England and Wales High Court (Technology and Construction Court) Decisions

IN THE HIGH COURT OF JUSTICE
QUEEN'S BENCH DIVISION
TECHNOLOGY AND CONSTRUCTION COURT

		5 November 2004

 


Introduction

1. This action concerns a residential training college in Coventry which was built for Eastern Telegraph Company Ltd. (ET) by Birse Construction Ltd. (Birse or BCL), to the design of the well-known architects, MacCormac Jamieson Prichard (MJP). It is a case about defects. ET is a wholly owned subsidiary of Cable and Wireless plc. ET had decided to move its training college from Porthcurno in Cornwall to somewhere that was more accessible to students from this country and many other countries where Cable and Wireless operates. The site was bought from Coventry City Council in November 1990. It is now no longer required by Cable & Wireless and ET is trying to sell it.
2. MJP's brief was to provide a high quality building of unique design. The college was not a conventional college. The building had to house a large amount of teaching accommodation, ranging from small rooms for classes in management to large, highly technical laboratories and workshops. The teaching buildings are arranged in a series of parallel strips to the south of the site with the residential accommodation in a long row to the north of the site. To the east of the site is a separate leisure centre which contained recreational facilities. The residences and the teaching blocks are united by an oval courtyard overlooked by the college dining room and the common rooms. A particular feature of the buildings was the wave roofs that were constructed over the teaching facilities. There were six sections of overlapping roofs. This was the first time that roofs of this profile and construction had been built in the United Kingdom. MJP gained architectural awards for the college. It thus comprises a number of structures, including: two teaching blocks with lecture and training rooms and a library; an administration building and restaurant; two three storey residential buildings where up to 160 students lived; a leisure centre with ancillary buildings; and four outbuildings. Miss Rosemary Jackson, for Birse, described the college in her submissions as "an arresting development with clean lines, materials selected for their particular aesthetic qualities, and eye-catching details such as the wave-form roofs on the teaching wings".
3. Birse entered into a contract with ET. It was made under seal by deed dated 6 February 1992 and incorporated the JCT Standard Form of Building Contract 1980 edition with Quantities together with the Contractor's Designed Portion Supplement. MJP were the Architect under the contract and Northcroft Neighbour & Nicholson (Northcrofts or NNN, now NNN Ltd) were the Quantity Surveyor. Birse's obligations under the contract were secured by a Bond (also of 6 February 1992) given by the New Hampshire Insurance Company in favour of ET for an amount up to £1,433,065.63.
4. The college was built between about October 1991 and the end of 1993. On 22 December 1993 MJP issued a certificate recording that practical completion had been achieved on 17 December 1993. However as a consequence of delays in construction (some of which resulted in extensions of time up to 16 April 1993) from 4 August 1993 MJP began to issue the first of many partial possession certificates so as to enable ET to take possession prior to practical completion of the whole in order that it could start fitting out using its contractors and could start teaching on 6 September 1993. (Students arrived a few days before on 4 September.) This was probably a little earlier than was advisable, given the number of defects that had to be attended to subsequently. However it was in the interests of both parties that ET should move in. Some seven certificates were issued, three after 4 September 1993. I ought here to record, first, that I did not get the impression that the delays in completion affected the standards of work, as sometimes is the case, and, secondly, although the trial necessarily focussed on the quality of workmanship, the documents and evidence did not establish that the overall performance of Birse was below average, although, as will appear, there were too many defects. The contemporary criticisms were not out of the ordinary. Birse's management might well have been better but it was not particularly bad. Miss Jackson had reason, in my judgment, to preface her opening submissions with a quotation from "Concrete" when the college was in 1995 the Winner in the Building Category Awards:

"The finishes, both in situ and precast, are of extremely high standard throughout, and the use of variously coloured Armouralia Venetian cement-based renders, which could so easily have become an architectural cliché, creates a series of identifiable punctuation marks within the complex. The Judges extend their congratulations to the architects, contractors and all the specialist suppliers and subcontractors for the undoubted care and attention that went into the realisation of an extremely exciting project"

It should be noted this took place in 1995. Obviously the judges were taken with MJP's remarkable designs, but the citation also extends to those who executed them.

5. Prior to the issue of a partial possession certificate snagging had been carried out and lists were prepared, although it later became unclear whether they had always been attached to the relevant certificate. Difficulties ensued. Birse faced problems in putting right defects from the lists, especially since the buildings were by then occupied by students. Access depended on ET making arrangements. From December 1993 until July 1996 Birse carried out a large amount of remedial work, including (according to it) work which was not its responsibility. There was some evidence of the amounts that Birse said that it had spent, culminating in an agreement between the parties. Certain print-outs identified external costs incurred in the period after practical completion in the order of £250,000 to £350,000, together with internal costs, although since there was no proof of payment of such costs, ET made no admissions as to the works to which those costs might have related. The defects had been recorded on numerous snagging lists. They were so numerous and not well kept so that ET did not have complete sets of these documents which made it difficult for it to see what Birse ought to have done. The defects were eventually pleaded in a document termed "Schedule 4" (as it stemmed from schedule 4 to the counterclaim). It included every defect that ET contended was included in snagging lists issued to Birse in 1993 or 1996 (unless put right by Birse) plus some other later items added in later. It thus overlapped some, though not all, of schedules 1, 2 and 3, as well as claims in the main body of the counterclaim. As Miss Jackson pointed out in opening, this schedule contained a huge number of defects that ET had not yet rectified but which it said that it intended to put right. The items started as low as around £9. They included an item for cleaning mud from a door which had been left untouched since 1993 but for which ET now claimed £36.69; an item for removing rubbish found in a fire hydrant in 1996 (but not noted as a snagging item on completion in 1993) which had also been left untouched for seven years but for which ET now claimed £26.11; as well as items for the replacement of doors and frames for which the sum claimed exceeded £200,000.
6. In March 1994 an agreement was reached between ET and Birse. It is the basis of Birse's claim. The agreement was set out in a letter dated 2 March 1994 from ET to Birse. It read:

"Dear Sirs

We refer to the contract of 6 February 1992 ("the Contract") between Birse Construction Limited ("BCL") and The Eastern Telegraph Company Limited ("Eastern Telegraph"). The architect named in the Contact is MacCormac Jamieson Pritchard ("the Architect"). Words and expressions used but not defined in this letter shall have the meaning assigned to them by the Contract where the context so permits.

We set out below the terms agreed between BCL and Eastern Telegraph.

1. Eastern Telegraph shall pay to BCL the sum of £1,800,00 (one million eight hundred thousand pounds) together with VAT thereon at the appropriate rate within fourteen days of today's date subject to receipt of a satisfactory VAT invoice from BCL.

2. Eastern Telegraph shall pay to BCL the sum of £2,261.405 (two million two hundred and sixty one thousand, four hundred and five pounds) together with VAT thereon at the appropriate rate on 31 March 1994 subject to receipt of a satisfactory VAT invoice from BCL.

3.1 BCL shall make good as soon as possible the defects listed on the "snagging" sheets ("the snagging defects") attached to or referred to in any certificate or statements issued pursuant to clauses 17 or 18 of the Contract to the satisfaction of the Architect. Eastern Telegraph shall pay to BCL the sum of £470,000 (four hundred and seventy thousand pounds) within fourteen days after being notified by the Architect that all such defects have been made good to his satisfaction. If in the opinion of the Architect the snagging defects have been substantially but not fully made good by 31 March 1994 Eastern Telegraph shall pay to BCL the said sum of £470,000 less such sum as in the absolute discretion of Eastern Telegraph represents the cost of remedying any outstanding snagging defects (but for the avoidance of doubt no sum shall be deducted under this clause 3.1 if it may instead be deducted under clause 3.2 of this Agreement).

3.2 If within 14 days after receipt of a written notice from the Architect requiring compliance with the obligations set out in clause 3.1 hereof BCL does not comply therewith, then Eastern Telegraph may employ and pay other persons to make good any outstanding snagging defects and all costs incurred in connection with such making good may be deducted by Eastern Telegraph from the balance of the said sum of £470,000 retained by Eastern Telegraph or from any moneys due to BCL under this agreement.

3.3 In the event the retained sum is not adequate to pay for the costs of making good any such outstanding snagging defects the balance of these costs may be recoverable from BCL by Eastern Telegraph as a debt.

4. The Retention under the Contract shall be dealt with as follows:

4.1 On 30 September 1994, or within fourteen days after the issue by the Architect of the last (in time) certificate issued under Clause 18.1.2 of the Contract in respect of those parts of the Works which are identified in Schedule One attached, ("the Part One Works"), whichever is the later, Eastern Telegraph shall pay to BCL the sum of £145,000 (one hundred and forty five thousand pounds).

4.2 On 31 October 1994, or within fourteen days after the issue by the Architect of the last (in time) certificate issued under Clause 18.1.2 in respect of those parts of the Works which are identified in Schedule Two attached ("the Part Two Works"), whichever is the later. Eastern Telegraph shall pay to BCL the sum of £126,000 (one hundred and twenty six thousand pounds).

4.3 On 31 December 1994, or within fourteen days after the issue by the Architect of the Certificate of Completion of Making Good Defects under Clause 17.4, or within fourteen days after the issue by the Architect of the last (in time) certificate issued under Clause 18.1.2 in respect of those parts of the works which are identified in Schedule Three attached ("the Part Three Works"), whichever is the latest, Eastern Telegraph shall pay to BCL the sum of £212,275 (two hundred and twelve thousand, two hundred and seventy five pounds).

5. The Architect shall not issue any further certificates under Clause 30 of the Contract.

6. Save as the terms of the Contract are altered, added to and varied by this agreement, the Contract shall remain in full force and effect.

7. The payments previously made to BCL by Eastern Telegraph totalling £18,785.320 and the payments to be made under this agreement totalling £5,014.680, all of which payments aforesaid total £23,800.000, and the terms set out herein, are in full and final settlement of all claims which BCL has or may have now or in the future against Eastern Telegraph under or arising out of the Contract, and BCL shall have no further or additional claim against Eastern Telegraph for any sums whatsoever under or arising out of the Contract.

8. Eastern Telegraph shall if required by BCL enter into a deed with BCL and New Hampshire Insurance Company (New Hampshire) which shall provide that the amount of the bond executed by BCL and New Hampshire on 6 February 1992 shall reduce by 50%, that is to say from £1,433,065.63 to £716,532.82: provided that Eastern Telegraph shall not be obliged to enter into such deed unless New Hampshire either executes or agrees to execute (as Eastern Telegraph may require) the deed referred to in clause 11.2 of this Agreement.

9. BCL shall indemnify Eastern Telegraph against all losses, costs expenses, actions, claims or proceedings which Eastern Telegraph may incur or suffer as a result of BCL's failure to discharge any of its obligations to any of its subcontractors or otherwise.

10. Eastern Telegraph may assign the benefit of the Contract and this agreement without the consent of BCL.

11. BCL if and as so requested by Eastern Telegraph shall forthwith:

11.1 Execute and deliver to Eastern Telegraph a novation agreement between Eastern Telegraph, BCL and Cable and Wireless PLC in the form annexed to this agreement (or in such substantially similar form as Eastern Telegraph may specify);

11.2 Procure the execution by New Hampshire of a deed in such form as Eastern Telegraph may specify whereby the bond referred to in clause 8 of this Agreement shall apply to such novation agreement as it applies to the Contract;

11.3 Itself execute and deliver to Eastern Telegraph after its execution by New Hampshire the deed referred to in clause 11.2 of this Agreement; and

11.4 Procure the execution by Birse Group plc of a deed in such form as Eastern Telegraph may specify whereby the parent company guarantee executed by Birse Group plc on 6 February 1992 shall apply to such novation agreement as it applies to the Contract.

Please sign and return the enclosed copy of this letter by way of acceptance of the terms of settlement contained herein.

This letter is executed by Eastern Telegraph and is signed by [two directors][a director and the secretary] of the company."

The letter was so executed.

7. On 15 March 1995 a certificate of making good defects was issued by MJP for parts of the works. It read:

"Under the terms of the above-mentioned Contract,

We hereby certify that the defects, shrinkages and other faults specified in the schedule of defects delivered to the Contractor as an instruction have in our opinion been made good."

The parts of the works were then identified by reference to the certificates of partial possession.

8. Thereafter no further certificates were issued but on 22 April 1996 Mr Middlemore and Mr Allsop of Birse met Mr Stephen Cherry of MJP. There was a divergence of views about what happened at and the upshot of this meeting, and even as to whether Mr Cox was there (he couldn't recall) but I think that he was, as I accept Mr Allsop's account in this respect. The meeting was described, ironically, by Miss Jefford for ET, as "the high point of this case" (referring to Birse's case). However in the light of the evidence I did not understand Miss Jackson in the end to suggest that any agreement of consequence was reached at this meeting. As was only to be expected when a routine meeting is given a significance which was not appreciated by those participating in it, when there were no determinative records, and when it took place many years ago, there were differences in recollection. I do not accept that there was any acceptance as such by MJP or Mr Cox that the works were essentially complete apart from the Pozament render (which was still under investigation and not thought by MJP to be Birse's responsibility), two areas of mosaic re-tiling that Birse was willing to carry out but that could not be made available by ET, and minor items which were attended to between April and July 1996). The purpose of the meeting was to try, again, for finality. In a letter of 1 April 1996 to Mr Cherry Birse said:

"1. Completion of the Snagging Work

Together with the clerk of works we have identified those items on the snagging lists that are to be referred to you for a decision. We will contact you for a date when you can inspect the works.

We are still awaiting advice from Mr Berry of the value attributable to each of the areas."

The letter then went on to deal with specific points, including an offer of an allowance in lieu of making good the mosaics. Birse's objective was stated in the last paragraph – the release of retention. I found Mr Cherry's account entirely plausible. Mr Cherry was to "adjudicate" (as used in a later fax) in the sense of settling the differences between Mr Cox and Birse and he did so act. Not everything could be covered – hence not every building was inspected. A number of matters were for the client, ET. He told me:

"Yes, and then I had to make the decision, when we got back to the office or discussing with the client, because a lot of these remedials were not -- they were visual ones, shall we say, for want of a better expression rather than technical failing ones. For example, the leaking squash court had to be fixed by Birse even though it was not done at the time and the consequential damage by that leak had to be fixed, like the junkers flooring.

Other items like setting out of trims or cutting bits and pieces, whatever it might be, that no further work could be done or would be done; we had to decide what to do with that particular item.

So far as Birse were concerned, that is signed off by them, they did not have to do any work to it and we would have to make a decision about whether our client was taking possession of something that was not strictly in accordance with the contract."

The meeting was informal and those present evidently were still on reasonably good terms (as they had been generally). Mr Cherry however had no authority to commit either MJP to a certificate of making good defects under the agreement of March 1994. It is clear that at that time Birse certainly thought that ET held too much retention. Birse had to get a certificate; it knew that it might have to make financial allowances. There were however reports that by value at least the defects might not be great. MJP thought that ET held too much retention. Mr Prichard had told ET so in a letter of 17 January 1996 (and this remained the case as appears from the minutes of meeting of 25 June 1996). However it seems that no steps were taken by MJP to agree any list of defects to record items not to be rectified or where there was a dispute, as Mr Middlemore invited them to do, for example, in his letter of 17 May 1996 and fax to Mr Cherry of 4 June 1996, although it seems that Mr Cox inspected the Leisure Block by 25 June 1996.

