which is on Throgmorton Avenue, approximately 76-85 m to the north of the collapse site (Footnote: ³⁰). According to Rear-Admiral Ross (whose office was on the ground floor of Drapers’ Hall):

I heard an enormous bang. My immediate instinctive reaction was that a bomb had gone off.

.. The entire building shook briefly, the windows rattled and the brass candelabra suspended from the ceiling of my office moved side to side. It seemed to me to last for about 2 to 3 seconds. I particularly recall the very loud bang, the vibration throughout the building, the rattling of the windows and the movement of the candelabra.

I went to the reception area. Other people were already gathering there from other parts of the building, alerted by the noise and movement of the building ..

Even so, Rear-Admiral Ross did not suggest in his evidence that any actual physical damage had been caused to Drapers’ Hall.

In an email dated 18 October 2005 sent to the Managers of The Livery Companies Mutual Limited insurance company, the Director of Finance of the Merchant Taylors Company reported that the effects of collapse had also been noticed at Merchant Taylors Hall. Merchant Taylors’ Hall is approximately the same distance away from the site of the collapse as Drapers, Hall, but to the east. According to that email:

.. We did have a problem on this date and are building did appear to move. I have spoken to The Clerk about this and we were planning to get a Structural Survey carried out to see if there was any damage ..

There was, however, no evidence before me that any such survey had in fact been carried out. Nor was there any evidence that any damage had actually been caused to Merchant Taylors’ Hall.

The evidence before me did include a report of damage to the Royal Exchange building, a masonry building which is located approximately 75 - 125 m to the south of the site of the collapse. As recorded in the experts’ Supplementary Joint Statement (Footnote: ³¹), DTZ Debenham Tie Lung complained in a letter dated 14 December 2005 to Delva Patman Associates (Footnote: ³²) that the collapse had caused 15 cracked glass panels in the circular windows at lower level facing onto the Atrium at the fourth floor, along with some minor cracking to the plasterwork in various locations. However, Delva Patman Associates’ reply dated 21 December 2015 (Footnote: ³³) rejected that complaint, saying that they “find it extremely unlikely that the collapse of a ground floor area some 25-30 m away would cause cracking to windows located on the fourth floor of a building without causing considerable damage elsewhere”. There was no evidence before me that the complaint relating to the Royal Exchange was thereafter pursued.

The buildings to which I have so far referred are all considerably closer to the site of the collapse than Grocers’ Hall, and none of them had the benefit of any shielding effect that the Bank of England building may have provided.

However, there was also evidence that the effects of the collapse were noticed at the time in the offices of Türkiye is Bankasi AS at Princes Court in Princes Street, which is directly in front of the Grocers’ Hall. According to Mr Wood, who was working on the first floor at the time, he perceived:

.. A loud explosion -like bang and tremor, which shook the office .. My initial thought was that a bomb had been detonated within the vicinity of the Bank .. As I sat there, there was a very obvious tremor that shook the building .. I remember speaking to colleagues who also believed that a bomb had exploded nearby and confirmed that they felt the tremor ..

In oral evidence, Mr Wood said that the experience of shaking went on for “a couple of seconds”, though he acknowledged that, after all this time, it was difficult for him to remember precisely. Again, however, Mr Wood did not suggest that any actual damage had been suffered by his building.

At Grocer’s Hall

Brigadier Rawlins’ evidence of how the effects of the collapse were felt from his office on the Fourth Floor of Grocers’ Hall was as follows:

.. I recall quite clearly sitting at my desk in the Clerk’s office and feeling a sudden sharp jolt, accompanied by a dull thud. A colleague had brought her dog with her to the Hall that day, and my immediate thought was that the dog had entered my office unseen and bumped against the back of my chair. When I saw that this was not the case, I concluded that there must have been a mild earthquake. I then went next door to ask my PA whether she had felt anything, and she confirmed that she too had felt a jolt ..

Some “second-hand” evidence about how the effects of the collapse were perceived at Grocers’ Hall can also be found in the reports which were written not long after the event by the consultants who were retained to investigate the subsequent problems. In particular, in his report dated 14 December 2005, Mr Phillips of Burgoynes noted that:

.. [Brigadier Rawlins]was in his office at the southern end of the top (fourth) floor when he became aware of a sudden, single jolt. He described it to me as being a “single jolt” and not a continuous tremor. It resulted in his computer monitor swaying. His secretary, who was sitting in the office to the north, was also aware of the incident, although people seated further north in the floor were not.

The June 2007 report by Mr Ballard of GBG contains a summary to similar effect:

Anecdotally, the event was noted by staff at the Grocers’ Hall as a perceptible jolt, and in discussion with other persons in the area either a sound or a slight jolt was felt by several people at the same time. Others did not notice it ..

.. There are no reports however from the staff that the “jolt” gave rise to any sound of building distress, which would certainly have been the case if the whole building was shaken to the extent that it cracked ..

In oral evidence (Footnote: ³⁴), Mr Griffin confirmed that, when he returned to work on Monday, 15 August 2006, no one mentioned anything to him about having heard an explosion or about having experienced any jolt or tremor on the previous Wednesday, 10 August 2006.

In his oral evidence, Mr Griffin also expanded on a passing reference in his Witness Statement concerning the effect of the “jolt” on glasses in a display cabinet at Grocers’ Hall. According to Mr Griffin, it was brought to his attention “much later than the actual flood itself”, in about September 2005 (Footnote: ³⁵), that these glasses had moved. Mr Griffin’s recollection was this (Footnote: ³⁶):

[A] .. There are glass shelves and they are part of the sort of major collection that we had within the Hall they are positioned in a certain way so that the cabinets are dressed in a particular style, and the style is that they would be setback against the back of the cabinet and in the middle of the cabinet, not on the edge of the cabinet when the door is locked. So the danger was if the door had been opened, the glasses might have toppled over.

[Q] So they have moved forward a little?

[A] Correct.

Again, there is some roughly contemporary (albeit “second-hand”) evidence on this aspect of the matter in Mr Phillips’ 14 December 2005 report:

.. There are a number of display cases on the second floor with cherished glassware. Following the incident these glasses were found to have moved along their shelves so that glasses were in contact where previously they had been apart. It is assumed that the movement occurred during the “jolt” ..

The damage to the cistern

As I stated at the outset of this judgment, the flood was discovered by Mr Hogwood when he came to re-open Grocers’ Hall on Monday, 22 August 2005, some 12 days after the collapse.

The cistern from which the initial escape of water occurred was in the bathroom adjacent to the Committee Room and Sitting Room in the south-east corner of the third floor of Grocers’ Hall. It was a low-level cistern, close coupled to the lavatory pan beneath, and fixed to the tiled breezeblock wall behind by screws. These screws passed through oversize holes in the back of the system into rawlplugs in the wall. There were washers between the screw heads and the inside face of the cistern, but no washers or other buffers between the rear face of the cistern and the wall to which the cistern was attached.

Mr Allen described the arrangement as follows (Footnote: ³⁷):

.. In a close coupled cistern you have the toilet pan. At the back, it has a sort of flat area with a nice moulding. Around the main water entry point there is a four-inch rubber ring put in ..

.. You then put the cistern on top and there are two holes at the bottom which you have rubber washers around. Straight through there, into two holes in the toilet pan, underneath which you have two wing nuts. You pull the cistern down onto the toilet pan - making it seal - so all of the weight of the cistern is supported by the toilet pan itself. The two holes at the back of the wall and the two fixings of the toilet pan to the floor are literally just a little bit of stability, because the whole thing is fairly stable by itself. In fact we can see that because the current one has a big gap behind it.

No evidence was available to show how tightly or how rigidly the cistern was fixed to the wall. By the time that any thought was given to the possibility of a claim against anyone, the damaged cistern had been removed and replaced (this time with rubber washers as buffers between the back of the cistern and the wall), and the original cistern and its pieces had been thrown away.

Mr Hogwood’s description of what he observed at the time was recorded by Mr Phillips in his 14 December 2005 report, as follows:

Mr Hogwood described the cistern as having fractured at its left end, with a crack running down and left from the hole for a securing screw towards the top left corner of the rear face of the cistern. .. So far as I am aware, the cistern was not photographed. It was removed and replaced. The remains were discarded.

Mr Hogwood’s description in his witness statement of what he found when he went into that bathroom adds some details to that contemporary description:

.. The bathroom floor was waterlogged. On the floor, part of the base of the cistern lay broken in a few pieces. Water was running freely out of the base of the cistern. I immediately removed the cistern lid, picked up the toilet brush and jammed it under the ballcock to stop the flow of water ..

He expanded on this a little in his oral evidence (Footnote: ³⁸):

[The debris from the broken cistern] appears to have fallen straight to the floor .. as we are looking at it, to the left of the toilet pan itself ..

The description in Mr Griffin’s witness statement is consistent with that given by Mr Hogwood:

In the bathroom on the third floor, I saw how Mr Hogwood had jammed a toilet brush under the stopcock to prevent further water being drawn. I also saw the cistern, which was cracked down its back and at the base. It was clear that the water had been leaking through the bottom of the cistern which had broken and partly fallen away.

On 14 July 2014, Master McLeod refused the Grocers’ somewhat belated application to adduce evidence concerning the possible causes of the failure of the cistern from an expert in ceramics.

The cracking at Grocers’ Hall

During the days which followed the flood, the Grocers did all that they could to get their Hall ready to re-open in September. This involved what Mr Griffin referred to as “major disruption and disturbance” (Footnote: ³⁹). As Mr Griffin described in his witness statement:

.. The clean up operation continued, with cleaning staff mopping and vacuuming carpets to try to extract the water. Dehumidifiers were hired, and the carpet in the Court Room, which had by then shrunk by 7 or 8 inches and no longer reached the walls, was replaced. Its removal was a huge task that required the very large table in the Court Room to be dismantled and moved. The damaged cistern was quickly replaced ..

A skip was hired and the carpet from the Courtroom placed in it. I presume that the damaged cistern was also disposed of in this way. To make matters worse, water continued to cascade sporadically into the ground floor office .. through the ventilation grills in the ceiling .. This led to the discovery .. of [a] cracked waste pipe in the ceiling above the reception and ground floor office.

On or around 24 August 2005 the cast iron pipe was replaced with a new plastic one. This was itself a huge task with the pipe measuring about 30 foot. It resulted in further delays to the repair and readying of the ground floor ahead of re-opening.

By early to mid-September, the hall was mainly dry but the dehumidifiers remained in place ..

By mid-September, the Hall reopened for business ..

In about late September 2005 (Footnote: ⁴⁰), cracks were noticed in the walls on the second, third and fourth floors of the building (Footnote: ⁴¹). As summarised in the Experts’ Joint Statement, these cracks comprised:

There was no record in the evidence before me of any cracking having been observed in the Grocers’ Hall before 22 August 2005. That, of course, does not necessarily mean that cracking did not exist prior to that date.

It was at or about the time when this cracking was first observed that it was suggested that the collapse may have been responsible both for the cracks and for the flooding. Following the involvement of insurers and solicitors, reports were prepared by Mr Gilham of Watts and Partners (11 November 2005), Mr Phillips of Burgoynes (14 December 2005), Mr Ballard of GBG (June 2007) and WhitbyBird/Ramboll (19 July and 13 November 2007).

There is evidence, both from the contemporary documents and from those reports, that cracking continued to occur at Grocers’ Hall in 2006. For example, an email dated 3 February 2006 sent between insurance intermediaries referred to a concern of Royal and Alliance about “possible continued movement”, and stated that “it is clear the cracks are continuing to appear”. A little later, in a letter dated 12 April 2006 to Mr Griffin, Mr Gilham drew attention to ongoing high moisture meter readings, and recommended the application of monitoring “tell-tales” on the grounds that “it would appear the Hall is experiencing on-going movement”.

According to Mr Griffin, problems continue to surface in 2007 (Footnote: ⁴²). In his June 2007 report Mr Ballard (the seismologist appointed by Royal and Alliance) noted that he had located a number of cracks which had not been observed either by Mr Gilham or by the Grocers’ Hall staff. In Mr Ballard’s view “it is equally likely that these were either missed previously or are the result of an ongoing mechanism and have appeared subsequent to [Mr Gilham’s] inspection”. He also observed that

.. There is a general trend for the cracks to be wider, more frequent and more disfiguring towards the Northern end of the building, and at higher levels. At roof level there are some open and extensive cracks in the plant rooms, which penetrate the block/brickwork and are indicative of movement and “racking” of the whole structure ..

.. In the plant room, there is clear evidence of the relative movement between the finishes and the underlying steel frame .. It appears that the frame has been slightly distorted, [and] shows evidence of possible sway or asymmetric displacement ..

WhitbyBird’s July 2007 report was a full structural survey. They reported that, of the cracks monitored by the “tell-tales” installed in April 2006, movement was only perceptible in the 3 of the monitored cracks which were adjacent to bedrooms 8 and 9 on the Third Floor, and in these cases only after the full year. Their conclusion was that:

.. The main area of cracking that is of concern is at third floor level of the North wing, and to a lesser extent the same area on the fourth floor above. This is where the cracks are most severe and “tell-tales” indicate the cracks are opening slightly. There is also visible evidence that the cracks are opening on the third floor as they are elongating ..

Their recommendation was that “if movement of the internal walls to rooms eight and nine on the third floor continues to worsen, .. the slab in rooms eight and nine [should be] exposed and inspected, with particular attention being paid to the support conditions of the slab on the beams”.

That recommendation was followed, and the raised timber floors to rooms 8 and 9 were partially removed in August 2007, to expose the slab below. This led to the discovery of the holes that had been made in the stub walls. It also led to the discovery of cracks in the slab, indicating that some deflection had taken place. WhitbyBird’s recommendation was that the cracks should be monitored, and that further investigation should be carried out when refurbishment of the third floor is undertaken.

WhitbyBird also recommended that vibration monitoring should be started as soon as possible, as works of demolition and construction were beginning at No 1 Lothbury, a site immediately to the north of Grocers’ Hall. On 11 October 2007, Mr Griffin sent an email to WhitbyBird, asking them when the monitoring equipment will be in place, because:

.. We have suspicions that there is further movement with doors that were opening and closing freely suddenly sticking or not closing. More cracks are appearing in bedroom 12 [on the 3^rd floor] and possibly in Shower Room A on the 4^th floor ..

In his oral evidence, Mr Griffin explained that the sticking doors to which he referred were on the Third Floor (Footnote: ⁴³).

Vibration monitoring was duly carried out by Hann Tucker Associates, who submitted reports in December 2007, and February and April 2008. Unfortunately (as WhitbyBird recorded in their 13 November 2007 Report) “the monitoring equipment was not in place while the worst shaking was felt at the beginning of October”. It appears that these works of demolition and construction did not, in the event, cause any significant damage to Grocers’ Hall (Footnote: ⁴⁴).

In 2009, Sandberg LLP Consulting Engineers carried out both intrusive and non-intrusive investigations at 10 specified areas of floor slabs and heads of walls on the Third and Fourth Floor levels of Grocers’ Hall. They subsequently carried out additional investigations to confirm the depth of reinforcement in the Third Floor level concrete floor slab construction in bedrooms 7 and 9. These investigations revealed the full extent of the construction defects in the floor slabs that I have previously noted. Thereafter, recommendations were made for remedial works, and steel beam restraints were installed.