9. On 1 July 1996 Birse received from MJP AI 80:

"Pursuant to clause 17.2 of the building contract, you are instructed until further notice not to make good any defects, shrinkages or other faults specified in the schedule of defects."

Mr Prichard of MJP said that the instruction was issued on the instructions of ET with advice from Northcrofts (ET's quantity surveyors) as to the best way to proceed. (By that time MJP had little to do with the project.) Although Birse in a letter of 9 July 1996 told ET that it expected a further inspection and a certificate of making good defects, neither of those steps occurred. Birse left the site. In August MJP sent ET (via Northcrofts) a long list of "latent defects" with a code which was to distinguish between various items (e.g. original snagging, items to be left, items made worse, items the responsibility of ET, items to be counter-charged, etc). Although I shall refer to these lists (the 1996 lists) they are of little value. Strangely no one from MJP (or Mr Cox) accepted responsibility for the lists. Mr Prichard was not involved as the details of the list were left to others; Mr Wisniewski had left the project; Mr Cherry wrote the letter and produced the code but he said that it was Mr Cox's list; Mr Cox said however that Mr Cherry produced the lists following an inspection on 11 July. There are also inconsistencies between the 1996 lists and other documents, such as the inclusion of defects in areas which had been the subject of the Certificate of Making Good Defects and earlier lists (not being new defects). Tracking certain defects by reference to rooms was an interesting, if unrewarding exercise. I believe that Mr Cox was the author of the 1996 lists; and their quality is because they were left to him. I agree with Miss Jackson that, in general (with exceptions) they are not an accurate (or comprehensive) account of the state of the works in the summer of 1996. They were however relied on by Northcrofts which had to use them to evaluate the defects.

10. ET moved to surer ground since Ms Elizabeth Tooth, an architect, and partner in Crocker Tooth Priestley (CTP), was then instructed by ET. She initially investigated the crazing and cracking of the Pozament render but her engagement was extended to other defects generally. She was ET's principal witness but she was not called to give expert evidence, although it had been contemplated that she would do so. In order to deal with that situation the parties agreed:

"The statements of Mrs Tooth are admitted as evidence of her contemporaneous observations, investigations and discussions and as to the opinions that informed her actions and decisions as to the appropriate exploration and remedial works, but not as opinion evidence that Birse's workmanship was defective or in breach of contract, or that MJP's design was defective or in breach of contract."

I was greatly helped by Ms Tooth's evidence and her telling descriptions of what she found. ET carried out some remedial work on her advice. Correspondence and meetings took place between ET and Birse in attempts to resolve the disputes..

11. In addition to Ms Tooth, who provided statements concerning her extensive work, ET called Mr Jonathon Priestley, another partner in CTP who had been involved in the investigations and work in 1999-2000; Mr Ronald Burns, Facilities Manager at the college who was there from July 1993; Mr David Sanderson, who inspected the Pozament render and Blanc de Bierges paving; Mr Roger Fletcher, a construction manager, who was at the college from July 1999 in connection with the remedial works (including inspection, identification and recording of defects); Mr Gordon Cox, who was Clerk of Works under the building contract from October 1991 and who continued after practical completion until June 1997 who gave evidence about the identification and recording of defects including the snagging lists and the later lists of July 1996; Mr Igors Rukuts, quantity surveyor and director of Northcroft Neighbour and Nicholson (NNN Ltd.); Mr Anthony Burgess, Planning and Project Manager of C&W since September 2000 who gave evidence about the condition of the college and its sale; Mr Stewart Pitteway, a quantity surveyor and director of Northcrofts; and, in relation to the sale: Mr Simon Marshall who was retained to handle the sale of the college and Mr Eric Bremner who was the Manager of the college (and who was to move to the new owners. Host Corporation, should the college be sold to it). Birse called the following employees or former employees: Mr Robert Allsop who was on site from July 1995 to July 1996 dealing with the rectification of defects the release of retention; Mr Ivor Roberts, a foreman on the residences who was on the site from February 1993 and who returned in 1995 to deal with snagging items; Mr Robert Taylor who was there from September 1995 dealing with snagging as it affected M&E services; Mr Christopher Barraclough who dealt, briefly, with Birse's costs; Mr Michael Greenwood, a bricklayer and foreman during the contract, and who attended inspections in 2000-2001; Mr Ralph Middlemore, Regional Commercial Manager who covered the period 1995-1996; Mr Michael Duffy, a foreman; Mr Paul Seamer, site agent from December 1992 to December 1993; Mr David Ashley, Site Manager from 1991; Mr Paddy Barber, section agent on the Teaching Wing, Leisure Block, and external works; Mr Peter Bowers who was project manager until early 1993 and thereafter prepared the final account and Mr Brian Nearney the site quantity surveyor during the contract. MJP called Mr David Prichard, one of their partners; Mr Dorian Wiszniewski, the senior project architect until he left MJP in 1994 and Mr Stephen Cherry, the associate in charge of the project. ET's experts were Dr Ian Sims (materials) and Mr Alex Fleming (building and architectural matters); Birse's experts were Dr D. S Leek and Professor John Knapton (materials) and Mr Barry Woodward (building and architectural matters). (MJP had left the case before the experts gave evidence.)
12. Some observations need to be made about the witnesses. First, not surprisingly, all (or nearly all) the witnesses found it difficult to recall the events and especially any detail. MJP's witnesses had however better recollection. There was little helpful material to assist them to remember (Birse's documents in particular were remarkably scant). This was particularly noticeable in the case of Mr Cox who clearly no longer recalled much and probably too readily agreed with suggestions. I have however to say that, making allowances, I came to the conclusion that Mr Cox, especially after 1993, did not keep proper records so that both Birse, MJP and ET knew which items of snagging had been cleared at least to his satisfaction. Thirdly, some allowance had to be made for a number of Birse's witnesses whose evidence was, understandably, not altogether objective. MJP's witnesses were balanced and informative.
13. These proceedings are therefore about defects in the construction of the college which ET claims are the liability of Birse and which either were outstanding in 1996 or have since been identified by Ms Tooth and others. ET's claims against MJP and Birse came to about £1.5 million in respect of both remedied and unremedied defects. The amount claimed against both was about £586,557 and that claimed against Birse only was about £.883,228. The defects alleged by ET are numerous. Miss Jackson, in opening Birse's case, observed, correctly, that ET had made no attempt to remedy items valued at more than £1,000,000 on the grounds that the College was to be sold. ET maintained a claim for diminution in value but it was struck out on 18 July 2003, without prejudice to ET's right to apply to reinstate it. ET tried to reintroduce such a claim by way of amendments consequential on amendments that Birse had been permitted to make but I did not allow them. Attempts to confine the trial to those that really mattered in financial terms were not successful, although defects under £1,000 were to be put to one side. Nevertheless ET required decisions on a myriad of minor items. The pivotal issues are therefore almost all factual. ET's original complaints were about the striking "wave form" and flat roofs and the insulation of about the use of a Rockwool insulation product in those roofs. When Birse was brought into the action concerning these complaints it resurrected a claim for outstanding retention which had been dormant. Quite why a contractor such as Birse did not press such a claim was never clarified but no inferences adverse to Birse are necessarily to be drawn. Birse's claim is for £422,000 (plus VAT - £73,850). This is the unpaid balance of the final account agreed under the building contract as varied by the agreement of March 1994. (Unless indicated otherwise all figures in this judgment exclude VAT which may have to be added to anything due to Birse.)
14. The arrival of Birse's claim led ET in turn to defend itself by relying on other defects which are the subject of these proceedings. Although Birse made the claim that resulted in the present trial, procedurally, ET was ordered to be the claimant on the basis of its defence and counterclaim. Birse and MJP each sought contribution from the other in Part 20 proceedings. MJP is no longer a party to any proceedings as during the trial the claims concerning it were settled. The contribution claims were discontinued. That settlement does not affect the resolution of the dispute between ET and Birse. By the time of the settlement MJP had called witnesses (but not its experts); cross-examination continued after the settlement was announced. I shall be dealing with liability and, to an extent, causation, but not the quantum of any defect for which Birse may be held liable (there are figures agreed which may be used). I shall deal with certain other issues concerning quantum arising out of the agreement of March 1994 and the law. I shall not therefore answer all 83 questions that Miss Jackson helpfully annexed to her closing submissions.
15. The settlement of the claim against MJP resulted in a payment by MJP to ET of damages of £127,000 plus a contribution of £68,000 towards ET's costs. Since ET's counterclaim was for the cost of remedying defects alleged to be the liability of Birse only, or of MJP only, or of both, an adjustment has to be made to arrive at the amounts which ET could recover from Birse. I consider that Miss Jackson's submissions on the adjustments are right. The effect of the settlement between ET and MJP is that ET's claim in respect of Pozament render must be reduced by £11,044.29 and that in respect of Armouralia by £4,654.61. (There may be other adjustments.)
16. At this point I will interpose a digression both to record that it is clear that Birse had encountered considerable problems in the running and management of the project and to mention a matter that affected ET. At the beginning of 1993 Birse brought in new management team under Mr Kevin Gander. Mr Gander wrote a memorandum dated (nominally dated 8 October 2002) addressed to Mr Jim Elders, Birse's Commercial Director, upon which ET placed reliance. Birse did not call Mr Gander so he was not cross-examined on this document, nor was it or its contents, as such, ever put to any other witness. (Some of the facts contained in it obviously were in issue and were the subject of questions of other witnesses.) There is also some uncertainty as to whether it was ever received by Mr Elders (who was not called). It is also clear that it is concerned to justify or explain matters, perhaps, as Miss Jefford suggested, to rebut criticism of a sub-contractor or sub-contractors. Whilst there is material in it apparently adverse to Birse there is material that either assists Birse or explains its position. For the purposes of the issues in this case the contents have no real probative value but, with that qualification, they do provide some indication of some of the background to the events which I have to consider. It read:

"Jim

On 6 September 1993 we convinced the client that his building was ready for occupation, which in reality it was not! Roughly 60% would be nearer the truth and we made the external works look completish. The Blanc de Bierge Subcontract was never capable of carrying out the amount of work against the original Contract programme, let alone the hard hitting finishing programme which we were working to from March 1993.

Just prior to the client occupation, we had only nearly completed Residences West, the Teaching Wings. The entire East Residences, the Leisure Centre and a vast amount of the external works were incomplete. We were in fact still plastering house 'J' and had only just started decorating in house "F". The general labourers and bricklayers were involved in the completion of the Blanc de Bierge, and the re-laying of areas which had been previously laid to give the impression of completed areas. I have attached a histogram to demonstrate the activities carried out by Parsons. In addition we received in the same period 204 C.V.I./Instructions both from M.J.P. and O.A.P. which also hampered our progress.

There are major points which must be kept in mind, especially all the work carried out after September 1993 which was in an occupied building, not an easy task at the best of times, but even more difficult at Cable & Wireless. To keep the client convinced that we were almost finished and that there was very little work left to do was in itself some achievement.

The other major factor which has seriously affected the labour level, is the amount of latent (if you can call them that?) will continue to do so for a very long time! Let's take a few examples, severe rising damp in both squash courts, this problem to this day has not been resolved, despite every effort to eradicate the problem, perhaps the only real answer is to take down the entire outer skin and rebuild the blockwork with the D.P.C's in the right place and undamaged.

The failure of the D.P.M's over the Administration and Leisure roof lights has already cost us 20k. Unfortunately we were not successful. We are now at this moment taking the entire walls down and we will re-build it properly, but who knows what else we will find once we have exposed the leaking areas.

Training duct missed out entire runs, all had to be hand dug, most of this was done out of hours so as not to alarm the college and to avoid re-instatement of large areas of landscaping works. We still have at least one to sort out.

Damp penetration of cavity walls, especially in the Leisure Block, all caused by dirty wall ties and cavities full of bricklayers crap. Not just a few mortar droppings either, some areas of cavities had up to 400mm above the cavity trays. To date we have cleaned out at least 50-60 areas, but an additional 60-70 areas have been identified in the recent driving rain. This is a nightmare as it seems this type of defect will never stop manifesting itself!

Six weeks trying to locate the leaks in the central water feature, eventually we found that the return pipe had been damaged because it had never been surrounded in concrete. The main cause of leakage was "honeycombing" in the concrete structure, especially behind the light fittings. We eventually had to divide the water course into sections of 5-6 metres, so that we could pin point the problem areas. In addition to carrying out these works, 90% of the Blanc de Bierges had to be removed.

Drainage that did not work properly when in full use because it had been put in at the wrong levels, we adjusted the easy runs in an attempt to improve the flows, but this will always be a worry I'm afraid. It would have necessary to tunnel under the Administration building to put things right and break out floor slabs and foundations in both East and West Teaching Wings. Some nights I don't sleep so well!

The list of defective works is almost endless. The ones that worry me the most are the one's we do not know about yet! I hope that I am not around when the really bad one's come to the surface. I can only tell lies to the client for so long.

We have also carried out work because either the Subcontractors won't come back, or been thrown off the site or gone bankrupt: i.e.

CARPENTRY - The original Subcontractor, Stotts, were thrown off the Contract and replaced by a carpentry Subcontractor called Whitlock. When Whitlock had completed his main Contract works, the question was of who would carry out the snagging works on the work which Stotts had carried out, Parson's rates were more economical and that's the reason why we used them. The extent of snagging/defect work that was relating to Stotts was huge and also had to be carried out in a fully occupied building, we all know that this is never an easy thing to do.

DECORATIONS Much the same story as the carpentry, it was more economical to use Parson's to repair works and to back charge others. Damage was caused by all the sundry especially B4 during their fit-out works, which should have been done after handover from us, but due to the revised finishing programme and the occupation required date, we had no choice but to overlap our activities.

Generally we have supplemented labour across a broad spectrum of operations where Subcontractors have defaulted. When this has occurred, all the relevant information has been passed to the Surveyors for recovery.

We have tried our very best to maintain a level of production, so as to maintain the confidence of the client, which by and large has worked. Unfortunately this will not work much longer, we are still having great difficulty in getting the remaining Subcontractors to perform and to complete their snagging works. I'm afraid it is indicative of this Contract and Contracts of a similar nature. They are so fed up with the way they have been made to work and the way they have been treated that any measure of goodwill is just not there any more.

As from a personal point of view Cable & Wireless has had an almost devastating effect upon myself and my family, and that is going to take a great deal of time and patience to heal.

I feel very proud of the great efforts put into Cable & Wireless by both the site team and the Surveyors and indeed the end result. I have heard the innuendos levelled at myself and some of the site team, especially in relationship to Parsons.

I was asked to come to Cable & Wireless to sort it out, I did this with very little help. I was expected to know all the answers, but unfortunately I did not know all the answers, nor did anybody else. If it was not for the efforts of a very few we would never have made it i.e. Paul Seamer, Paddy Barber, Neil Wilson, Neil Thompson and last but not least Richard Green."