Mr Allen’s report (Footnote: ⁴⁵) records that, in the course of his site visit on 9 January 2012, he noticed that the cracking between doors 8 and 9 on the third floor of Grocers’ Hall had recurred, and that there was a significant deflection to the door heads on either side of the crack. He also noticed that other crack repairs had similarly reopened. That evidence (which I accept) was not challenged.

The background to the experts’ reports

In order to make sense of the views of the expert witnesses, Mr Jessep and Mr Allen, it is first necessary for me to explain some of the concepts and expressions employed by them, and to set out some of the contents of the authoritative literature upon which they have relied.

Peak particle velocity

I begin with the concept of Peak Particle Velocity (“PPV”). When energy is transmitted to the ground, by blasting or piling, or (as in the present case) by a collapse, that energy is transmitted (or “propagated”) outwards through the ground in waves of three types: compressive waves, shear waves, and Rayleigh waves (Footnote: ⁴⁶). The particles in the soil experience a combination of effects from these different energy waves, and this produces a particle motion which rapidly reaches a peak value before reducing as the waves pass. Attenuation of the generated wave motion, and therefore of the peak particle motion, takes place through the geometric enlargement of the various wave fronts as they move away from the source, and by material damping (Footnote: ⁴⁷).

“Particle velocity” is simply the velocity, at any given point at any given time, of the soil particles that have been set in motion: and “Peak Particle Velocity” is the maximum instantaneous particle velocity, at any given point at any given time. Soil particles can, of course, move in three dimensions. “Peak Component Particle Velocity” (Footnote: ⁴⁸) is the maximum value of any of these three velocities measured during a given time interval (Footnote: ⁴⁹).

It was common ground between the experts that measurement of particle velocity provides the best single parameter for assessing human and structural responses to ground-borne vibrations (Footnote: ⁵⁰).

The calculation of PPV

Unfortunately, no monitoring data was available for the vibrations arising from the demolition of the former Stock Exchange as a whole, or for the collapse. It was therefore not possible for the expert witnesses to establish from records the actual nature and magnitude of the vibrations which the collapse caused (Footnote: ⁵¹).

However, case history data from monitoring the ground borne vibrations caused by demolition of buildings, falling weights and blasting has been used, in the specialist literature (Footnote: ⁵²), to develop formulae for assessing the likely attenuation of PPV with increasing distance from the source of the vibration (Footnote: ⁵³). Both experts relied, to an extent, upon these formulae in order to assess the likely PPV experienced at Grocers’ Hall as a result of the collapse.

Most attenuation calculations derive from an equation originally developed by Wiss which, as set out in section 2.6 of CIRIA, is as follows:

This equation, however, involves two empirical factors, “k” and “x”, which have a wide range of possible values. Factor “k” depends on ground profile and, according to CIRIA, can vary from 0.1 to 1.5 (for “E” in Joules). Also according to CIRIA, factor “x” varies from 0.8 to 1.5, depending on soil characteristics. This wide range of possible values for these two empirical factors means that, at least without on-site testing, the use of this equation involves a considerable degree of estimation, and consequent uncertainty.

Human responses to vibration

The British Standards

The two principal British Standards that deal with the human perception of construction vibration are BS 6472-1 (2008) - Guide to evaluation of human exposure to vibration in buildings. Part 1: Vibration sources other than blasting; and BS 5228-2 (2009) - Code of practice for noise and vibration control on construction and open sites. Part 2: Vibration. Both experts accepted these documents as authoritative, and relied upon them.

The guidance in BS 6472-1 is expressed in terms of “vibration dose value” and weighted acceleration (Footnote: ⁵⁵). That in BS 5228-2 is expressed in terms of PPV, on the basis that it is “more appropriate .. since this parameter is likely to be more routinely measured .. [and] .. since many of the empirical vibration predictors yield a result in terms of PPV” (Footnote: ⁵⁶).

As BS 6472-1 points out, the way in which people perceive building vibration depends upon various factors, including the vibration frequency and direction (Footnote: ⁵⁷). Generally, vertical vibrations are more perceptible than horizontal vibrations (although at very low frequencies this tendency is reversed). Whether a person in a building senses a given vibration magnitude also depends on the general background level of noise and vibration, what that individual is doing, where the individual is in the building, and how the person is “coupled” to the building. A person moving around is less likely to notice vibration than a person who is still. If a person is standing on the floor, the coupling is most direct. If the person is seated, the coupling is less direct, particularly if the seat is upholstered. However seat resonance might increase vibration (Footnote: ⁵⁸). Perception thresholds also vary widely from individual to individual: and the extent to which any given individual reacts may also depend upon related or unrelated effects which occur at the same time, such as structure-borne noise, airborne noise, induced rattling (for example of windows, furniture, loose fittings or ornaments), visual effects (such as the swinging of suspended light fittings, or movement of reflections on computer monitor screens or mirrors) (Footnote: ⁵⁹).

According to BS 5228-2, human beings are known to be very sensitive to vibration. The threshold of perception is typically in the PPV range of 0.14 to 0.3 mm/s (Footnote: ⁶⁰). Vibrations above these values can disturb, startle, cause annoyance or interfere with work activities. At higher levels, they can be described as unpleasant or even painful (Footnote: ⁶¹). BS 5228-2 includes the following table (Footnote: ⁶²):

The Modified Mercalli Intensity Scale

The Mercalli intensity scale is a scale used for measuring the intensity of an earthquake, by reference to its observed effects. It was originally created at the end of the 19^th century by the Italian volcanologist Giuseppe Mercalli, but has been revised and refined on a number of occasions since then. As so revised, it is usually known as the Modified Mercalli (“MM”) scale.

A copy of the MM scale, giving 13 levels, was annexed to the report produced by Watts and Partners in November 2005. A slightly different version, with only 12 levels, was exhibited by Mr Ballard to his June 2007 report, as follows:

Watts and Partners’ November 2005 stated that “the reported observations by the Grocers Company as to the scale of the jolt” included perceived grades up to and including category VIII on the MM scale (Footnote: ⁶³). Mr Ballard, in his June 2007 report, expressed doubt about the correctness of this categorisation of the perception of the event at Grocers’ Hall (Footnote: ⁶⁴).

The particular relevance of the MM scale to the present case is that a 1999 paper by Wald et al (Footnote: ⁶⁵) indicates a regression relationship between the perceived intensity on the MM scale and PPV, enabling the one to be translated very approximately into the other. Again, both experts relied, to an extent, upon this relationship as a method of assessing the PPV experienced at Grocers’ Hall as a result of the collapse. However, it was also common ground between them that, since both the MM scale and the regression analysis were concerned with earthquakes rather than the effects of a collapse, the results could give no more than a general indication which required to be confirmed by other evidence.

Structural responses to vibration

The science behind the prediction of the effect of ground borne vibration on buildings and other man-made structures has advanced over the years.

A detailed review of the then available studies and standards (including those from France, Germany, Sweden, Switzerland and the UK) was set out in CIRIA in 1992 (Footnote: ⁶⁶). Having reviewed all of the sources, CIRIA noted that there was then no definitive British Standard, and so proposed some “broad, tentative guidelines” as follows (Footnote: ⁶⁷):

It would appear that damage is unlikely (even improbable) at true peak particle velocities of less than 2 mm/s measured on structural foundations. Exceptions could be previously damaged or dilapidated buildings and monuments. The effects of vibrations also depend, however, on the predominant frequency: for a given value of PPV, the risk decreases for higher frequencies and increases for lower frequencies ..

.. For PPV values of between 2 and 10 mm/s generated by impact piling there is the increasing possibility of plaster cracking, of the enlargement of existing (visible) defects and of cracking in finishes. Undamaged buildings of modern construction to a high standard, and which have been well maintained, should not be affected by intermittent vibrations.