17. ET's claims are for the costs incurred in remedying defects that it has put right and for the estimated costs of remedying defects that it has not yet put right. The latter include defects which ET does not intend to put right but in the main ET says that it intends (or intended) to put them right. It should be noted that after the draft judgment was sent to the parties' representatives, the parties settled the proceedings. This judgment has certain main sections (leaving aside any sub-sections):

1. The Contracts (para 15)

2. The Agreement of March 1994 (para 22)

3. Recoverability for Unremedied Defects (para 37)

4. Remedied Defects (para 56)

A. Pozament Render (para 57)

B. Armouralia (para 92)

C. Blanc de Bierges (para 100)

D. External Walling and Copings (para 126)

E. Ceilings (para 162)

F. Curtain Walling, Windows, Doors and Glass (para 178)

G. Felt Roof Work, Lead Flashings and Zinc Roofing and Flashings (para 179)

H. Paintwork to External Metalwork (para 180)

I. Light Fittings and Other Electrical Items (para 183)

J. Miscellaneous Remedied Defects (para 186)

K. Lake Aeration System (para 187)

L. Cost of Clerk of Works (para 193)

5. Unremedied Defects (para 196)

A. Mosaic Tiling (para 199)

B. Joinery (para 211)

6. Certificate of Making Good Defects (para 218)

7. Damages (para 219)

8. Conclusion (para 220)

1. The Contracts

18. The contract between ET and Birse was made under seal and dated 6 February 1992. It incorporated the JCT Standard Form of Building Contract 1980 edition with Quantities together with the Contractor's Designed Portion Supplement. MJP were the Architect under the contract. The Contract Sum was £14,330,656.30.
19. In the event the total of the agreed final account was £23,800,000.00; I am not concerned with why there was such an increase of 66%. The Date for Completion was 31 December 1992. The contract between ET and MJP was also made under seal. It was dated 9 February 1992. In so far as it relevant, clause 4.2 of MJP's contract restated the obligations of an architect:

"4.2 The Consultant shall:

4.2.1 in complying with all his obligations under this Agreement exercise the reasonable skill, care, attention and diligence to be expected of a properly qualified and competent architect experienced in performing services in respect of works of a similar size, scope and complexity to the Project".

20. The contract between ET and Birse included in its Conditions:

"8. Work, materials and goods

8.1.1 All materials and goods for the Works other than the part thereof comprised in the Contractor's Designed Portion shall, so far as procurable, be of the kinds and standards described in the Contract Bills provided that materials and goods shall be to the reasonable satisfaction of the Architect where and to the extent that this is required in accordance with clause 2.1.4. All materials and goods for the Contractor's Designed Portion shall, so far as procurable, be of the kinds and standards described in the Employer's Requirements or, if not there specifically described, the Contractor's Proposals or documents referred to in clause 25.1.1 provided that the Contractor shall not substitute anything without the Architect's consent in writing, which consent shall not be unreasonably delayed or withheld, but no such consent shall relieve the Contractor of his other obligations.

8.1.2 All workmanship for the Works other than the part thereof comprised in the Contractor's Designed Portion shall be of the standards described in the Contract Bills, or, to the extent that no such standards are described in the Contract Bills, shall be of a standard appropriate to the Works, provided that workmanship shall be to the reasonable satisfaction of the Architect where and to the extent that this is required in accordance with clause 2.1.4. All workmanship for the Contractor's Designed Portion shall be of the standards described in the Employer's Requirements, or, if not there specifically described, the Contractor's Proposals but to the extent that no such standards are described in the Employer's Requirements or in the Contractor's Proposals shall be of a standard appropriate to the Contractor's Designed Portion.

8.1.3 All work shall be carried out in a proper and workmanlike manner.

8.2.1 The Contractor shall upon the request of the Architect provide him with vouchers to prove that the materials and goods comply with clause 8.1.

8.2.2 In respect of any materials, goods or workmanship, as comprised in executed work, which are to be to the reasonable satisfaction of the Architect in accordance with clause 2.1, the Architect shall express any dissatisfaction within a reasonable time from the execution of the unsatisfactory work.

8.3 The Architect may issue instructions requiring the Contractor to open up for inspection any work covered up or to arrange for or carry out any test of any materials or goods (whether or not already incorporated in the Works) or of any executed work, and the cost of such opening up or testing (together with the cost of making good in consequence thereof) shall be added to the Contract Sum unless provided for in the Contract Bills or unless the inspection or test shows that the materials, goods or work are not in accordance with this Contract.

8.4 If any work, materials or goods are not in accordance with this Contract the Architect without prejudice to the generality of his powers, may:

8.4.1 issue instructions in regard to the removal from the site of all or any of such work, materials or goods; and/or

8.4.2 after consultation with the Contractor (who shall immediately consult with any relevant Nominated Sub-Contractor) and with the agreement of the Employer, allow all or any of such work, materials or goods (except those which are part of the Contractor's Designed Portion) to remain and confirm this in writing to the Contractor (which shall not be construed as a Variation) and where so allowed and confirmed an appropriate deduction shall be made in the adjustment of the Contract Sum; and/or

8.4.3 after consultation with the Contractor (who shall immediately consult with any relevant Nominated Sub-contractor) issue such instructions requiring a Variation as are reasonably necessary as a consequence of such an instruction under clause 8.4.1 or such confirmation under clause 8.4.2 and to the extent that such instructions are so necessary and notwithstanding clauses 13.4, 25 and 26 no addition to the Contract Sum shall be made and no extension of time shall be given; and/or

8.4.4 having had due regard to the Code of Practice appended to these Conditions (following clause 41), issue such instructions under clause 8.3 to open up for inspection or to test as are reasonable in all the circumstances to establish to the reasonable satisfaction of the Architect the likelihood or extent, as appropriate to the circumstances, of any further similar non-compliance. To the extent that such instructions are so reasonable, whatever the results of the opening up for inspection or test, and notwithstanding clauses 8.3 and 26 no addition to the Contract Sum shall be made. Clause 25.4.5.2 shall apply unless as stated therein the inspection or test showed that the work, materials or goods were not in accordance with this Contract.

8.5 Where there is any failure to comply with clause 8.1.3 in regard to the carrying out of the work in a proper and workmanlike manner the Architect, without prejudice to the generality of his powers, may, after consultation with the Contractor (who shall immediately consult with any relevant Nominated Sub-Contractor), issue such instructions whether requiring a Variation or otherwise as are reasonably necessary as a consequence thereof. To the extent that such instructions are so necessary and notwithstanding clauses 13.4 and 25 and 26 no addition to the Contract Sum shall be made and no extension of time shall be given in respect of compliance by the Contractor with such instruction.

8.6 The Architect may (but not unreasonably or vexatiously) issue instructions requiring the exclusion from the Works of any person employed thereon.

…..

10. Person-in-charge

The Contractor shall constantly keep upon the site a competent person-in-charge and any instructions given to him by the Architect or directions given to him by the clerk of works in accordance with clause 12 shall be deemed to have been issued to the Contractor.

11. Access for Architect to the Works

The Architect and his representatives shall at all reasonable times have access to the Works and to the workshops or other places of the Contractor where work is being prepared for this Contract, and then work is to be so prepared in workshops or other places of a Domestic Sub-Contractor or a Nominated Sub-Contractor the Contractor shall by a term in the sub-contract so far as possible secure a similar right of access to those workshops or places for the Architect and his representatives and shall do all things necessary to make such right effective. Access in accordance with clause 11 may be subject to such reasonable restrictions of the Contractor or any Domestic Sub-Contractor or any Nominated Sub-Contractor as are necessary to protect any proprietary right of the Contractor of any Domestic or Nominated Sub-Contractor in the work referred to in clause 11.

12. Clerk of works

The Employer shall be entitled to appoint a clerk of works whose duty shall be to act solely as inspector on behalf of the Employer under the directions of the Architect and the Contractor shall afford every reasonable facility for the performance of that duty. If any direction is given to the Contractor by the clerk of works the same shall be of no effect unless given in regard to a matter in respect of which the Architect is expressly empowered by the Conditions to issue instructions and unless confirmed in writing by the

…….

17. Practical Completion and Defects Liability

17.1 When in the opinion of the Architect Practical Completion of the Works is achieved, he shall forthwith issue a certificate to that effect and Practical Completion of the Works shall be deemed for all the purposes of this Contract to have taken place on the day named in such certificate.

17.2 Any defects, shrinkages or other faults which shall appear within the Defects Liability Period and which are due to materials or workmanship not in accordance with this Contract or to frost occurring before Practical Completion of the Works or other failure of the Contractor to comply with his obligations under this contract in respect of the Contractor's Designed Portion, shall be specified by the Architect in a schedule of defects which he shall deliver to the Contractor as an instruction of the Architect not later than 14 days after the expiration of the said Defects Liability Period, and within a reasonable time after receipt of such schedule the defects, shrinkages, and other faults therein specified shall be made good by the Contractor at no cost to the Employer unless the Architect with the consent of the Employer shall otherwise instruct; and if the Architect does so otherwise instruct then an appropriate deduction in respect of any such defects, shrinkages or other faults not made good shall be made from the Contract Sum.

17.3 Notwithstanding clause 17.2 the Architect may whenever he considers it necessary so to do, issue instructions requiring any defect, shrinkage or other fault which shall appear within the Defects Liability Period and which is due to materials or workmanship not in accordance with this Contract or to frost occurring before Practical Completion of the Works or other failure of the Contractor to comply with his obligations under this Contract in respect of the Contractor's Designed Portion, to be made good, and the Contractor shall within a reasonable time after receipt of such instructions comply with the same and at no cost to the Employer unless the Architect or with the consent of the Employer shall otherwise instruct; and if the Architect does so otherwise instruct then an appropriate deduction in respect of any such defects, shrinkages or other faults not made good shall be made from the Contract Sum. Provided that no such instructions shall be issued after delivery of a schedule of defects or after 14 days from the expiration of the Defects Liability Period.

17.4 When in the opinion of the Architect any defects, shrinkages or other faults which he may have required to be made good under clauses 17.2 and 17.3 shall have been made good he shall issue a certificate to that effect, and completion of making good defects shall be deemed for all the purposes of this Contract to have taken place on the day named in such certificate (the "Certificate of Completion of Making Good Defects").

17.5 In no case shall the Contractor be required to make good at his own cost any damage by frost which may appear after Practical Completion, unless the Architect shall certify that such damage is due to injury which took place before Practical Completion.

17.6 The Architect shall not be obliged to issue a certificate of Practical Completion of the Works until the Contractor has supplied all operating manuals, warranties, as-built drawings, test certificates and other documents specified in the Contract Documents relating to the Works.

18. Partial possession by Employer

18.1 If at any time or times before the date of issue by the Architect of the certificate of Practical Completion the Employer wishes to take possession of any part or parts of the Works and the consent of the Contractor (which consent shall not be unreasonably withheld) has been obtained, then notwithstanding anything expressed or implied elsewhere in this Contract, the Employer may take possession thereof. The Architect shall thereupon issue to the Contractor on behalf of the Employer a written statement identifying the part or parts of the Works taken into possession and giving the date when the Employer took possession (in clauses 18, 20.3, 22.3.1 and 22C.1 referred to as "the relevant part" and "the relevant date" respectively).

18.1.1 For the purpose of clauses 17.2, 17.3, 17.5 and 30.4.1.2 Practical Completion of the relevant part shall be deemed to have occurred and the Defects Liability Period in respect of the relevant part shall be deemed to have commenced on the relevant date.

18.1.2 When in the opinion of the Architect any defects, shrinkages or other faults in the relevant part which he may have required to be made good under clause 17.2 or clause 17.3 shall have been made good he shall issue a certificate to that effect.

Other relevant provisions are also the same as or similar to the JCT conditions, such as clauses 2, 30 and 41 (and Article 5) and therefore need not be set out.

2. Agreement of March 1994

21. The agreement of March 1994 represented a compromise of the final account and established a final account sum and a mechanism for payment of the balance outstanding. The contract sum had been £14,330,656.30. The total of the agreed final account was £23,800,000.00; I am not concerned with why there was such an increase of 66%. (All figures in this judgment exclude VAT.) The agreement also effected a compromise of the claims by both parties for delay, e.g. ET's claims for liquidated damages, given that Birse had only an extension of time to April 1993. I am also not concerned with why there was delay or with whose responsibility it was. I have referred to Birse's problems. I deal here with points on the interpretation and effect of the agreement. The agreement also provided for the reduction of the amount of the bond from 10% to 5% of the original contract sum, i.e. to £716,532, itself still a large sum and significant in considering the terms of the agreement. Thus agreement of the final account resulted in £5,014,680 that might yet be paid to Birse. Some of that was to be paid: £1,800,000 by 16 March 1994 (clause 1) and £2,261,405 by 31 March 1994 (clause 2). Those payments were made. Payment of the remainder was the subject of clauses 3 and 4.