At PPV values between 10 and 50 mm/s, damage is an increasingly high risk, but again much depends on the geometry (structural form), the materials and the construction quality of the building e.g. well stiffened industrial buildings are less likely to be seriously damaged ..

Above 50 mm/s some form of damage is likely, which might include adverse effects on load-bearing units.

These guide values should be reduced to about 50% for continuous vibrations and further reduced if the buildings are dilapidated or have been damaged beforehand.

As was pointed out in CIRIA (Footnote: ⁶⁸), the response of buildings to ground-borne vibrations is governed by a number of factors including the relationship between the natural frequencies of the building and its elements and the characteristic frequency of the ground vibration; the magnitude of vibration; the stiffness of the building and its elements; the damping characteristics of the building; and other factors such as the height and dimensions of the building and the types of materials used in its construction. CIRIA also noted that:

Cases where vibration has caused serviceability damage to buildings are rare and those reported note that levels of vibration become unpleasant and even painful to the occupier before such damage results.

When investigated, aesthetic damage, such as plaster cracking or loosening of roofing tiles, which was initially attributed to vibrations, is often found to have been produced by other causes. In many cases, such defects have only been noticed as a result of inspection through concern produced by the onset of vibrations - but were there previously. In other cases, it is possible that there were incipient defects made apparent by the vibrations ..

Shortly thereafter, in May 1992, BS 5228-4 (1992) Noise control on construction and open sites – Part 4: Code of practice for noise and vibration control applicable to piling operations was published. In section 8.4.1, it noted (as CIRIA had also done) that documented proof of actual damage to structures or their finishes resulting solely from piling vibrations was rare. However it went on to recommend that:

… for soundly constructed residential property and similar structures which are in generally good repair, a conservative threshold for minor or cosmetic (Footnote: ⁶⁹) (i.e. non-structural) damage should be taken as a peak particle velocity of 10 mm/s for intermittent vibration .. Below these vibration magnitudes, minor damage is unlikely to occur.

Current experience suggests that these values may be reduced by up to 50% where the preliminary survey reveals existing significant defects (such as a result of settlement) of a structural nature, the amount of the reduction being judged on the severity of such defects.

.. Buildings constructed for industrial and commercial use exhibit greater resistance to damage from vibrations than normal dwellings, and it is recommended that light and flexible structures (typically comprising a relatively light structural frame with infill panels and sheet cladding) should be assigned thresholds of 20 mm/s for intermittent vibrations .. whereas heavy and stiff buildings should have higher thresholds of 30 mm/s for intermittent vibrations ..

.. Where buildings appear not to conform precisely to one or other of the descriptions given in this subclause, the thresholds may be adjusted within those stated ..

Section 8.4.1 also noted that that “the range of frequencies excited by piling operations in the soil conditions typical in the UK is between 10 Hz and 50 Hz”, and that “acceptable values of PPV may need adjustment for predominant frequencies outside this range”. A note to this section suggested that “At low frequencies (below 10Hz), large displacements and associated large strains necessitate lower ppv values (50% lower), whereas at high frequencies (above 50Hz), much smaller strains allow the ppv limits to be increased (100% higher)”.

BS 7385-2 (1993) Evaluation and measurement for vibration in buildings - Part 2: Guide to damage levels from ground borne vibration was published the following year, in November 1993. In Annex A, it listed some of the many causes of cracking in buildings and noted that:

.. cracks normally exist to varying degrees in buildings not subjected to vibration and are not, in themselves, an indication of vibration-induced damage ..

In its Table 1 (set out below) BS 7385-2 gave a series of suggested “limits for transient vibration, above which cosmetic damage could occur” (Footnote: ⁷⁰). These were slightly different to those given in BS 5228-4 (1992), for reasons later explained in BRE Digest 403 of March 1995 (in part to do with the derivation of the standards, and the use in BS 7385-2 of peak component particle velocity rather than peak particle velocity).

The nature of the vibration guide values given in BS 7385-2 was explained in paragraphs 7.4.1 and 7.4.2:

The vibration levels suggested are judged to give a minimal risk .. of vibration-induced damage.

Some data .. suggests that the probability of damage tends towards zero at 12 mm/s peak component particle velocity. This is not inconsistent with an extensive review of the case history information available in the UK …

.. Minor damage is possible at vibration magnitudes which are greater than twice those given in Table 1, and major damage to a building structure may occur at values greater than four times the tabulated values ..

This 1993 guidance was reproduced unchanged in 2009, in table B2 of BS 5228-2 (2009). Paragraph B.3.1 of BS 5228-2 again pointed out that:

Extensive studies carried out in the UK and overseas .. have shown that documented proof of actual damage to structures or their finishes resulting solely from well-controlled construction and demolition vibrations is rare. There are many other mechanisms which cause damage, especially in decorative finishes, and it is often incorrectly concluded that vibrations from construction and demolition sites are to blame.

In some circumstances, however, it is possible for the vibrations to be sufficiently intense to promote minor damage. Typically this damage could be described as cosmetic and would amount to the initiation or extension of cracks in plasterwork, etc, rather than the onset of structural distress. In more severe cases, falls of plaster or loose roof tiles or chimney pots can occur ..

Paragraph B.3.2 of BS 5229-2 also explained that:

.. The damage threshold criteria presented in BS 7385-2 are based upon systematic studies using a carefully controlled vibration source in the vicinity of buildings. Strains imposed in a building by ground motion will tend to be greater if lower frequencies predominate. The relative displacements associated with cracking will be reached at higher vibration magnitudes with higher frequency vibration. BS 7385-2 provides frequency-dependent threshold levels which are judged to give a minimal risk of vibration-induced damage ..

The conflicting views of the expert witnesses

Against that background, I now turn to set out the conflicting views of the expert witnesses.

Mr Jessep

The Arup model

Mr Jessep arranged for the well-known firm of consulting engineers Ove Arup & Partners (“Arup”), to produce a computer model of the collapse, so as to predict the vibrations that would have reached Grocers’ Hall. A visual representation of this model was played to me as part of the evidence. Arup’s Technical Note dated 26 January 2012 is annexed to Mr Jessep’s first report. As recorded in that Note (Footnote: ⁷¹):

.. The accurate and reliable prediction of vibration transmission, given the input information available and the complexity of foundation structures lying on the transmission path and other factors, is not possible.

Our approach, as has been agreed by [Mr Jessep], is to make our best estimate of vibration transmission, accompanied by a description of sources of uncertainty and where possible estimate of their effects in order to assess an upper bound of vibration level at Grocers’ Hall.

Arup observed that there was “significant uncertainty associated with many of the modelling assumptions, including the leading characteristics in the soil” (Footnote: ⁷²). They therefore modelled 3 collapse scenarios, 4 loading durations, 3 levels of soil stiffness, and 3 levels of damping, all producing predicted results at three different locations in Grocers’ Hall. These variables produced predicted PPVs at Grocers’ Hall of between 0.00 mm/s (assuming, inter alia, a loading duration of 2.0 seconds) and 12.1 mm/s (assuming, inter-alia, a loading duration of 0.2 seconds and a 50% reduction from their base assumption as to the stiffness of the soil).

From these figures, Arup noted that both loading duration and soil stiffness had particularly significant effects on the calculated response. On the basis of their Collapse Scenario 3 (300t falling 8.45m as a single mass), and a 0.2 second loading duration, Arup reported that their model predicted a PPV of between 3 mm/s and 12 mm/s, depending upon the stiffness of the soil (Footnote: ⁷³).

Arup reported that these figures compared well with figures taken from the attenuation chart produced by Woods & Jedele in 1985, which suggested a PPV of between 5.3mm/s and 13mm/s at the relevant distance (Footnote: ⁷⁴). However, they noted that the prediction from their model represented “an upper bound estimate, and is not intended to represent the most likely event”.