Clause 3.1

22. Birse claim £13,725 under clause 3.1 of the agreement since in 1994 ET had paid all but that amount of the £470,000 referred to in that clause. ET said that no more is due since (1) Birse did not fulfil its obligations to make good all "snagging defects" as required by clause 3.1; (2) ET is entitled to withhold against £13,725 "such sum as in the absolute discretion of Eastern Telegraph represents the cost of remedying any outstanding snagging defects". ET says that it is irrelevant for these purposes whether ET intends to remedy those defects. Thus if ET has any unremedied "snagging defects" which Birse failed to remedy, the cost of remedying which, in ET's absolute discretion, would exceed £13,725 then ET is entitled to withhold that amount. Miss Jackson conceded that the reference to ET's absolute discretion enabled ET to set off against that £13,725 its estimate of the cost of putting right unrectified defects under clause 3.1. The £13,725 is therefore available against which to set off the claim in Schedule 4 in respect of any item which was listed in a 1993 snagging sheet, is a defect (i.e. a breach by Birse) and has not been remedied. Miss Jackson attached to Birse's closing submissions a list of admissions in respect of defects, the value of which admitted defects (even at 1994 values) was enough for ET to set off against the balance of £13,725.
23. However Birse said that the right of set-off under clause 3.1 only applied before remedial works were actually carried out, when clause 3.2 applied. In addition ET could not set off more than the balance unpaid. It could not attempt to recoup what had been paid on the basis that it had assessed a value greater than £13,725 in respect of works not yet put right. Furthermore there was no means whereby ET could withhold retention monies due under clause 4 under its rights under clause 3.1.
24. Miss Jefford maintained that if an item were a "snagging defect" as defined (i.e. "the defects listed on the "snagging" sheets"), Birse was liable to remedy it and, until the work was done, ET was entitled to withhold the cost of doing so and did not have to prove that the "defect" was the result of a breach of contract on the part of Birse. Miss Jefford also submitted that ET was also entitled to rely on clause 3.3 to recover by counterclaim the sums that it estimates (or that were agreed) to be the cost of the remedying unremedied "snagging defects" as clause 3.3 provided that if the sum retained under clause 3.1 was not adequate to pay for the cost of making good such outstanding snagging defects, the balance of these costs was recoverable from Birse by ET as a debt. The parties thus contemplated that if Birse did not get on with the snagging the amount retained under clause 3.3 might not be adequate. Clause 3.3 protected ET (and gave Birse an incentive to get on) by giving ET a right to recover the balance as a debt.
25. Miss Jackson submitted that the definition of snagging defects did not require Birse to rectify or to be liable for not rectifying something for which it was not liable under the building contract. Clause 6 said that terms of the building contract remained in effect and there was nothing in the agreement which altered or varied them. There was also a clear distinction between clauses 3.1 and 3.2. The agreement had been arranged to provide ET with security for the estimated costs of work to be done (should it not be done by Birse) but if ET had to do the work and spend money then it was limited to the amount actually retained under clause 3.1. Clause 3.3 was linked to clause 3.2 and required actual expenditure before there was a liability. Miss Jackson elicited evidence, notably from Mr Cherry that when the lists were drawn up he did not necessarily expect all the items to be made good (as Birse might jib at them). It was therefore submitted that the lists could not be treated under the agreement as recording matters for which Birse was liable. However that evidence is obviously to be disregarded in considering the meaning of the written agreement.
26. I consider that a pragmatic approach must be applied to the settlement agreement as it was intended to bring the discussions to an end and not to perpetuate argument. It is not necessary to rehearse the evidence about how it came to be made. It is easy to see from reading it that the parties did not intend that in order to make it work it would or might be necessary to go back to the building contract and get an answer to a question arising under that contract. Thus I have no doubt whatsoever, first, that the terms of the building contract were substantially altered, added to and varied and that no reliance can really be placed on clause 6; and, secondly, that clause 3.1 means that Birse agreed to put right the snagging defects as defined whether or not it was actually liable under the building contract. The defects clearly include items which were almost certainly not Birse's liability but which a contractor in the position of Birse would not jib at making good. They include items which Birse would ordinarily have been obliged to put right before practical completion, such as cleaning up (i.e. not necessarily bad work as such but performance of an outstanding obligation), but, in the circumstances in which ET needed to take possession, which could be left to be dealt with later. Thus ET was right to give as an instance the cutlery server (Schedule 4, item 2-11) which was installed in three pieces. It was still perfectly serviceable, as Mr Fleming agreed but as it was listed as a "snagging defect". It may then have suited Birse not to do such work and a responsible contractor (as Birse professed to be) would recognise that it could not take advantage of the certificate which enabled ET to regain possession. In addition the interpretation suggested for Birse would effectively nullify the "absolute discretion" given to ET by clause 3.1 if ET had first to be sure that the cost which it wished to withhold was in respect of a defect that was either admitted or undoubted. The snagging sheets resolved that. In my judgment, by assenting to the definition of the defects under the agreement, Birse either admitted liability or waived its right to contest liability. It is therefore not open to Birse to assert that it had no liability under the building contract for its terms have been altered to preclude that being done, e.g. by way of challenging MJP's opinion and seeking to have an opinion or certificate about such a defect or which would be consequential on or connected to a decision as to whether an item of work was or was not defective (including incomplete or omitted work). Accordingly I consider that ET is right to describe such defects as agreed or admitted by Birse and that such phrases (which I shall also use in the same way) do not signify only that Birse had in 1994 agreed that the item had been listed.
27. I also reject the submission that ET could use clause 3.3 to recover from Birse the estimated costs of putting right defects to the extent that such costs were greater than the amount retained. Clause 3 is arranged in a logical and progressive order. It starts with the position where defects (as defined) had not been made good, two estimates of the possible costs of making them good are made. The first is the amount of £470,000. That was arrived at by ET with the assistance of Northcrofts as part of the total outstanding which was not to be paid immediately (as Miss Jefford accepted and as Mr Rukuts said), so it has to be treated as what ET regarded as sufficient should Birse not comply with its obligations. It is self-evidently a considerable sum. (If evidence were needed Mr Rukuts so described it.) It has also to be remembered that ET had the benefit of a bond which even though reduced in value under was still worth nearly £750,000 and that clause 3 did not deal with retention (to which I shall come). ET had therefore two other means of protection. This therefore justifies the declaration: "(but for the avoidance of doubt no sum shall be deducted under this clause 3.1 if it may instead be deducted under clause 3.2 of this Agreement)".
28. Secondly, should Birse not do so (in the opinion of MJP), ET had then effectively to decide how much to decide to withhold. Next, if ET were obliged to step in to put right defects then it could use whatever it had retained which might be less than £470,000, depending on what had happened after 31 March 1994. Thirdly, in such event, if ET found that it did not have enough in hand by way of the retained sum (or what was left of it) then Birse would be liable to pay the shortfall: the term "the balance of these costs" was used. ET is therefore only entitled to recover from Birse any excess over £13,725 which it has in fact incurred in the exercising its rights under clause 3.2. Clearly if ET had released part of the retained sum of £470,000 (as it has done) then ET will in effect be clawing back what it paid.
29. The main difference between ET's rights under clause 3.3 and its general rights under the building contract is that Birse's liability is expressed as a debt and it stems from its liability in respect of "snagging defects". Thus ET does not have to prove Birse's liability under the building contract. The written notice from MJP is of course subject to challenge by Birse as is ET's quantification of its recoverable costs since the agreement varies the building contract and thus any dispute about such matters would be the subject of clause 41. In the circumstances this is of little practical consequence. Birse take no point on the absence of a written notice from MJP.
30. However it is clear that when ET came to release £456,275 of the £470,000 (which it did in instalments of £435,000 on 25 April 1994, £8,130 on 30 August 1994 and a further £13,145 on 12 September 1994), it did so on the advice of MJP and Northcrofts (as Mr Cherry said) and in full knowledge of the facts. For example 13 April 1994 Mr Cherry wrote to ET:

"On Tuesday 13 April with Gordon Cox (CofW) and John Berry (NNN) we made a formal assessment of the outstanding works and progress of snag items. Although BCL have not finished, we were pleased at the progress made over the last month and their renewed determination to complete the works to everyone's satisfactory.

A summary of the current situation is as follows:

Snagging

Externals: Teaching, Administration and Residential buildings are approximately 95% complete. Leisure is slightly less but has not been officially desnagged. Most of the outstanding items are minor. A recurring snag is marks (pencil, bitumen, red wax, stains etc.) on the concrete. Also approximately 250 roof tiles require replacement as they have been unacceptably touched-up on the exposed face.

Internals: All buildings are over 90% cleared. Some of the outstanding items will be done at the end of the defects period, whilst other are numerous tidying-up of rebates, paint and junctions.

External Works: This has not been desnagged, and areas are still being worked on by BCL. Also BCL are refusing to complete a number of items. Brophy's also have some outstanding work to completed.

Remedial Works

Curtain Walling: 95% of general items are cleared. Outstanding items are still glass, skeg pins, bar door stops and work to LD20 and 21. A more difficult issue is the set of as constructed drawings, we will discuss a sensible resolution of this with BCL which may involve further time by ourselves at BCL's cost.

….."

Similarly on 5 July 1994 Mr Cherry provided a two page list of outstanding items and concluded by saying: "I trust this is helpful and you will agree further payment on this basis with Birse. We will write separately to Birse regarding outstanding works". In his letter to Birse (of 12 July 1994) Mr Cherry said:

"We attach our reply dated 5 July 1994 to your valuation submission sent to Northcrofts last week. As you can see we do not agree with all of your percentage completions. We are also acutely aware that completion of these and other outstanding items is painfully slow. At the current clear-up rate, release of all retention monies is unlikely with yet further site meetings from which, in previous experience, very little progress will be made.

A selection of items (not a full list) that must be addressed are as follows:

Test certificate for emergency lighting and smoke detection – this has been outstanding for a year and is preventing the issue of a fire certificate for the Administration building.

Teaching wing finials not fixed – remember the recent storms.

Electrical as built drawings complete with drawing list.

West courtyard drainage.

Curtain walling – especially missing and incorrect face caps, plus numerous items on the snag sheets. Recent inspection has revealed that only 10-15% of snag items have been completed since our last inspection in January 1994.

Br screen doors and LD20/21 remedials not carried out.

Mast still not repainted.

Squash court is still leaking.

Pozament remedials poor in gardens compound.

Gardener's compound not complete.

Mosaic – no work done.

Front of College reinstatement to road not started – this work has been outstanding since December 1993.

Grit blasting of new blockwork to administration and leisure walls as agreed. Indeed are the blocks the correct mix?

Slate stepping stone not fixed – propped on wood!

VIP WC floor still not replaced.

Boardroom bridge maple not repaired.

Residential stairs – no work commenced.

Residential doors – none replaced.

CofW directions – still some outstanding.

Insulation certificates not returned.

Still numerous snag items outstanding which will be added to the defects schedule.

Numerous maintenance sheets required (separate list forthcoming).

Your urgent attention is required to all these and other items."

31. Accordingly by the time that ET came to make these payments it must have formed a view about the defects which it regarded as outstanding for the purposes of the agreement of March 1994 (which includes therefore all known and notified defects under the building contract since the definition of defects under the 1994 agreement was wider than the defects under the contract). Although the word "withhold" does not appear in clause 3, I accept Miss Jackson's submission that the wording and structure of clause leads to the conclusion that it was not intended to be operated in respect of defects for which some part of the £470,000 might have been (but was not) in effect withheld by ET using its rights under clause 3.1. In my judgment the March 1994 agreement was intended to bring finality to the snagging defects then listed whereby ET's rights under the building contract in respect of them were replaced by a new code.

Clause 4

32. Birse claim £408,275 under clauses 4.1, 4.2 and 4.3. It is the major part of its claim. ET's defence is that until a certificate of making good defects has been made good the claim must fail. ET submitted that clause 4 was clear. It provided for the payment of retention on the issue of the last of the certificates of making good defects that relate to the three parts of the works for which certificates of partial possession had been given. Nor does Birse become entitled to the release of any retention upon the issue of a certificate of making good defects for a part of the Works (for which there was a certificate of partial possession). The terms were expressly that the total sum becomes due and payable on the issue of the certificate that is "last in time". ET accepted that it might be decided that the relevant certificates ought to have been issued but Birse had to satisfy me of that. It was not for ET to show that Birse was not entitled to these certificates. No certificates should be issued (the extent of defects remaining in 1996 demonstrated that MJP's certificate of making good defects should be revised). Furthermore clause 4 did not make retaining money dependent on the employer carrying out or having an intention to carry out the same.
33. For Birse Miss Jackson submitted that the proposition that the absence of certificate of making good defects disentitled Birse to any release of retention for that tranche, was contrary to construction practice and the operation of the JCT conditions. Retention was money held back and the primary entitlement of the contractor to receive payment of the contract sum, as adjusted and less any recoverable damages, remained. ET's argument made the holding of retention under the agreement not just security and leverage but a forfeiture provision since the Birse would not get the money retained if there were slight defects not made good that led to the certificate of making good not being issued. In addition ET's stance was unjustified. At the end of the defects liability period MJP had listed the defects including any outstanding snagging items. A certificate of making good defects was issued on 15 March 1995. It covered five of the eight areas within Schedule One (the Part One Works) and two of the six areas in Schedule Two. MJP was thus satisfied that snags and defects in the relevant blocks and areas had been made good. ET paid £75,000 of the retention, leaving £408,275. Birse claimed a certificate of making good defects should have been issued for all of the work. Although the certificate of 15 March 1995 could be revised, Birse argued that this was not necessary, since Birse was entitled to the retention monies subject only to ET's right set off any damages suffered by ET as a result of any breaches by Birse of the building contract, such as clause 17.2. (This contention leads to a possible issue about what ET or Birse has to prove.)
34. In my judgment the agreement basically restated the position under the building contract. Retention is money earned but not payable. It is a proportion of the value of the work done (in the opinion of the architect) as certified in interim certificates under clause 30. It is held back as security for the due performance of the contractor's obligations in case the contractor should default and the employer suffer loss or damage. Since its release is linked to the architect being satisfied with the quality of the works the employer's rights of recourse are commonly thought to be limited to claims in respect of defective work. That is a misconception, since, subject to the terms of the contract, retention is security for default in any respect (damages for delay, claims by third parties met by the employer for which the contractor is to indemnify the employer etc.). It is however clear (and well established) that payment of retention held after practical completion under a building contact incorporating the JCT conditions is conditional on the contractor obtaining a certificate of making good defects. A final certificate cannot be issued until such a certificate has been issued (see clause 30.8). In my judgment that is clearly so under clause 4. No further certificate was to be issued under clause 30 of the building contract (see clause 5 of the agreement). The final account had been agreed. Thus the policy of clause 30.8 had to be implemented in a different way: the money retained by virtue of clause 4 would be held as security (additional to that under clause 3 and the bond) until Birse obtained the relevant certificate for the relevant Part (which corresponded to the areas the subject of partial possession certificates). Thus I reject Birse's starting point.
35. However it does not follow that Birse cannot be paid the retention or any of it. Clause 4 is not (or need not be) in the nature of a forfeiture provision. First, as is accepted, under clause 41.4 a certificate may be issued (although, for reasons which will become apparent, I shall not do so). Secondly, if Birse's primary obligation to make good defects has not been fully performed, then, as a matter of the application of general law, that obligation is replaced by an obligation to pay damages in lieu. Accordingly, under clause 41 or otherwise, an award or decision will assess those damages and, if they are less than the amount of retention claimed, release the balance. I deal later with the facts. Thirdly, and in any event, the effect of the steps taken by ET and MJP in 1996 in issuing AI 80 whereby Birse was instructed under clause 17.2 "not to make good any defects, shrinkages or other faults specified in the schedule of defects" put it out of Birse's power to comply with its obligations and thereby to obtain the requisite certificates under clause 4. ET is no longer able to rely on the requirement to obtain a certificate under clause 4. Fourthly, ET's action would not assist Birse if no certificate would have issued had AI 80 not been issued, so ET would have to have had good reason not to complete the works to a satisfactory state and, and as will appear, no such reason exists. Thus AI 80 and ET's subsequent inaction have the effect of requiring ET to pay the retention less that to which it is entitled for any breach by Birse of its contractual obligations, by way of abatement, set off or counterclaim (the method is immaterial as such niceties and their discussion is and has for some time been distinctly passé). Accordingly ET is liable to pay the outstanding retention save to the extent that it has a valid ground not to do so, and thus according to it, to recover the balance of all or part of the £865,000 that it claims. Miss Jackson was essentially right in her submission that that it would be absurd and contrary to practice – I would say untenable in law as practice is irrelevant - if the agreement of March 1994 resulted in Birse being both being liable to pay ET up to £1.5 million and not being able to recover or offset any of the money that would have been due under the agreement to it had there been no outstanding defects.

3. Recoverability for Unremedied Defects

36. As I have already said, ET intend to sell the college. This section considers what damages ET may recover for defects which it has not rectified (on the assumption that Birse is liable), in the light of ET's intention and other facts and factors. Mr Bremner confirmed that the decision to sell had been taken as long ago as September 2000. He said:

"As I say, the first thing I knew about the college being sold was when a fellow called Ben Milner arrived from head office in September and told two of us that we are selling the college in November. No doubt somebody in London somewhere, it was not me, who had discussions with [GVA] Grimley, who represent Cable & Wireless in a broad range of property matters…."

ET was then advised by GVA Grimley that as the college was "a unique, high-quality specialist property" it was difficult to value. A range of £9m to £16.5m was suggested. At some time in 2003 a sale was agreed. I have no doubt that this would have been advantageous to ET. Mr Burns told me that at present the college was only used 10% of the time by Cable & Wireless and that 90% of the bookings were by outside organisations. Since the conclusion of the hearing the sale then examined has fallen through, as appears from the occasional reports received on the progress of the sale and ET's intentions as part of ET being required to comply with its obligations as to discovery. It was not at all clear to me from the reports whether ET then irrevocably abandoned the proposed sale and set about remedial work or whether it still wished to sell the college. I have to decide the issues on the basis of the evidence that I received at the trial.

37. In a witness statement of 15 May 2003 Mr Burgess had said that there was interest in acquiring the college, and that discussions were at an advanced stage for its possible sale. He said that therefore remedial works had "once again" been put on hold (with certain "run of the mill" defects being attended to in the normal way). He said that the court would be kept fully informed of what eventually transpired. In evidence Mr Marshall agreed that by that time Mr Richard Nichol of Host Corporation had been given the status of a preferred bidder that a price had been established. (Mr Marshall had been retained by Cable & Wireless in August 2003 solely to manage the sale and was to leave on its completion.) Moreover Mr Marshall said that ET would not deal with other bidders. Host was buying the college as a going concern so the price offered included factors other than the value of the building, as Mr Marshall agreed. It seemed to me from the documents and the evidence of the discussions that the buyer always intended to refurbish the college and had made an offer and negotiated without assessing the cost of putting right any defects, other than, perhaps, the roof. It is neither necessary nor desirable in ET's commercial interests to mention the price and I shall not do so, except to say that it was at a very substantial discount to the final account.
38. Despite the decision having been taken in 2000, ET did not give discovery of the documents relating to its intentions until Birse had applied for specific discovery (as Birse had learned of them from another source). That application led to a (third) witness statement of Mr Burgess and a statement of Mr Blaza (ET's conveyancing solicitor) from which Birse then learned of the preferred bidder and of the negotiations then underway with a view to exchange of contracts, then forecast for the autumn. (ET had still to be told at the outset of the trial to keep disclosing documents relating to the sale.)
39. Alternative prices had originally been sought depending on whether ET did or did not remedy the roof. There is nothing in documents disclosed which dealt either with a reduction in price to reflect defects or with ET's professed intention to remedy defects. Mr Bremner described in detail how Mr Nichol was taken round the building and given lists of defects by ET (but not until October 2003), the discussions about what might need to be done but he could not say what effect there was, if any, on the price. The list of defects contained the pleaded costs rather than the lower costs agreed by the quantum experts in August 2003. Mr Marshall was in a similar position. He had however procured a letter from Mr Nichol on 8 October 2003 in which Mr Nichol recorded, amongst other things:

"I am now in receipt of the detailed schedule of remedial issues relating to the above-mentioned property. Whilst it would have been helpful to have received the details in this form earlier in our discussions I do understand the reasons. They have been discussed and investigated in some depth and we have agreed to transact at the agreed figure and undertake remedial works at our own expense. It occurs to me that there will be reports and surveys."

Mr Nichol was not called although Mr Marshall said he was available. There was no evidence of any reduction in any offer on account of any survey.

40. In addition Mr Burns and Mr Burgess said that no outside organisation was apparently deterred from booking the college by the state of the buildings in their unremedied condition, although Mr Burgess thought that some organisations might think that the college was not up to the highest standard. Mr Bremner was quite firm that, apart from water penetration and the like, everybody as of the view that the other defects had no material consequence on the use of the building. Certainly it is clear that ET did not think it necessary to carry out work to the rooms in order to ensure that the college was up to the standard expected of it.
41. Birse thus challenged ET's professed intention to carry out remedial works. The defects in question are marked with an X in Schedule 4 in the "to be rectified" column. They include defects which appeared in snagging lists prepared in 1996 and 1996 alone, those which were not in any lists but have been inserted later. The defects include some which must have existed earlier but were never listed. That casts doubt on whether they were truly regarded as defects or ought to be so regarded. I deal with certain classes later. Birse admits responsibility for some of these defects, as will appear, but subject to liability under clause 3.1 of the agreement of March 1994 (see above) says that no loss has been or will be suffered by ET as a result of either these admitted unrectified defects or any other defects for which Birse may be found liable and that ET is only entitled to nominal damages. This topic is considered in section 7 below: "Damages".
42. Birse maintain ET's conduct shows that it has no real intention to carry out the works. ET had not carried out any work to the roofs even though the trial had been settled in October 2002. Not even cosmetic works have been carried out to the roof. Mr Burns acknowledged there were still trees and other small plants growing in the waveform roofs. Birse submitted ET would sell the college as its interests were protected by an agreement with Host that ET would lease back two areas within the teaching wings to be used for technical training for which steps had been taken in September 2003.
43. In these circumstances it was plain to me that ET had no intention of carrying any remedial works. In 2000 it decided to sell the college. It has only put right the minimum. It has not even rectified the roof. I do not believe that at any stage prior to the conclusion of the trial it has ever intended to put rights the defects which have not been remedied, as was pleaded – see for example paragraphs 8.66 and 29.40 of the counterclaim and the quantification of ET's claims on that basis. Its pleadings in this respect appear to have been unsustainable when served and are now seen not to have credibility. It is surprising, to say the least, that any real intention to rectify was ever maintained. Furthermore ET's conduct in not disclosing the documents relating to the sale and its reluctance to comply promptly or at all with its obligations as to discovery seem to me, if not simply the consequence of poor judgment or advice (which it may well be), to be consistent also with a wish not to reveal the true state of affairs. In my view there is considerable force in Birse's submission that Schedule 4 was advanced solely as a device to avoid paying the retention due under the agreement of March 1994 and was not a genuine claim. For example, Mr Fleming did not inspect all the defects and had to produce further supplementary reports. Birse submitted also that if ET had a genuine claim it would have been advanced at the same time as its claim against MJP in respect of the roof (put at £532,227.08) especially if it was worth £2,595,262.39 as pleaded, since that was five times the amount of the claim on the roof. (Against this one has to ask why Birse did not claim what was due to it if it thought that it was blameless.)
44. On that basis ET would have suffered no loss and until the decision of the House of Lords in Ruxley Electronics v Forsyth [1996] AC 344 it is clear that no damages (other than nominal) would have been recoverable. (Diminution in value cannot be awarded to ET.) ET submitted that damages were recoverable. For the purposes of this section it ought to be recorded that ET claims damages for the breach of the building contract; for damages for breach of clause 17.2; a debt due under clause 3.3 of the Agreement and damages for breach of clause 3.1 of the Agreement. As to the latter, ET's remedy is contained in clauses 3.1 and 3.2 (although damages for losses or cost consequential on its primary expenditure might also be recoverable).
45. ET submitted that if a party sells his property the proper approach is to consider the measure of damages on the principles in Ruxley. Miss Jefford contended that where a party has contracted for the provision of goods and services of a particular quality and goods and services of that quality are not supplied, the innocent party suffered a loss in the sense of the loss of a bargain in that the innocent party has not got what it paid for (or is liable to pay for). It was not necessary for the innocent party to establish some other further or consequential financial loss in order to recover substantial damages. Thus the normal measure of damages in construction cases is the cost of remedial works. In many such cases, a building that is "less good" in some respect will nonetheless be functional and capable of being used for its purpose. Nevertheless the innocent party will have suffered a (financial) loss in that he has not got what he bargained for and that loss will usually be quantified by reference to the cost of remedial works. Miss Jefford relied on the speech of Lord Lloyd in Ruxley at page 366 as he there accepted that the cost of reinstatement is the ordinary measure of damages for defective performance under a building contract even where there is little or no difference in value or where the difference in value is hard to assess. She submitted that that loss is not any the less "real" if the defect is remedied and a cost incurred in doing so; nor is it a loss which somehow ceases to be real if the defect is not remedied and no cost is incurred. The loss is, and is caused by, the defective performance provided and is not a function of or the consequence of the action or inaction of the innocent party. Ruxley was also concerned with the application of that approach and was principally concerned with the issue of the quantification of that loss: (a) whether that quantification should be by reference to the cost of remedial works and (b) the circumstances in which that is not the proper measure of damages. She said that the issue, as expressed by Lord Jauncey (at page 357) was a question of whether the loss extends to the need to reinstate:

"If it is unreasonable in a particular case to award the cost of reinstatement it must be because the loss sustained does not extend to the need to reinstate".

Lord Lloyd similarly made the first inquiry "to ascertain the loss the plaintiff has in fact suffered by reason of the breach" (at page 365). In addition the cost of reinstatement being the ordinary measure of damages in a building case, Having referred to Jacob & Youngs Inc v Kent (1921) 230 NY 239) as endorsing the cost of reinstatement as the ordinary measure of damages in a building case, he said:

"Cardozo J's judgment is important because it establishes two principles which I believe to be correct and which are directly relevant to the present case: first, the cost of reinstatement is not the appropriate measure of damages if the expenditure would be out of all proportion to the good to be obtained, and secondly, the appropriate measure of damages in such a case is the difference in value, even though it would result in a nominal award. "

He later restated these principles, expressed in terms of reasonableness, at page 368:

"Once again one finds the court emphasising the central importance of reasonableness in selecting the appropriate measure of damages. If reinstatement is not the reasonable way of dealing with the situation, then diminution in value, if any, is the true measure of the plaintiff's loss. If there is no diminution in value, the plaintiff has suffered no loss. His damages will be nominal.

These principles are recognised in the leading English authority, East Ham BC v Bernard Sunley & Sons Ltd [1966] AC 406…..".

She suggested that the answer to those questions, (a) and (b), above therefore involves the consideration of the reasonableness of the cost of remedial works as the measure of damages. She contended that amongst the matters relevant to reasonableness were: (1) whether the cost of carrying out the remedial works is disproportionate to the benefit to be obtained from carrying out the remedial works; if it is, then the loss does not extend to the need to reinstate or the cost of remedial works is not the reasonable measure of damages; (2) whether the injured party intends to reinstate. They were related in that intention is a relevant because it may be some evidence of whether the cost of carrying out remedial works is disproportionate to the benefit to be obtained and/or reasonableness. She relied on the speech of Lord Lloyd in Ruxley at page 373:

"The relevance of intention to the issue of reasonableness is expressly recognised by the respondent in his case. In para 37 Mr Jacob says:

'The Respondent accepts that the genuineness of the parties' indicated predilections can be a factor which the court must consider when deciding between alternative measures of damage. Where a plaintiff is contending for a high as opposed to a low cost measure of damages the court must decide whether in the circumstances of the particular case such high cost measure is reasonable. One of the factors that may be relevant is the genuineness of the plaintiff's desire to pursue the course which involves the higher cost. Absence of such a desire (indicated by untruths about intention) may undermine the reasonableness of the higher cost measure.'


I can only say that I find myself in complete agreement with that approach, in contrast to the approach taken by the majority of the Court of Appeal.


Does Mr Forsyth's undertaking to spend any damages which he may receive on rebuilding the pool make any difference? Clearly not. He cannot be allowed to create a loss which does not exist in order to punish the defendants for their breach of contract. The basic rule of damages, to which exemplary damages are the only exception, is that they are compensatory not punitive."

Thus, Ruxley says that the court is not normally concerned with what the injured party actually intends to do with the damages awarded.

46. Miss Jefford therefore submitted that any proposition that the cost of remedial works is not the proper measure of damages in any case where the innocent party does not intend to carry out remedial works (or may not do so) must be wrong for if it were tenable, Mr Forsyth's intentions would have been central. The reason why the innocent party does not intend to carry out remedial works is also irrelevant. Miss Jefford developed ET's case on proportionality. She submitted that this topic should be examined in terms of the thing contracted for and what would be necessary to give the innocent party the thing contracted for. In Ruxley the House of Lords considered that the relationship between the remedial works and the improvement of the quality of the thing in issue was relevant - see the citation of Jacob & Youngs v. Kent and Bellgrove v. Elldridge (1954) 90 CLR 613 (by Lord Jauncey, at page 367). She submitted that Channel Island Ferries Ltd. v. Cenargo Navigation Ltd. ("The Rozel") [1994] 2 Lloyd's Rep. 161 (although decided in the light of the decision of the Court of Appeal in Ruxley) is an example of the same approach since Phillips J. held (at page 168) that the right question was

"… what, if any commercial implications flowed from the mention in the memoranda and whether, having regard to those implications, it would be reasonable to replace the entablature in order to remove the mention".

Miss Jefford maintained that the fact that the innocent party will (or even may) not carry out the remedial works because of a sale is not relevant to the issue of proportionality or reasonableness. Thus what ET has or has not done is not relevant other than as evidence of proportionality.

47. Accordingly it was said that the issue in relation to the unremedied defects is the proportionality of carrying out remedial works and that can only be answered by reference to the particular defects and not on some wider basis. Miss Jefford also relied on the circumstances in which the loss arose – the March 1994 agreement and its aftermath. In addition, ET was entitled to general damages for loss of amenity: many of the unremedied defects are ones that affect the general appearance, comfort and amenity of the College, including the mosaics, the joinery and the ceilings.
48. For Birse, Miss Jackson first said that it was not Birse's case that ET could only recover damages for the unrectified Schedule 4 defects if it actually carried out remedial works as such a contention was wrong in law. Birse faced an allegation that ET intended to remedy the defects and thus should recover the cost of so doing. It denied that ET has any such intention. Nevertheless ET still claims to be entitled to recover the cost that it would have incurred if it did remedy the defects. Miss Jackson submitted that the correct measure of damages was the classic expression by Lord Blackburn in Livingstone v. Rawyards Coal Co. (1880) 5 App.Cas. 25 at page 39:

".. that sum of money which will put the party who has been injured, or who has suffered, in the same position as he would have been in if he had not sustained the wrong for which he is now getting his compensation or reparation."

It was applied by Megaw, L.J. in Dodd Properties v Canterbury City Council [1980] 1 WLR 433 at page 451 where he also said:

"In any case of doubt, it is desirable that the judge, having decided provisionally as to the amount of damages, should, before finally deciding, consider whether the amount conforms with the requirement of Lord Blackburn's fundamental principle. If it appears not to conform, the judge should examine the question again to see whether the particular case falls within one of the exceptions of which Lord Blackburn gave examples, or whether he is obliged by some binding authority to arrive at a result which is inconsistent with the fundamental principle. I propose to carry out that exercise later in this judgment."

It would be wrong in principle if ET were to be awarded damages equivalent to the cost of carrying out remedial works when the works were not be carried out by it and where the defects had no effect on the sale price. ET would get a windfall as it would get, if completely successful, about £1.5 million in addition to the sale price. It had not carried out the repairs in July 1996 after Birse had been instructed to stop and to leave the site. It did not carry them out in 1999 and 2000 at the same time as the other remedial work. The facts were not comparable with Dodd or any other case.

49. Miss Jackson maintained that Ruxley established that where the expenditure on remedial works is out of all proportion to the benefit to be obtained, the appropriate measure of damages is not the cost of reinstatement but the diminution in the value of the work occasioned by the breach even if that would result in a nominal award. She referred to the speech of Lord Bridge (at page 353) as helpful:

"The circumstances giving rise to the present appeal exemplify a situation which one might suppose to be of not infrequent occurrence. A landowner contracts for building works to be executed on his land. When the work is complete it serves the practical purpose for which it was required perfectly satisfactorily. But in some minor respect the finished work falls short of the contract specification. The difference in commercial value between the work as built and the work as specified is nil. But the owner can honestly say: "This work does not please me as well as would that for which I expressly stipulated. It does not satisfy my personal preference. In terms of amenity, convenience or aesthetic satisfaction I have lost something." Nevertheless the contractual defect could only be remedied by demolishing the work and starting again from scratch. The cost of doing this would be so great in proportion to any benefit it would confer on the owner that no reasonable owner would think of incurring it."