Mr Jessep’s interpretation of Arup

In his report and his oral evidence, Mr Jessep sought to narrow the uncertainties taken into account by Arup. He concluded (partly on the basis of his view of the evidence as to the perception of the event at Grocers’ Hall) that the loading impulse would probably have been between 0.2 and 0.5 seconds, which would give predicted PPV’s of between about 7.5 mm/s and about 1.3 mm/s, on Arup’s base case assumptions as to soil stiffness. Since “doubling the stiffness of the soils leads to unrealistic [ie too low to produce the effects noticed by Brigadier Rawlins and others at Grocer’s Hall] values” (Footnote: ⁷⁵), Mr Jessep discounted those figures in Arup’s analysis, recommending instead that “the effect of halving the stiffness of the soil .. is worthy of consideration” (Footnote: ⁷⁶).

Mr Jessep’s overall conclusion was accordingly that the PPV of the vibrations at the Grocers’ Hall caused by the collapse was likely to have been in the range of about 1.5 mm/s to 12 mm/s.

The causes of damage

That entire range falls well below the vibration guide values given in the tables in BS 7385-2 (1993) and BS 5228-2 (2009) as “judged to give a minimal risk of damage”. Its highest point only just reaches the figure of 12mm/s, below which BS 7385 (1993) notes that “some data .. suggests that the probability of damage tends towards zero”. It was nevertheless Mr Jessep’s view that the effect of the collapse caused or contributed to the cracking of the walls at Grocers’ Hall, and caused the fracturing of the cistern.

Mr Jessep gave four principal reasons for this view:

Mr Allen

Observation and common-sense

In Mr Allen’s view, the collapse would have had no significant impact on the Grocers’ Hall. He relied principally upon the following two observational factors in support of his conclusion.

Calculations

Mr Allen did not commission an alternative computer model to that produced by Arup. However, he did provide his own calculations:

Mr Allen also tested his own and Mr Jessep’s proposed PPV figures by putting them into the “Wiss” equation, to see what figures they yielded for the empirical constant “k”.

The cracking

In Mr Allen’s view, the level of vibration experienced by Grocers’ Hall, whilst within the range perceptible to some of the people in it, would not have been at a level sufficient to cause cracking in the building itself.

In his first report, Mr Allen expressed the view the pattern of cracking observed and reported by WhitbyBird is “typical of that associated with shear and yield line stress patterns, where undersized floors are utilised in buildings”. Furthermore, a large number (and the most significant) of the cracks occur in the northern part of the building, where it appears that the “jolt” from the collapse was not noticed. Many of the cracks are recorded in locations such as cupboards and the lift well, which are infrequently decorated and infrequently examined: and cracking has continued to occur in certain locations many years after the date of the collapse in 2005.

Mr Allen therefore concluded that the major cracks were the result of “long-standing distortion and movement” consequent upon the poor quality and strength of the original construction of the building, and so had “occurred gradually from the date of construction in 1970, possibly having been repaired during redecoration in the mid-1990s”. Other cracks were simply the result of differing rates of thermal expansion and differing effects from moisture variation, of the kind that always occur in buildings whose components are constructed from differing building materials.

In his Supplemental Report, Mr Allen revised this view on the basis of the evidence in the witness statements of Mr Hogwood and Mr Griffin (that he had by then seen), and expressed the opinion that “the [new] cracking to the walls occurred most probably in late September or early October [2005]”, and was a side-effect of the flood.

The damage to the cistern

In Mr Allen’s view, the cracking at the head of the bath near the cistern was probably caused by someone leaning on the tiles. He would have expected to see cracking of the tiles behind the cistern if there had been sufficient vibration to crack the cistern itself.

As for the probable cause of the damage to the cistern, Mr Allen identified two possible causes, each of which he considered to be a more likely cause than the collapse. Those causes are: (1) a manufacturing defect leading to “unexplained and sudden failure of ceramics many years after manufacture” (Footnote: ⁷⁹); and (2) “misuse, or at least someone being heavy-handed when using the toilet” (Footnote: ⁸⁰).

Mr Jessep’s response

Mr Jessep’s responses to the views expressed by Mr Allen were, in summary, as follows.

Mr Jessep took issue with Mr Allen’s use of the MM scale. In Mr Jessep’s view, the collapse was not an earthquake, and the use of the MM scale and the Wald conversion was therefore questionable. At best it could only be a broad guide.

Mr Jessep also pointed out that Mr Allen had only used a frequency of 10 Hz. According to Mr Jessep, at frequencies between 1 Hz and 8 Hz the PPV for MM Scales II and III (Mr Allen’s estimated levels) would be between 0.33 mm/s and 21.86 mm/s. At 5 Hz, the PPV would be somewhere between 3.12 mm/s and 9.37 mm/s. It was Mr Jessep’s view that these were consistent with those derived from Arup’s analysis, and “within the range that are sufficient enough they could cause damage to the Grocers’ Hall”.

However, Mr Jessep in any event disagreed with Mr Allen’s placing of the perception of the event at Grocers’ Hall on the MM scale. Mr Jessep’s assessment (taking into account the evidence concerning the sensation of a “jolt”, the movement of the glasses, the display screen equipment shaking, pictures moving, and cracks to the plaster and walls) was that the event taken as a whole was “probably in principle in the broad range of MM IV to VII, most likely Scale IV or possibly V”.

As to the absence of damage to buildings closer to the collapse than the Grocers’ Hall, this would (in Mr Jessep’s view) “not necessarily be expected because the PPV would be lower than the threshold for cosmetic damage for reinforced and framed buildings and heavy commercial buildings”.

Mr Jessep rejected the criticism made by Mr Allen of Mr Jessep’s suggested PPV of 12 mm/s by reference to the “Wiss” equation and the figure which that produced for the empirical concept “k”. Mr Jessep at first sought to defend his figure (as he had already done in paragraph 4.20 of his Supplemental Report) by reference to the values given for “k” in the data given in Appendix 1 of CIRIA. However, as was pointed out to him (and as he was obliged to agree), the values in Appendix 1 are expressed in Kilojoules, whereas the range given in the section of CIRIA dealing with this calculation expressly states that it uses energy values, not in Kilojoules, but in Joules, which produces a lower “k” value (Footnote: ⁸¹). Mr Jessep’s attempted defence of his figures on this particular basis was therefore misconceived.

Mr Jessep was not re-examined on that point. In cross-examination of Mr Allen, however, the point was made that the “Wiss” equation involves not just one, but two empirical constants. In his calculations, Mr Allen had used the figure of 1.5 for “x”, on the basis that that was what was recommended in CIRIA for clay soils. It was suggested to Mr Allen that some of the data in CIRIA gave a value for “x” derived from driving piles through gravel or silt into London clay of around 0.8 or 0.9. It was also put to Mr Allen (and he agreed) that running the same calculation on the basis that “x” = 0.86 produced a figure for “k” (on a PPV of 12) of 0.263, comfortably inside the bottom end of the range suggested in CIRIA.

Mr Jessep also criticised Mr Allen’s use of the “Wiss” equation to predict PPV figures for distances as close as 25 m from the collapse. In his view, the resulting figures were “well outside the range of data recorded in case histories from blasting”, and Arup’s figures of 20 mm/s to 50 mm/s at that range should be preferred. Since Arup’s figures are below the threshold at which damage would be expected, the absence of damage to other buildings was not, in Mr Jessep’s view, significant. In any event, in Mr Jessep’s opinion, the Royal Exchange building did suffer some damage.