Thus an intention to remedy is only one factor. Intention was however the subject of part of Lord Jauncey's reasons (at page 357):

"Damages are designed to compensate for an established loss and not to provide a gratuitous benefit to the aggrieved party from which it follows that the reasonableness of an award of damages is to be linked directly to the loss sustained. If it is unreasonable in a particular case to award the cost of reinstatement it must be because the loss sustained does not extend to the need to reinstate. A failure to achieve the precise contractual objective does not necessarily result in the loss which is occasioned by a total failure. This was recognised by the High Court of Australia in the above cited passage in Bellgrove v Eldridge 90 C.L.R. 613, 617-618, where it was stated that the cost of reinstatement work subject to the qualification of reasonableness was the extent of the loss, thereby treating reasonableness as a factor to be considered in determining what was that loss rather than, as the respondents argued, merely a factor in determining which of two alternative remedies were appropriate for a loss once established. Further support for this view is to be found in the following passage in the judgment of Sir Robert Megarry V.-C. in Tito v Waddell (No 2) [1977] Ch. 106, 332:

"Per contra, if the plaintiff has suffered little or no monetary loss in the reduction of value of his land, and he has no intention of applying any damages towards carrying out the work contracted for, or its equivalent, I cannot see why he should recover the cost of doing work which will never be done. It would be a mere pretence to say that this cost was a loss and so should be recoverable as damages."

The Vice-Chancellor was as I understand it there saying that it would be unreasonable to treat as a loss the cost of carrying out work which would never in fact be done".

Miss Jackson therefore submitted that ET's claim for £865,000 or so was out of proportion to the benefit, if any, to be obtained from remedying these items. She referred also to Lord Bridge (at page 354): "… to hold in a case such as this that the measure of the building owner's loss is the cost of reinstatement, however unreasonable it would be to incur that cost, seems to me to fly in the face of common sense".

50. On reasonableness (or proportionality) Miss Jackson pointed to ET's failure to put right work which it claimed affected the amenity of the college. She said that many of the defects could easily have been put right by ET in the ordinary course of maintenance. Mr Burns did however say in his statement that he thought it likely that after 1996 outstanding items on the snagging list were remedied by his department in the normal course of their duties, such as door handles and window openings. However in cross-examination he agreed that nothing had been done about visible defects such as paint splashes on lintels and metal lintels that have numbers stamped on them and needed to be cleaned with a wire brush and he had no real explanation as to why the items about mud on a door and rubbish in a fire hydrant had not been attended to.
51. In my judgment it is now clear from Ruxley (and indeed many of the cases referred in the speeches) that the normal measure of damages for defective works is the cost of reinstatement (i.e. the cost of remedial works) but in every instance it has to be reasonable to apply it. Thus where that measure is out of proportion to the claimant's real loss then some other measure should be used. This is the case where there has been a modest effect on the utility of the works and where it would be reasonable to assess the loss on the basis of diminution in value. A pragmatic approach may have to be applied although the claimant is not to be too readily deprived of the ordinary measure of compensation. However in my judgment Lord Bridge encapsulated the essence of the approach in his reference to the use of common sense.
52. Even allowing for the unique nature of the building and the demands of MJP's design (to which I shall come) I shall assume for present purposes (but there is little doubt about it) that Birse did not clear the snagging lists with the determination and application required, especially after the agreement of March 1994. That gave them a defined incentive in terms of the unpaid balance of the account yet it took until early 1996 before a degree of satisfaction was achieved (here I make an assumption in Birse's favour). After Birse had been instructed to stop, the position changed in as much as that the way was clear for ET to put right the defects and to recoup from the money held back under the terms of the March Agreement or, on the basis of a breach or default, from Birse itself or the bondsman, without even putting them right. ET did deal with the items which Ms Tooth considered significant but it did not deal with the remainder which are the subject of this section. Whilst there are explanations for not dealing with some of the items (such as inaccessibility) it defies belief that there were any items which truly affected the general appearance, comfort and amenity of the College. If there were any such items then any reasonable owner with resources (and ET had them) would have put them right promptly either at the same time as the main works or soon thereafter as part of a planned scheme. ET did not do so. On that basis alone a claim for general damages based on the totality of the unremedied defects must fail. (I shall consider if any individual item or group of items justify such an award.) From the latter part of 2000 the reason must have been the proposed sale, but equally I infer that the items were not such as to warrant attention so that the complex was spruced up so as to improve its prospects. To award ET the cost of putting them right seems to me to be unreasonable and, on the basis of the pleaded claim of £865,000, or, indeed the agreed figures, out of all proportion to ET's loss which appears to be minimal, in any terms, financial or economic. There is no evidence that ET's business has suffered or even that the work of any of its students has been affected.
53. Moreover if the college is sold, which was probable at the date of trial (and by all accounts is still probable), ET's supposed loss will have been avoided. Just as a claimant's loss may increase between the date of breach and the date of trial (when damages are to be assessed) it may decrease. A loss that has been avoided or is reasonably avoidable is not recoverable. It is not in law right or reasonable to compensate the claimant for such a loss. The fact that remedial work has not yet been carried out does not mean that the loss has been avoided. But a contractor which sought to offset a claim for defective work against the final account of a sub-contractor when the employer had foregone any claim against the contractor would receive short shrift. It would not be reasonable to award the contractor damages for breach of the sub-contract. It would be quite another question if the contractor remained exposed to a claim from the employer and the work had been put right. It is clear from Ruxley that, in the absence of any factor that would make it unreasonable, the measure of damages would be the cost of reinstatement. If a building owner disposes of property with defects attributable to some breach of duty by the defendant and for which the cost of reinstatement was the appropriate measure but does so without any reduction or loss on account of its condition then the loss that the law supposes is avoided and no damages are recoverable. In some cases it may be reasonable (or even proportionate) to award an amount so that the contractor does not get paid for what was not done (if it was not done at all, then either an appropriate contractual reduction in the price or a comparable award of damages should be made).
54. Miss Jefford's submissions based on recovery for "loss of a bargain" are, in the light of Ruxley, soundly based in principle (and are also justifiable by academic analyses). I can see that it could well be thought to go against the grain and might fly in the face of common sense if Birse were to be paid amounts retained under clause 4 of the agreement without reduction. However, leaving aside any pleading point, such an approach is not open to ET. First, under a building contract such as this such an approach is met by an appropriate valuation of the works since in part they have not been properly executed. The settlement agreement resolved all points on the final account and created new arrangements for defects. No further certificates were to be issued under clause 30 so there would no power under clause 41 or other means to revise the valuation that had in any event been agreed. ET did not reserve such a right when entering into the settlement. It is not preserved by clause 6 since, as I have said, the relevant provisions ceased to be effective for that purpose. Moreover the agreement of March 1994 contained more than adequate methods to protect ET: it held back very large amounts (and had the bond renewed); it could make a near arbitrary assessment of its likely remedial costs and hold that amount (which in practice would be at least tantamount to the application of the approach); it could put the work right itself and recoup the cost from the money withheld. In fact it released money to Birse at a time when, if its case is correct, there remained numerous unremedied defects. In these circumstances, even if the figures were available, it would also not be reasonable to use this approach as an alternative measure of damages to the cost of reinstatement. Nevertheless, depending on the outcome I may consider whether such an approach might be applied to a significant item or category of items. Otherwise ET is only entitled to nominal damages in respect of those unremedied defects for which Birse may be liable to ET (unless there is an exception).

4. Remedied Defects

55. The schedules about the defects are extensive. The defects break down into three categories: (1) Remedied defects, (2) Unremedied defects which ET intends to remedy and (3) Unremedied defects which ET does not intend to remedy. This section deals with the former. The latter two are considered together. ET makes its claims (1) under clause. 3.2/3.3 of the agreement of March 1994; as damages for breach of clause 3.1 of that agreement; as damages for breach of the building contract; and (4) specifically, damages for breach of clause 17.2 of that contract. In practice, except for the application of clause 3 of the 1994 agreement, there is little difference between any of these approaches.

4.A Pozament Render

56. MJP's design provided for brilliant white, eggshell smooth external walls. To achieve this a self-finished white render was to be applied. Pozament render was used. Miss Jackson said that they were to bisect and define the site. The walls were to be in the following position: the outbuildings – the gardener's store, the transformer room, the communications room and the sub-station; the garden walls, including the north wall and site entrance wall; the east and west garden seats; various elevations on the residences, including the pods on the south elevation of the residences and the "cruciform" windows on the north walls of the residences; the gable end walls of the teaching blocks; the north elevation of the administration block; the leisure block. Ms Tooth provided this useful description:

"One of the most important aesthetic features of the design of the College is the provision of white render panels to the main buildings on the site, set against a backdrop of fair-faced blockwork walling. The white render is extended to the four outbuildings, which are fully rendered. Rendered garden walls and seats are also important aesthetic features of the site, some of the walls and both of the seats being finished in the same white render (there are also feature garden walls with a green Armouralia render finish defining the site and building main entrances, a mast wall with a red Armouralia finish, and a curved feature wall with a blue paint finish)."

(The complaints in respect of the Armouralia follow.)

57. The white render has cracked and crazed. ET has carried out repairs to cracked and debonded render and overcoated it with an acrylic compound ("Sto acrylic"). The quantum experts have agreed the cost of the work carried out at £220,055.16 including fees. £3,381.38 had been claimed solely against MJP, £168,984.86 against Birse and MJP jointly, and £47,688.82 against Birse alone. The claim against Birse is therefore £216,673.78. However, as a result of the settlement between ET and MJP (which, as noted above, expressly included the sum of £11,044.29 for Pozament render, being 25% of the total claimed against Birse and MJP in respect of the outbuildings), the claim against Birse is now reduced to £205,629.49 inclusive of fees.
58. MJP initially specified Fibrocem as the render. The specification provided (Part 1, 1/35G ff):

1.11.3 EXTERNAL RENDERING

G Materials: 'Fibrocem' as manufactured by Cement Glass Products Ltd, 44 The Green, Warlingham, Surrey CR3 9NA Tel: 08832 6511.

[Page 1/36]

A Generally: See Part 2 for the specification of workmanship, subject to the following requirements:

B Smooth finish for 'Fibrocem' render: shall be achieved using a stainless steel trowel to minimise the risk of metallic blemishes.

C Samples: Before placing orders submit for approval representative samples of all work.

D Uniformity of colour and texture: Once samples of coatings have been approved do not change type or proportion of constituent materials. Ensure that supplies of materials are sufficient to give consistent and uniform colour and texture. Obtain each material from one source and mix different loads if necessary.

E Cement: As specified in the type of coating clause(s). Where ordinary cement is specified, use ordinary Portland, rapid hardening Portland or Portland blast furnace cement. All cements must comply with the appropriate British Standard and be manufactured by a BSI Registered Firm of Assessed Capability.

F Cement based mixes: may be retarded ready-mixed mortars provided they are:

of materials and proportions specified in this section and to BS 4721.

used within the working time recommended by the manufacturer and not remixed on site.

Obtained from an approved manufacturer.

G Admixtures: Do not use, other than air-entraining and water-retaining admixtures, unless specified or approved. Do not use admixtures of any type with proprietary mixes. Do not use calcium chloride or any admixtures containing calcium chloride.

H Mixing:

Measure materials by volume using clean gauge boxes. Proportions of specified mixes are for damp sand: adjust proportions if dry sand is used.

Thoroughly mix materials dry until uniform in appearance. Add water and mix to a consistence suitable for the work.

J Mechanical mixers:

Do not use for premixed plasters unless recommended by manufacturer.

Do not use mortar-mill type for mixing gypsum plasters.

Do not mix materials for longer than 5 minutes.

Wash out mixer four times daily if in continuous use, and after each batch if mixer is used intermittently, or if a different constituent is used.

K Adulteration: Avoid contamination of one type of material by another and by any set material.

L Retempering: Do not use mixes after initial set has taken place and do not retemper or reconstitute mixes unless permitted by manufacturer.

[Page 1/37]

A Scaffolding: Use independent scaffolding to avoid putlog holes and other breaks in coatings.

B Cleanliness: Protect all existing work and approaches with boards, dust sheets, etc. All droppings on to finished work to be cleaned off immediately.

C Cold weather: Do not carry out external work when air temperature is below 3 deg C on a falling thermometer or 1 deg C on a rising thermometer or while any coating material or background remains frozen.

D Cold weather: Take all necessary precautions to enable internal coating work to proceed without damage when air temperature is below 3 deg C. Do not apply coatings to frozen backgrounds.

E Protection: Adequately protect newly applied external coatings against frost and rain for the first 48 hours using polythene sheet, hung clear of the face, or other approved method.

F Acceptance of backgrounds: Before starting preparation and applying coatings ensure that:

Backgrounds are secure, adequately true and level to achieve specified tolerances, free from contamination and loose areas, reasonably dry and in a suitable condition to receive specified coatings.

All cutting, chasing, fixing of concealed conduits, service outlets, fixing pads and the like and making good of the background, are completed.

G In situ concrete surfaces: Scrub with water containing detergent to ensure complete removal of mould oil, surface retarders and other materials incompatible with coating. Rinse with clean water and allow to dry unless specified otherwise.

H Smooth concrete surfaces: Where no keying mix or bonding agent is specified, wet smooth concrete surfaces immediately before plastering.

I Plasterboard backings:

Plasterboard: To BS 1230: Part 1.

Ensure that perimeter and unbound or cut edges of boards are fully supported by additional noggings as necessary.

Ensure that noggings, bearers, etc required to support fixtures, fittings and services are accurately positioned and securely fixed.

Arrange boards with bound edges at right angles to supports, and joints staggered between rows and a gap of 3 mm between boards.

Working from the centre of each board, fix to all supports at not more than 150mm centres with 2.6 mm diameter galvanised clout nails of length not less than 3 times the thickness of the board being fixed. Set heads flush; do not break paper or gypsum core.

K Beads/stops for internal use. Galvanised steel to BS 6452: Part 1

L Beads/stops for external use: Manufactured from stainless steel to BS 1449: Part 2, grade 304 S15.

[Page 1/38]

A Beads/stops:

Provide beads/stops at all external angles and stop ends except where specified otherwise.

Cut neatly, form mitres at return angles and remove sharp edges, swarf and other potentially dangerous projections.

Fix securely, plumb, square and true to line and level. Fix at not more than 600mm centres, ensuring full contact of wings with background. Dabs, if used, to be of undercoat material, fastenings to be galvanised or stainless steel.

After coatings have been applied, remove coating material while still wet from surfaces or beads/stops which are to be exposed to view.

B Dissimilar solid backgrounds for plaster: Where plaster is to be continued without break across joints between dissimilar solid backgrounds which are rigidly bonded or tied together, cover the joints with a 200 mm wide galvanised mesh strip (backgrounds in the same plane) or with galvanised corner mesh (internal angles) fixed at not more than 600 mm centres along both edges, unless specified otherwise.

C Dissimilar solid backgrounds for rendering: Where rendering is to be continued without break across joints between dissimilar solid backgrounds which are in the same plane and rigidly bonded or tied together, cover joints with a 150 mm wide strip of building paper to BS 1521 overlaid with 300 mm wide galvanised steel lathing fixed at not more than 600 mm centres along both edges, unless specified otherwise.

D Movement joints: Form joints in coatings to coincide with movement joints in background using:

Movement joint beads with sealant infill for joints in external render.

E Service chases: Cover with steel mesh strip fixed at not more than 600 mm centres along both edges.

F Conduits: bedded in undercoat to be covered with 90 mm wide jute scrim bedded in finishing coat mix, pressed flat and trowelled in. Do not lap ends of scrim.

G Scrimming: Where specified, bed 90 mm wide jute scrim centrally over joints using same plaster as for the first coat. Do not lap ends. Press well in, trowel flat and smooth and allow to set but not dry out before applying coating.

H Joints in Gypsum lath:

Fill joints between boards which are in the same plane with same plaster as for first coat. Allow to set but not dry out before plastering.

Scrim joints at internal and external angles (except where coincident with a metal bead).

J Joints between boards and solid backgrounds which are both to be plastered: fill and scrim except where specified otherwise.

[Page 1/39]

A Application generally:

Apply each coating firmly to achieve good adhesion and in one continuous operation between angles and joints.

All coatings to be not less than the thickness specified, firmly bonded, of even and consistent appearance, free from rippling, hollows and ridges.

Finish surfaces to a true plane, to correct line and level, with all angles and corners to a right angle unless specified otherwise, and with walls and reveals plumb and square.

Prevent excessively rapid or localised drying out.

B Accuracy of plaster 13 mm thick or more: The variation in gap under a 2 m straight edge (with feet) placed anywhere on the surface to be not more than 3 mm.

C Metal mesh lathing: Work undercoat well into interstices to obtain maximum key.

D Undercoats generally to be not less than 8 mm thick, with thicknesses greater than 16 mm applied to two equal coats. Rule to an even surface and across scratch each coat to provide a key for the next hand-applied coat.

E Gypsum/lightweight plasters: Apply final coat as soon as undercoat has set, is firmly bonded to background and has developed reasonable suction.

F Smooth finish: Trowel or float to produce a tight, matt, smooth surface with no hollows, abrupt changes of level or trowel marks. Do not use water brush and avoid excessive trowelling and over polishing.

G Rendering application generally:

Apply each coating firmly to achieve good adhesion and in one continuous operation between angles and joints.

All coatings to be not less than the thickness specified, firmly bonded, of even and consistent appearance, free from rippling, hollows and ridges.

Finish surfaces to a true plane, to correct line and level, with all angles and corners to a right angle unless specified otherwise, and with walls and reveals plumb and square.

Prevent excessively rapid or localised drying out.

H Accuracy of rendering to receive tiles fixed with adhesive. The variation in gap under a 2 m straight edge (with feet) placed anywhere on the surface to be not more than 3 mm.

J Undercoats:

Apply first undercoat or dubbing out coat by throwing from a trowel.

Coats to be not less than 8 mm thick, with thickness greater than 15 mm applied as two coats, the second thinner than the first. On weak backgrounds first undercoat to be not less than 10 mm thick. Comb surface of each coat as specified.

Brush down each undercoat to remove dust and loose particles and wet thoroughly before application of next coat.

[Page 1/40]

A Anchored mesh reinforcement: The first undercoat must be applied through and round the mesh to fully bond with the solid background.

B Final coat – plain floated finish: Finish with wood or other suitably faced float to give an even, open texture. Do not apply water while working up. Do not draw excessive laitance to surface (either by overworking or by use of steel trowel).

Installation

C Nail fixings for metal lathing to timber: When not specified otherwise, use 38 2 mm galvanised clout nails to BS 1202: Part 1 or 32 x 2 mm galvanised staples.

D Plain expanded metal lathing

Stretch lathing and fix securely in accordance with manufacturer's recommendations to give a taut, firm base for plaster/rendering.

Fix with the long way of the mesh at right angles to supports and with all strands sloping in the same direction.

Lap side edges not less than 25 mm. Lap ends 50 mm at supports and 75 mm between supports. Laps must not occur within 100 mm of angles or bends.

Tie all edges and ends together with 1.2 mm wire ties at not more than 150 mm centres.

E Lathing on solid backings: Fix securely in accordance with manufacturer's recommendations, increasing length and frequency of fastenings as necessary in weak areas. Inform before applying plaster/rendering."

59. Mr Prichard of MJP said that at quite a late stage in the design (RIBA Stage E or F) he had sought advice from Bickerdike Allen Partners (BAP). Mr John Streeter of BAP sent the following letter of advice to Mr Whitehead (one of the project architects but who was not in the event called) dated 15 November 1991:

"Dear Mr Whitehead

CABLE & WIRELESS TRAINING COLLEGE

I refer to our meeting on 22 October 1991, your letter dated 23 October 1991 and our subsequent telephone conversations.

Fibrocem High Performance Renders

I have discussed the properties and performance of this proprietary rendering system with colleagues at BAP and other interested parties including Mr Panther of BBA (involved in testing for Agrément Certificate 88/2060), Dr Majumdar of BRE and Mr Gilbert of Cemfil International (manufacturers of the fibre reinforcement).

There is a general consensus of opinion that there has been no significant failures of this type of rendering which can be attributed to the material itself or its constituent parts, or in particular to the Cemfil glass fibre reinforcement.

However, it is clear that in this country the product has for the most part been "niche marketed", specifically for internal use where high impact strength is required.

Consequently, it does not appear to have been used to any great extent in external situations but where it has been so used it is normally decorated with a masonry paint and not left in a self-finished state.

On the limited evidence we have seen, and the opinion of others there would seem to be no fundamental reason why the material should not perform as the manufacturer's claim.

The absence of specific information about comparable "performance in use" for this material presents great difficulties when trying to evaluate its likely performance and durability.

Normally, we would only recommend the use of such a detailing and workmanship sensitive material in a prominent and exposed situation on the basis of reference to previous examples of successful installation in similar circumstances, i.e. assessing the material by its track record – rather than by manufacturer's claims or necessarily limited laboratory testing.

As no such proven track record of use appears to exist, you must rely on close compliance with the generally accepted principles of good design and workmanship for similar conventional smooth renders.

If the material does out-perform normal smooth renders, in particular with regard to shrinkage cracking, as claimed, then the overall performance of the application may be satisfactory.

We would most strongly advise against any lowering of normal standards of good practice in design, detailing and workmanship in the belief that this material will in itself have 'extraordinary' properties which will fully compensate for any other failings.

Thus, if all recommendations for 'normal' rendering are adopted the special properties of this material may give some increased protection.

I would stress the importance of applying the material strictly in accordance with the manufacturer's instructions, as clearly set out in the current Agrément Certificate.

The detailing and workmanship used in the application of the render will fundamentally determine the performance and durability of the material. The render should be considered in all respects as a conventional smooth render and follow the guidelines given in BS5262, unless otherwise advised by the manufacturer and with clear explanation of why variation is necessary.

The advice in the following documents should be also followed where it is applicable to smooth render finishes:

- CCA Appearance Matters 2

'External Rendering'

BRE Defects Action Sheets 37 and 38

I would also refer you to the joint guidance on the use of cavity insulation.

"Cavity insulated Walls 0 Specifiers' Guide"

CIW – 2 January 1987

(BDA, Eurisol and others)

This reiterates the important points of detailing and workmanship to consider with the use of full and partial fill cavity insulation.

General Points on detailing

The following comments apply generally to the details of the rendering/external walling I have seen.

It is not advisable to render the wall surface below lowest dpc level. This area of wall will be subject to moisture moving up from the ground by capillary action, and is a splash zone. Render will severely restrict evaporative drying out of the underlying masonry especially where cavities are fully filled with insulant material. This, therefore, increases the risk of frost action and delamination of the render from the substrate. In any event the render will rapidly become dirty and discoloured due to splashing. Also this zone of the wall is likely to be at most risk from impact damage.

The lower edge of the render of dpc level should always be formed to a 'bell cast' profile to ensure water running down the wall face is shed away from the base of the wall and away from the edge of the dpc which preferably should project slightly and on no account should be pointed over or otherwise bridged by render or jointing mortar.

If flush cappings to parapet walls are unavoidable it must be accepted that the risk of staining and water penetration at the head of the wall are significantly increased, with the consequential risk of delamination of the render and possible water penetration to the inner leaf of the wall. Staining will be most pronounced at joints between copings, cappings or lengths of flashings.

The risk of damage and deterioration can be somewhat reduced, but not entirely eliminated by the use of an inset bed joint flashing below the capping dpc to protect the head of the render. This flashing must have welted sealed joints to ensure water is not directed into the wall, but some drip staining must be anticipated at any joint however well and tightly formed (see CCA publication fig 21).

I suggest that render stop beads, bell case beads etc. should all be of stainless steel and not uPVC. Galvanized mild steel has a lower coefficient of thermal expansion than either stainless steel or uPVC – much closer to that of the render itself thus the risk of cracking at these details would be must reduced. However, because of the risk of corrosion and rust staining in the longer term the durability of stainless steel is to be preferred with slightly increased risk of movement cracking of the render – but still with less risk than with uPVC.

Metal lathing used to bridge over lines of potential movement in the substrate – where this is otherwise absolutely unavoidable (e.g. by the use of corresponding joints in the render) should be adequately fixed, and sufficiently proud of the wall to permit a proper render key to be achieved. However, the render must also be further separated from the background by the use of a slip strip of building paper, or similar material sandwiched between the expanded metal and the wall. This will ensure that through bonding to the masonry substrate cannot occur as the render is applied into the lathing.

Areas of render should be 'manageable' i.e. day work joints in areas of render will be almost impossible to disguise in smooth, self-finished work. Alternatively, ensure day work joints coincide with, or are formed as, a feature such as a recessed joint. In particular the radiussed wall shown on dwg CW 7214 seems excessive for a 'single' one stage jointless application.

All window cills in areas of rendering should have upstand, or stooled ends to ensure that water run-off is directed to and only occurs from the front edge of the cill.

This edge must project sufficiently to throw the water well away from the vulnerable top edge of the render (below the cill) and prevent water 'blow back' underneath. The greater the slope, the outward projection and the depth of downturn front edge of the cill, the better protection will be afforded to the rendering.

The ends of cills are likely positions at which movement cracking may occur, and thus potentially risky regarding water penetration. Also by ensuring, as far as possible, that water drips uniformly off the front edges of cills it may help to avoid concentrated runs of water staining on the render at cill ends.

Vertical dpc's at jambs should project into cavities by at least 25 mm (i.e. increase overall width of dpc) to comply with recommendations of BS 5628. The jamb dpc should be similarly overlapped by 25 mm by cavity try dpc's at head of openings. All discontinuous cavity tray dpcs to have stop ends at least 75 mm high.

…."

60. Nevertheless, in spite of this advice, MJP wanted a smooth finish (also described by Mr Wisniewski in his witness statement and in court as eggshell) even though smooth renders are not suitable for external use – see BS 5262 which says that steel trowels must not be used for external renders. This is because the use of a steel trowel brings laitance to the surface which is rich in cement and thus liable to craze. However a smooth finish can only be achieved by the use of a steel trowel and by repeatedly working the render until the aggregate (including the fibres incorporated to provide strength) is sufficiently depressed to leave a smooth layer of laitance. Therein lies the paradox: as Ms Tooth said, "It is not what the smoothness is but it is associated with it. You do not have one without the other". However Mr Cherry noted in April 1996: "We required a sharp appearance, not soft roundness as with traditional renders". Mr Prichard agreed that he decided to specify Fibrocem as the render because traditional render gives a very soft and rounded appearance and he wanted something sharp and smooth and because other, more traditional renders are more prone potentially to cracking and thus require overpainting. Fibrocem was at the time the only self-coloured product that would achieve the result MJP sought. Mr Prichard did not know at that time that BS 5262 said that a steel trowel must not be used for the finish.
61. On 19 June 1992 Birse proposed using Pozament as an alternative because of the financial position of Fibrocem:

"Following the demise of Cement Glass Products who were the sole producers of Fibrocem in the UK in 1991, Birse Construction felt it necessary to investigate the viability of using the product.

The investigations have led us to believe that a large risk could be taken; firstly by accepting the product from a manufacturer who has only started production within the last 12 months and secondly, by placing an order with a company part run by a previous director of Cement Glass Products, who due to their poor management went into receivership.

We therefore request to change the supplier of the material from Fibrocem/Wetherby Stone to Pozament, and we confirm that there will be no financial increase to the rates already in the Bill of Quantities.

The technical literature to this product is enclosed and I believe you have already accepted a sample panel on site."

As recorded in the letter, Birse had by then prepared sample panels which had been seen by MJP (see MJP's letter of 4 June 1992) and which were approved (see CVI of 26 June 1992 and MJP's letter of 30 June 1992) on the basis that the render would be applied in accordance with the manufacturer's instructions. MJP's letter of 30 June 1992 said

"Fibrocem

With reference to your letter dated 19 June 1992 and received on 29 June 1992 we confirm our approval of Pozament XF based on the render sample (No 2) on site. We record that this panel did not have an "expamet area" detail and suggest that it may be prudent to produce one on the back of the sample wall.

We note the literature states optimum 10-12mm cover where as the sample panel was 16-17mm. Please note our details require a 10mm cover in certain key areas and we alert you to this requirement.

We trust it will be applied in accordance with manufacturers instructions."

The experts concerned with materials agreed that "to all intents and purposes, Fibrocem and [Pozament] GR Render were the same product". (The GR render is the final coat and is applied to the XF render.) Mr Cherry said that Birse had constructed on site large samples (approximately 4m² each) of both materials for comparison; that the samples were prepared under the supervision of Birse; that MJP was assured that they were prepared in accordance with the supplier's instructions and constructed by the same rendering sub-contractor; that a perfectly smooth white hard finish was achieved to both samples; that the samples remained on site for a couple of months; that they were sprayed with water to see if they were subject to any colour change when wet and to check if the render was unduly wet afterwards; that both samples performed well and at that point MJP allowed Birse to change the supplier of the external rendering material from Fibrocem to Pozament. There is no suggestion that that sample panels exhibited crazing or cracking. Obviously MJP had not approved render that was cracked or crazed. The manufacturers of Pozament produced information sheets for GR renders and XF renders which are relevant. Those for the GR renders (1992 version) included:

"Pozament GR Renders

Information Sheet

Specification for rendering Brickwork and Concrete Blockwork

with normal suction

The background should be constructed to a strict tolerance, clean and dry.

When mixed to a working consistency GR Render Grade HM should be applied to the background as a floating/scratch coat at between 6-10mm thickness. Lightly devil float to provide a key. If the alignment is poor, two coats may be required.

The finishing coat of GR Render Grade HM should be applied whilst the first coat is still green (within 24 hours) at a thickness of 3-6mm and trowelled to a flat finish. When the 24 hours limit is exceeded then the first coat should be primed with a PVA bonding agent (interior) or SBR bonding agent (exterior) according to the manufacturer's instructions. The finishing coat should then be applied whilst the bonding agent is still tacky.

When the surface is to be left undecorated, then a stainless steel trowel should be used to minimise the risk of metallic blemishes.

Do not allow the render to dry out too quickly. Drape with a polythene sheet or spray with clean water until fully cured.