As to the cracks around the bath, Mr Jessep laid stress on the cracks in the tiles attached to the walls, which (in his view) were “consistent with the floor on which the bath (and adjacent toilet/cistern) is situated moving and rotating differentially from the walls”. This “apparent rotation of the floor would be expected to cause tension forces in the wall fixings of the cistern, which would be transferred through the screwed fixings and so not necessarily cause any shear stress in the adjacent wall ties” (Footnote: ⁸²).

The law

Counsel for Keltbray laid stress upon the fact that, in a case such as this, the claimant bears the burden of proof throughout, and that speculation, even sophisticated speculation, is not proof. He drew my attention to the well-known observations of Lord Brandon of Oakbrook in Rhesa Shipping Co SA v Edmunds (“The Popi M” (Footnote: ⁸³)):

My Lords, the appeal does not raise any question of law, except possibly the question what is meant by proof of a case “on a balance of probabilities.” Nor do underwriters challenge before Your Lordships any of the primary findings of fact made by Bingham J. The question, and the sole question, which Your Lordships have to decide is whether, on the basis of those primary findings of fact, Bingham J. and the Court of Appeal were justified in drawing the inference that the ship was, on a balance of probabilities, lost by perils of the sea.

In approaching this question it is important that two matters should be borne constantly in mind. The first matter is that the burden of proving, on a balance of probabilities, that the ship was lost by perils of the sea, is and remains throughout on the shipowners. Although it is open to underwriters to suggest and seek to prove some other cause of loss, against which the ship was not insured, there is no obligation on them to do so. Moreover, if they chose to do so, there is no obligation on them to prove, even on a balance of probabilities, the truth of their alternative case.

The second matter is that it is always open to a court, even after the kind of prolonged inquiry with a mass of expert evidence which took place in this case, to conclude, at the end of the day, that the proximate cause of the ship's loss, even on a balance of probabilities, remains in doubt, with the consequence that the shipowners have failed to discharge the burden of proof which lay upon them.

In answer to that submission, counsel for the Grocers relied upon the observations of Thomas LJ (as he then was) in Ide v ATB Sales Ltd (Footnote: ⁸⁴) that The Popi M was “very unusual”, in that it was a case in which all of the experts put forward explanations of the cause of the loss which were acknowledged to be highly improbable, and each explanation was supported as the most likely explanation only because any other hypothesis was regarded as almost (if not altogether) impossible. In the Grocers’ submission, this is not such a case, because it was not in any way “highly improbable” that the collapse could have been the cause of the damage.

The Grocers accepted that they bore the burden of proof: but they relied, in that connection, upon the often-quoted words of Lord Reid in McWilliams v Sir William Arrol & Co (Footnote: ⁸⁵), that:

.. [I]n the end when all the evidence has been brought out it rarely matters where the onus originally lay, the question is which way the balance of probability has come to rest ..

Taking all of these authoritative statements into account, it seems to me that the question which I have to answer is whether, taking all the relevant and admissible evidence into account, I am satisfied on the balance of probabilities that the collapse was the cause of the breakage of the cistern. Although the process of answering that question is (as in many such cases) both technical and complex, the question itself is fundamentally a simple one. As Toulson LJ (as he then was) helpfully explained in his Judgment in Milton Keynes BC v Nulty (Footnote: ⁸⁶):

[34] A case based on circumstantial evidence depends for its cogency on the combination of relevant circumstances and the likelihood or unlikelihood of coincidence. A party advancing it argues that the circumstances can only or most probably be accounted for by the explanation which it suggests. Consideration of such a case necessarily involves looking at the whole picture, including what gaps there are in the evidence, whether the individual factors relied upon are in themselves properly established, what factors may point away from the suggested explanation and what other explanation might fit the circumstances. As Lord Mance observed in Datec Electronic Holdings Ltd v United Parcels Service Ltd [2007] 1 WLR 1325, paras 48 and 50, there is an inherent risk that a systematic consideration of the possibilities could become a process of elimination “leading to no more than a conclusion regarding the least unlikely cause of loss”, which was the fault identified in The Popi M. So at the end of any such systematic analysis, the court has to stand back and ask itself the ultimate question whether it is satisfied that the suggested explanation is more likely than not to be true. The elimination of other possibilities as more implausible may well lead to that conclusion, but that will be a conclusion of fact: there is no rule of law that it must do so. I do not read any of the statements in any of the other authorities to which we were referred as intending to suggest otherwise.

35 The civil “balance of probability” test means no less and no more than that the court must be satisfied on rational and objective grounds that the case for believing that the suggested means of causation occurred is stronger than the case for not so believing ..

[37] .. In deciding a question of past fact the court will, of course, give the answer which it believes is more likely to be (more probably) the right answer than the wrong answer, but it arrives at its conclusion by considering on an overall assessment of the evidence (ie on a preponderance of the evidence) whether the case for believing that the suggested event happened is more compelling than the case for not reaching that belief (which is not necessarily the same as believing positively that it did not happen).

Analysis and conclusions

Against that background, I now turn to analyse the evidence and to express my conclusions.

Perception at Grocers’ Hall.

I begin with the perception of the collapse at, and near, Grocers’ Hall. It is not possible to establish from records the actual nature and magnitude of the vibrations caused by the collapse, because the collapse and its effects were not scientifically monitored or measured at the time. The best evidence of those effects is therefore the necessarily unscientific and approximate observations made by those who noticed those effects at the time. That is the approach adopted by both experts. Each of them, at various points, uses the contemporary observations as an element in, or as a check on, their reasoning and their calculations. I propose to do the same.

As I have already stated, it seems to me that Brigadier Rawlins, Rear-Admiral Ross and Mr Wood were all doing their best, in their evidence, to assist the Court to the best of their recollection. Their evidence was not challenged in cross-examination, and I accept without hesitation that each of them noticed the effects of the collapse at the time. Accordingly:

However, I am also satisfied, on the basis of the evidence contained in the reports of Mr Phillips of Burgoynes and of Mr Ballard of GBG, that there were other people who were present at Grocer’s Hall at the time of the collapse who did not notice it or any of its effects. Although this evidence is double hearsay, it appears in two apparently independent sources. It is consistent with what is said in BS 6472-1 (2008) and BS 5228-2 (2009) about the vagaries of human perception of vibrations. It accords with the probabilities, in that Grocers’ Hall is situated in the middle of one of the busiest areas of the City of London, where there is always a considerable amount of background noise and vibration (Footnote: ⁸⁷). Furthermore, this aspect of the reports of Mr Phillips and Mr Ballard was not directly contradicted by any other factual evidence: and, if evidence had been available to contradict it, it seems to me that the Grocers would probably have made some effort to locate it and to tender it to me.

Finally, on this aspect of the matter, I accept Mr Griffin’s evidence (supported by Mr Phillips’ report) that it was brought to Mr Griffin’s attention in about September 2005, much later than the actual flood itself, that glasses in a display case at Grocers’ Hall had moved. I am not, however, satisfied that the collapse was the direct cause of this movement. The process of cleaning up Grocers’ Hall after the flood caused major disruption and disturbance. A 30 foot length of cast iron pipe had to be removed and replaced with a new plastic one, the big table in the Court Room had to be dismantled and removed, so that the very large fitted carpet in the Court Room could be moved out, all the other flood-damaged items have to be removed, all the mess caused by the flood had to be cleaned up, and the relevant parts of the building dried, made-good and redecorated. These activities would themselves have generated a considerable amount of vibration within Grocers’ Hall. In my Judgment, it at least as likely that one or more of these activities in the period between the date of the collapse and the much later date when the movement of the glasses was discovered was responsible for that movement, than that that movement was directly caused by the collapse.

Taking all of these matters into account, I accept Mr Allen’s view that the event of the collapse, as perceived at Grocers’ Hall, fell into level II or level III on the MM Scale. At a frequency of 10 Hz, level II would equate to a PPV value of 0.26 mm/s, and level III would equate to a PPV value of 2.18 mm/s (Footnote: ⁸⁸). At a frequency of 5 Hz, level III would equate to about 4.37 mm/s (Footnote: ⁸⁹).

Mr Jessep argued in favour of the lower frequency on the basis of Arup’s Frequency Response Analysis, which showed that the response of the foundation slab at Grocers’ Hall was greatest between 1 Hz and 10 Hz, and Mr Bown’s analysis that the natural frequency of Grocers’ Hall, when undamaged, was between 2 Hz and 18 Hz (Footnote: ⁹⁰). Mr Allen argued in favour of the higher frequency, on the basis of a report of vibration at Grocers’ Hall caused by the demolition of the adjacent building at 1 Lombard Street equating to a frequency of 15 Hz. Mr Jessep countered by arguing that the vibration then measured was horizontal, whereas the vibration caused by the collapse was vertical. In my Judgment, no sufficient evidence is available to enable me to be confident in resolving that dispute one way or the other. I must therefore compare both figures with the other available evidence in reaching my conclusion as to the likely PPV resulting from the collapse.

In doing so, I must also bear in mind Mr Jessep’s observation that the collapse was not an earthquake, and that calculations by reference to the MM scale can only be an approximate guide.

Damage to other buildings

I now turn to consider the second factor relied upon by Mr Allen, the absence of reported damage to other, nearer buildings.

There is an attractive simplicity to Mr Allen’s argument that, if the effects of the collapse were strong enough to cause damage at Grocers’ Hall, they would necessarily have been strong enough to cause very substantial damage to the large number of buildings that were much closer to the collapse site.

However, it was the flood that made the events at Grocers’ Hall particularly damaging and which gave rise to this claim. Occupants whose buildings suffered less significant damage - some cosmetic or minor cracking to plaster surfaces, or even (as reported at the Royal Exchange) a few broken windows - may either not have noticed that damage at all, or (if they noticed it) may have thought that the amount involved was not worth the effort and cost of investigating the cause of the damage and of making a claim for the cost of repair.

After all, cosmetic damage (and even minor damage) is a regular occurrence in many buildings. There are many mechanisms which cause damage, especially in decorative finishes. Cosmetic or minor cracking typically passes unremarked at the time it occurs, is repaired as part of the usual cycle of redecoration, and recurs often quite soon thereafter. Moreover, investigating and making a claim is an expensive exercise (as this case has amply demonstrated).

In my judgment, therefore, Mr Allen is right to say that the absence of reported damage to any other building (apart from the Royal Exchange) sets an upper bound to the likely level of vibration generated by the collapse. However, that upper bound is higher than that suggested by Mr Allen, and is set by the level at which sufficient damage would probably have been caused to make it seem economically worthwhile for the building occupant or owner to initiate an investigation and to make a subsequent claim.

The Table in BS 7385-2 (1993) and BS 5228-2 (2009) indicates a PPV threshold of 50 mm/s for properly constructed and undamaged reinforced or framed structures and industrial and heavy commercial buildings, as the level below which cosmetic damage is unlikely. They also say that “major damage to a building structure may occur at values greater than four times the tabulated values”. On the basis of that authoritative guidance, the upper bound to which I have referred would, in my judgment, be a PPV of about 200 MM/s.

Calculations

The computer model produced by Arup is, in my judgment, a useful tool for understanding the likely vibration levels emanating from the collapse. Mr Allen did not strongly argue to the contrary. However, the limitations of the Arup model must be borne in mind. As Arup themselves noted, the many unknown and/or un-measurable factors involved mean that even the most sophisticated computerised model cannot give an accurate and reliable picture of what actually happened. What it can do - and what Arup have in fact sought to do - is to indicate a range of possibilities, and to indicate the probable effect on that range of changes to particular variables.

The two major variables identified by Arup are the loading impulse (sometimes referred to as the “loading duration”), and the soil properties.

As to the loading impulse, the experts’ agreement that the energy would have been imparted “essentially as a single shock”, Mr Allen’s description of the collapse as resulting in the collapsing slabs and rubble falling together “essentially as a single lump”, and the fact that the collapsing slabs and rubble fell onto a concrete slab supported on piled foundations, all suggest that the loading impulse would have been comparatively short. However, the collapsing slabs and rubble were dispersed over an area of about 130 m², and were (as Mr Jessep accepted) unlikely to have fallen perfectly uniformly.

Taking all of these factors into account, in my judgment the loading impulse would probably have been towards the upper end of the 0.2 - 0.5s range given by Mr Jessep, and so would probably have been about 0.35 to 0.5s.

As for the properties of the soil, I am not persuaded by the arguments put forward by Mr Jessep that any adjustment is required to the “base case” soil properties used by Arup in their model. These were derived from other Arup projects around the same area for the same soil type. Mr Jessep suggests that the effect of halving the stiffness of the soil is “worthy of consideration”, because the made ground “may not be as stiff as assumed by Arup” (Footnote: ⁹¹). However, at this point in his argument, it seems to me that Mr Jessep was at least in part engaged in a process of adjusting the figures to produce a result that he regarded as consistent with other evidence.

Conclusion as to energy level

It was not seriously in dispute that, of the three collapse scenarios considered by Arup, Scenario 3 was to be preferred.

Applying the conclusions which I have reached about the loading impulse and the stiffness of the soil to the modelling results which appear in Figure 16 of the Arup Note would suggest a PPV at Grocers’ Hall resulting from the collapse of somewhere roughly between about 1.3 mm/s and about 4 mm/s.

To reach a conclusion on this issue, it is then necessary for me to compare that range with the ranges suggested by the other evidence available to me.

Taking all of these matters into account, in my judgment the evidence as a whole establishes on the balance of probabilities that the PPV experienced at Grocers’ Hall as a result of the collapse was somewhere roughly in the range between about 1 and 4 mm/s.

The cracking

In reaching my conclusion as to the likely PPV at Grocers’ Hall, I have not taken into account the cracking which was observed there in about late September 2005, and which Mr Jessep attributes to the collapse. I have not done so, because to do so would be to risk reasoning in a circle. I cannot start from the assumption that the collapse caused the cracking in order to determine whether the vibration resulting from the collapse was sufficiently strong to cause the cracking.

In Mr Jessep’s view, a PPV of 3 mm/s or 4 mm/s (which would be at the upper end of the PPV range that I have found) would be sufficient to cause cracking on the third and fourth floors of Grocers’ Hall, because of the unique problems with the design and construction of that building (Footnote: ⁹³). In Mr Jessep’s opinion, “relatively modest vibrations of low-frequency that would not normally cause any damage to properly designed, constructed and maintained buildings, could cause damage to the Grocers’ Hall” (Footnote: ⁹⁴).

I accept Mr Jessep’s view to this limited extent: that PPVs even as low as 3 or 4 mm/s may perhaps sometimes cause some cracking. That view is supported to a degree by the comment in CIRIA that “for PPV values of between 2 and 10 mm/s generated by impact piling there is the increasing possibility of plaster cracking, of the enlargement of existing (visible) defects and of cracking in finishes” (Footnote: ⁹⁵), and (to a more limited extent) by Section 8.4.1 of BS 5228-4 (1992), which states that its suggested 10 mm/s threshold may fall to be reduced by up to 50% where the preliminary survey reveals existing significant defects (such as a result of settlement) of a structural nature, and by up to a further 50% where the vibration frequency is below 10 Hz (Footnote: ⁹⁶).

However, the type of cracking that vibration at levels of PPV between 1 and 4 mm/s might (in certain cases) cause is, in the nature of things, likely to be very limited: new hairline cracks at junctions or discontinuities of materials and/or the reopening of cracks in previously cracked finishes. Those levels of PPV are well below the figures given in the later (and so currently authoritative) BS 7385-2 (1993) and BS 5228-2 (2009) as the threshold below which even cosmetic cracking is unlikely. In my judgment, vibration at those sorts of levels of PPV is unlikely to have been sufficient to cause the more significant cracking that was noted after the flood, particularly at the northern end of the Third and Fourth Floors of Grocers’ Hall. That is so, even though Grocers’ Hall is a “poorly detailed” building that “would be prone to movement and distortion and consequent damage”.

I reach this view in part on the basis of the statements in BS 7385-2 (1993) and BS 5228-2 (2009) to which I have referred above. Grocers’ Hall may be “poorly detailed”: but it was not uniquely vulnerable to damage by vibration. There is no support in these British Standards (or, indeed, in any of the other technical literature to which I have been referred) for the proposition that vibration with a PPV of less than 5 mm/s has ever actually been observed to cause anything more than hairline cracks of the kind which I have described.

However, I also reach this view on the basis of the specific facts of this case.

First of all, the delay in observation of the cracks seems to me to be significant. It was part of Mr Jessep’s argument that the cracks “would be relatively obvious [so that he] would have expected them to be reported quite soon after they appeared” (Footnote: ⁹⁷). It was also Mr Griffin’s evidence that “the Hall is maintained to a very high standard” (Footnote: ⁹⁸): and Mr Hogwood’s evidence both that he himself carried out a daily inspection of the Hall, and that every floor of the building was checked each night by the person locking up.

Against that background, it is significant that the cracks which Mr Jessep attributes to the effect of the collapse were not noticed until late September 2005, some five or six weeks after the date of the collapse on 10 August 2005. A part of that period might perhaps be explained by the fact that the Grocers were preoccupied with clearing up after the flood. However, the same explanation cannot be given for the 9-day period between the date of the collapse and the date of the flood, during which no one at Grocers’ Hall thought to suggest that the collapse had caused any damage of any kind to the building. On the contrary, the collapse seems to have been treated, even by those (such as Brigadier Rawlins) who noticed it at the time, as just one of those things that happen in a crowded city such as London. It does not seem to have been regarded as significant. Mr Griffin was on holiday at the time of the collapse, but returned before the flood: and it was his evidence that no one told him about the collapse until about October 2005 (Footnote: ⁹⁹).

I am not persuaded by Mr Jessep’s suggestion that the fact that the cracks were not noticed before late September 2005 can be explained “by a general increase in crack width due to creep, making cracks more evident over time and also tearing wallpaper revealing previously hidden cracks” (Footnote: ¹⁰⁰). Nor am I persuaded by the evidence (for example in the initial report from Watts and Partners) that the cracks appeared new in late 2005. I accept Mr Allen’s view (for the reasons that he gives) that it is notoriously difficult to age cracks precisely.

Secondly, the location of the cracks seems to me to be significant, in two respects. First, a large number of the cracks (including the noticeable crack and deflection between the doors of bedrooms 8 and 9) appeared in the northern part of the building, in which the collapse seems to have gone largely unnoticed by the occupants. As Mr Allen points out, that is the part of the building where the beams have the longest span and which has the least robust detailing. Secondly, some of the most significant cracks appear in lift shafts and cupboards, and other places where they are very likely to have gone wholly unnoticed (or, if noticed, to have been ignored). That makes it difficult to draw any inference from the fact that the cracks had gone unremarked prior to the collapse about the time when those cracks first appeared.

Thirdly, the evidence establishes that the cracking was not a “one-off” event in August 2005. Cracks continued to occur and recur, and to grow, at least until the floors of the building were opened up and strengthening steelwork was put in place. I find it improbable that the effect on Grocers’ Hall of the collapse in August 2005 would still be manifesting itself in crack progression several years later. I do not accept that it would have taken that amount of time for the building to return to a steady state. That suggests to me (along with the other evidence to which I have referred) that these more major cracks long pre-date the collapse, but were only “discovered” as a result of the detailed attention which was paid to the fabric and finishings of Grocers’ Hall in the aftermath of the flood.

All of these factors combine to persuade me that the most probable explanation for the larger cracks is that given by Mr Allen, which is that they relate to the quality and strength of the original construction and would have occurred and become wider (or narrower) gradually after the building was constructed in 1970, depending upon loading and other transient circumstances.

In my judgment, therefore, the effects of the collapse did not cause the major cracking at Grocers’ Hall which was reported on following the flood.

Damage to the cistern

My finding that the collapse did not cause the more major cracking does not, in and of itself, answer the question whether the effects of the collapse caused the damage to the cistern.

There are logically only two possible causes of the damage to the cistern: inherent defect and/or subsequent mechanical damage. Grocers’ case, supported by Mr Jessep, is that the cause was mechanical damage, caused by differential rotation of the floor to which the lavatory pan was attached and the wall to which the cistern was screwed. In Mr Jessep’s opinion, that differential rotation was caused by the effect of the “jolt” on the weak structure of the Third Floor of Grocers’ Hall. Apart from his views as to the strength of the “jolt”, Mr Jessep relies upon two pieces of evidence in particular: the timing of the failure of the cistern; and the cracking that was recorded nearby.

Mr Allen’s view is that the most probable cause was either inherent defect (and that the timing is simply a coincidence), or mechanical damage caused by vigorous use and/or a combination of the two.

The timing of the escape of water from the cistern is a factor which could point in favour of Mr Jessep’s thesis. However, the weight that I can give to it is limited, because of the delay between the collapse and onset of the escape of water. Mr Jessep seeks to explain this away, on the basis that cracks in ceramics may take time to propagate. However, if that theory is correct, it seems to me that it could apply also to damage caused by over-tightening on installation or maintenance and/or by rough usage. It also seems to me to be at least equally likely that inherent defects could take years to manifest themselves. That was Mr Allen’s contrary thesis, and he produced some (albeit limited) external support for it.

As for the cracking around the bath, that would only be a relevant factor if I were satisfied that the collapse was the cause of that cracking. Largely for the reasons which I have already given in relation to the cracking in the walls, I think it more likely that Mr Allen’s account of the reasons for the cracking in the tiles is the correct one, and that those cracks were not caused by the effects of the collapse.

Two other factors also seem to me to be significant. The first is that each of the bedrooms on the Third Floor has a bathroom with a lavatory and cistern attached: yet the cistern in the bathroom attached to Bedroom 1/the Committee Room waiting area was the only one to fail. Mr Jessep did not suggest any reason why the floor at that particular location should have moved more strongly than anywhere else (Footnote: ¹⁰¹): and the evidence from the subsequent investigations indicates that the structure in the general area of the Committee Room was more soundly constructed than that further north in the building, which would suggest the contrary (a circumstance supported by the fact that the largest and most obvious crack was between the door heads of bedrooms 8 and 9). Nor did Mr Jessep give any convincing reasons why that particular cistern should have been particularly vulnerable to the effects of the collapse.

The second is my conclusion as to the limited amount of energy which would have reached Grocers’ Hall from the collapse. No evidence has been put before me to satisfy me that energy at that limited level would be likely to cause the damage which in fact occurred to the cistern.

The fact that the cistern was disposed of before it could be photographed or examined means that it is impossible to be certain of the cause of the damage. However in my judgment it is more probable that the cause was some combination of inherent defect and/or over-tightening on installation and/or rough usage (this being the lavatory for general use on that floor) than that it was caused by the effects of the collapse.

In my judgment, therefore, the Grocers have failed to make out their case that the damage to the cistern and the consequent flood were caused by the effects of the collapse.

Disposition

For these reasons, I dismiss this action.

I invite counsel to agree any consequential orders and to draw up a suitable Minute of Order. If any matters cannot be agreed, I will deal with them on the basis of written submissions, to be lodged within seven days of the date of the handing down of this judgment.

I am very grateful to counsel and their teams on both sides for their assistance.