Please use this specification in conjunction with the GR Render Product Data Sheet and ensure that you have the latest information.

Pozament GR Renders

Information Sheet

Specification for Rendering to Concrete and Low Suction Blockwork

The background should be constructed to a strict tolerance, clean and dry. In the case of new concrete, any mould oil contamination should be removed by washing with a detergent solution and rinsing with clean water. Allow to dry.

Prime the background with a PVA bonding agent (interior) or SBR bonding agent (exterior) in accordance with the manufacturers instructions.

When mixed to a working consistency GR Render Grade L should be applied to the background as a floating/scratch coat at between 6-10mm thickness. Lightly devil float to provide a key.

If the alignment is poor, two coats may be required.

The finishing coat of GR Render Grade L should be applied whilst the first coat is still green (within 24 hours) at a thickness of 3-6mm and trowelled to a flat finish.

When the surface is to be left undecorated, then a stainless steel trowel should be used to minimise the risk of metallic blemishes.

Do not allow the render to dry out too quickly. Drape with a polythene sheet or spray with clean water until fully cured.

Please use this specification in conjunction with the GR Render Product Data Sheet and ensure that you have the latest information.

62. The specification for the XF render was as follows. (Both documents were evidently submitted to MJP when Birse suggested the change from Fibrocem.)

Pozament XF Renders

Product Data Sheet

Description

Pozament XF Render is a premix of white Portland cement complying with BS 12, natural white aggregates, Cemfil alkali resistant glass fibre, polymer modifiers and workability agents. The product is designed to be used as a setting coat over GR Render.

Uses

As a setting coat for high performance GR Render when used as exterior render, internal plaster and squash court playing surfaces.

Advantages

Pozament XF Render, when used as specified enhances the whiteness and smoothness of the traditional GR Render without detracting from the other properties of the product.

Application

Pozament XF Render is designed to be applied as a setting coat on to a wet backing coat. The maximum thickness is 2mm and the material is applied by hawk and trowel within an hour or two of the backing coat. It is preferable to use a stainless steel trowel to reduce the risk of metallic blemishes.

Mixing

Pozament XF Render should be mixed only with clean water in a small free fall mixer or in a tub with a whisk. Do not over mix as this may damage the fibre. Never add extra water to a mix that has begun to stiffen i.e. after 1-2 hours. Use sufficient water to bring the mix to a creamy consistency.

Limitations

Should not be applied over a dry background.

Maximum thickness 2mm.

Cannot be used in temperatures below 4^oC or when unprotected from frost.

Coverage

At 2mm thickness approximately 4 kilos per square metre (240-250 square metres per tonne).

Packaging

25kg Valve sacks.

Quality Control

All Pozament products are subjected to quality control procedures in accordance with BS5750. Pozament is BS QAS approved.

Health & Safety

Cement Powder when mixed with water, releases alkali which may cause skin irritation – observe normal skin protection as for handling cement. A full Health & Safety data sheet is available.

Storage & Shelf Life

The sacks should be stored clear of the ground under dry cover, and stacked not more than 2 pallets high."

63. Birse retained a specialist subcontractor, Bellworth Plastering Ltd, to apply the renders. Bellworth produced Method Statements for its work. Birse submitted them to MJP on 9 July 1992:

"Cable & Wireless, Westwood Heath Road, Coventry

Re: Plaster/Render/Screed

We write to you concerning the above elements of work on our contract and attach Method Statements for plaster, render and screed, for your information/comments.

We will forward to you by 17 July 1992 further Method Statements on plasterboard ceilings, pozament render, acoustic plaster, Venetian marbling plaster, jambs and stud for your comments."

MJP replied on 20 July 1992 with no adverse comments. During the trial one of the documents submitted by Bellworth was treated as the method statement for the Pozament render although it is of general application. Although Bellworth's letter of 9 July envisaged the production of a method statement specifically for the Pozament render, none was found. I think that it probably did exist as MJP were keen on obtaining such documents. Mr Bowers said that he recalled it (although he was looking at the general statement). However since Bellworth's documents are simple and similar I imagine that it would have said much the same as the general statement, (e.g. "Render will be applied ….[as] recommended by the manufacturer".) The general method statement said:

"Bellworth Plastering Limited

Method Statement

RENDERING

Material Handling

All materials will be handled in accordance with manufacturer's instructions in respect of storage, mixing, application and safety. Materials are to be used in strict rotation and within prescribed periods.

Base Material

The base onto which render is to be applied will be checked for line, plumb, completeness and soundness. The compatibility of base and render specified will be confirmed.

Metal Components, i.e. beads, mesh

Beads will be bedded in S/C Render to a true line and plumb . Mesh will be either nailed to timber or bedded as for beads in all other instances.

Rendering

Render will be applied with firm pressure in the number and thickness of coats recommended by the manufacturer. Each coat is to be ruled to an even finish and lightly scratched before proceeding with the following coat.

….."

64. On 17 November 1992 the Minutes of a Quality Meeting attended by Mr Gordon Cox (the Clerk of Works) and Mr Whitehead recorded under "Quality of workmanship" that the Pozament render was "excellent". However the render crazed and, in places, it blew. The crazing first appeared on the north face of the administration building. Birse re-rendered repeatedly but always the problem recurred. Mr Cox listed the occasions when Pozament was applied and re-applied. In 1996 he produced lists for Ms Tooth. They were as follows:

"POZAMENT RENDER

TRANSFORMER ROOM

18.01.94 Removing render

20.03.94 Render complete

MAIN ENTRANCE WALL

05.09.93 Old render removed and new applied

11.03.94 Removing render

29.03.94 Render complete

GARDENERS COMPOUND

01.11.93 New render applied

25.01.94 Removing render

21.03.94 New render applied

09.03.95 Removing render

19.05.95 Render complete

15.11.95 Patching render

ADMIN NORTH WALL

20.04.93 Removing render

20.05.93 Render complete

17.01.94 Removing render

09.02.94 Render complete

SEAT BY SUB-STATION

15.11.93 New Render applied (last area to be completed)

06.12.93 Render completed

20.03.95 Render removed and new applied

SEAT AT END OF RESIDENCE WEST (DAMAGED BY J.C.B.)

09.03.95 Removing render

04.05.95 Render complete

NORTH GARDEN WALL

02.07.93 New render applied

02.08.93 Render complete

GAS METER ROOM/SUB STATION

04.08.93 Removing render 1st time (no date for applying new)

18.10.93 Removing render 2nd time

22.10.93 Render complete

01.03.94 Removing render (Gas Meter Room) 3rd time

22.03.94 Removing render (Sub Station) 3rd time

31.03.94 Render complete

15.02.95 Removing render 4th time

21.02.95 Render complete

28.02.95 Removing render 5th time

10.03.95 Render complete

TEACHING WEST

31.08.93 Applying render

31.01.96 BCL did have a visit by a Pozament Rep. to look at the problem. I did not see the report that he made, this was forwarded to BCL.

Gordon Cox

17.09.96

POZAMENT RENDER

1) 18.08.92 Commenced to Residence West Grid 'S' South Elevation

2) 24.08.92 Commenced to Residence East Grid 'S' House F South Elevation

3) 08.09.92 Gable end to Residence West Grid R1

4) 28.09.92 Commenced insulated panels to Residence West Grid 'T' North Elevation

5) 09.10.92 Gable end Residence West Grid 'R16'

6) 15.10.92 Residence West, Houses C & D Grid 'T' North Elevation

7) 16.10.92 Residence East, Houses H & J Grid 'T' North Elevation

8) 19.10.92 Residence West House A Grid 'T' North Elevation

9) 19.10.92 Gable end to Residence West Grid 'R1'

10) 21.10.92 Commenced to Administration Grid 'A1' North Elevation

11) 22.10.92 Residence West House E Grid 'R16' East Elevation

12) 23.10.92 Residence West House A – complete to West Gable

13) 26.10.92 Residence East House J Grid 'R29' East Gable

14) 28.10.92 Residence East Grid 'T' North Elevation

15) 30.10.92 Commenced to Administration grids A1-A2/T North Elevation

16) 03.11.92 Commenced to the Transformer Room

17) 23.11.92 Commenced to the Leisure Building – location unknown

18) 06.04.93 Administration – Service Yard

19) 07.04.93 Render to the Transformer Room

20) 08.04.93 Commenced to the Sub-station and Gas Meter Room

21) 14.04.93 Administration Grid Line 'U' North Elevation

22) 14.04.93 Commenced to Teaching West Soffites in areas to Grids D-E east end and on Grid Line 'D'

23) 15.04.93 Commenced to the Communications Room

24) 19.04.93 Administration Grid 'U' North Elevation

25) 25.05.93 Residence East. Porches to South Elevation

26) 25.05.93 Communications Room

27) 25.05.93 Residence West. Soffites to the South Elevation

28) 01.06.93 Residence West. Soffites to the South Elevation

29) 02.06.93 Administration. Curved area of the Dining Room

30) 03.06.93 Leisure Building. North Squash Court Wall

31) 08.06.93 Leisure Building. South Squash Court Wall

32) 07.06.93 Gardeners Store

33) 08.06.93 Central area. Soffite of the "Eye"

34) 08.06.93 Teaching West. Render to Gable West Elevation

…

36) 11.06.93 Teaching West Soffite to Gable Grid F-G East Elevation

37) 14.06.93 Teaching West Roof Soffites

38) 15.06.93 Internal walls of Gardeners Store

39) 15.06.93 Teaching West. Roof soffites

40) 21.06.93 Leisure Building. Soffites to Bar area

41) 29.06.93 Teaching East. Roof Soffites

42) 01.07.93 Main Entrance. Retaining Wall

43) 01.07.93 Teaching East. Overhang soffite to entrance

44) 02.07.93 Central area. Soffite of the "Eye"

02.07.93 Render to South Squash Court

45) 06.07.93 Teaching East Overhang soffite to entrance

46) 14.07.93 Render to North Garden Walls

47) 14.07.93 Residence West Soffites and upstands Grid 'S' South elevation

48) 14.07.93 Teaching Central. Soffit of the "Eye"

49) 16.07.93 Leisure Building North Squash Court soffite

50) 20.07.93 Administration Grids 'A1' and 'U'

51) 21.07.93 Residence West. Grid R16 East elevation

52) 21.07.93 Teaching East. Gable end. East elevation

53) 22.07.93 Teaching West. Gable end. West elevation

54) 22.07.93 Teaching West. Gable end. West elevation

55) 23.07.93 Administration – Service Yard ceiling

56) 26.07.93 Teaching West. Gable ends and soffites. West Elevation

57) 27.07.93 Teaching Central. Library Bridge Overhand

58) 28.07.93 Teaching West. Gable ends to Bays 2, 3 and 4 West Elevation

59) 02.08.93 Central area. Library Bridge Soffite

60) 04.08.93 Teaching East Gable ends. East Elevation

61) 11.08.93 Leisure Building soffite in front of swimming pool

62) 24.08.93 Residence East. Cloister soffites

63) 25.08.93 Residences East and West. Roof soffites

64) 27.08.93 Central area. library Bridge soffite

65) 02.09.93 Leisure Building. Rear doorways

66) 02.11.93 Gardeners Store

67) 18.11.93 Seat by the sub-station. East end of site

68) 02.12.93 Seat by Residence West. West end of site

G. Cox

Cable & Wireless College, Westwood Heath Road, Coventry

10.12.96

Pozament Render

Additional items located

22.10.92 Residence West, West Gable

22.04.93 Sub-Station, Gas meter Room

02.06.93 Teaching West. Soffite East end Grid D-E

08.06.93 Residence West. House E. Soffite to the Cloisters

15.06.93 Teaching West Roof Verges. Teaching East Roof Verges

26.06.93 Commenced fixing render beads to the Garden Wall

02.07.93 Leisure. South Squash Court

23.07.93 Teaching East. Gable ends

It is worth noting that by 5 July 1994 Birse had remedied all the areas originally required to be redone (but for an area in the gardener's compound and some panels in north garden wall that were to be done at the end of the defects liability period. Birse had achieved the eggshell finish required by MJP's specification. Mr Cox asserted that he kept a very close eye on the remedial work and if there were defects in the way that the Pozament being re-applied, he would have noted them and complained in writing (but there were no such complaints). The crazing and cracking in the Pozament render were not then considered by MJP to be the responsibility of Birse. They are not mentioned much in the snagging lists (and thus do not feature strongly in Schedule 4). In April 1996 Mr Cherry made a note about Pozament. In it he said:

"Pozament is now failing badly in two distinct ways, according to Bickerdike Allen and BRE: cracking due to the inability of the Pozament to cope with the movement of its substrates, especially but not exclusively around the stainless steel beads. 2, a chemical process sulphate attack which is turning the render to a paste in some places which, according to the BRE report, is an inevitable consequence of the composition of Pozament if the back of the render remains consistently wet."

"As far as we can tell, nothing implicates workmanship which is consistent with what we felt at the time some areas apart from which were dealt with separately. And with the two visits to site by Pozament, which we insisted on, Pozament had no negative comments on the way it was being applied."

By 7 July 1996, when MJP reconsidered the April 1996 report, I do not think that that MJP was expecting any more of Birse. In his evidence Mr Cherry agreed that, visually, workmanship did not then appear to be the cause (although investigations were still continuing). However he did say:

"Q. And it was not Birse being in default of an obligation to do anything as at July of 1996?

A. No. Birse replaced a number of panels. The reason why -- when it was going up for the first time -- is that some panels looked absolutely fine. We thought it was workmanship at the outset as I think we have been through, because when you apply the render and it looks fine and then you apply the next render and it crazes within a short period of time, you think why is that one fine and that one not?"

65. Miss Jefford submitted that the probable causes of the render crazing were (i) inadequate curing (that is, permitting the render to dry out too rapidly) and (ii) overworking of the render. Both amounted to breaches by Birse of the contract and of the provisions of the specification to which I have referred.
66. Birse's case was that the choice of the render was bound to lead to failure and that there was no (or no material) breach of contract on its part. Miss Jackson's contentions were numerous and detailed. The Pozament render had a very high shrinkability, and an even higher shrinkability in the crisp surface zone. Exemplary practices, i.e. beyond those normally required, were needed to minimise crazing and cracking but this was not specified. It was to be applied in contravention of BS 5262 with a steel trowel, thus achieving the laitance that designers normally strive (for good reason) to avoid in external applications. The smooth, eggshell finish necessitated re-working. Laitance is an essential characteristic of smoothness but is a prime cause of crazing as the material dries. Laitance does not indicate over-working. Nobody ever criticised Birse for overworking. The manufacturer's instructions did not warn against overworking. Pozament, unlike traditional renders, is a monolithic material and each coat is required to be applied much sooner after the previous coat than is usual. This prevents the undercoats from shrinking and the application was not likely to be achieved satisfactorily. The manufacturer's instructions for the finishing coat of XF render did not call for curing protection, when in fact the render was more susceptible to shrinkage than normal sand-cement renders At edges and corners, the render is taken over stainless steel beads where it was unlikely that the render would adhere fully to a smooth metal surface to produce crisp edges and corners. The stainless steel also has a different coefficient of expansion from the render, which will lead to differential movement. The manufacturer's instructions for the finishing coat of XF render did not call for curing protection, when in fact the render was more susceptible to shrinkage than normal sand-cement renders. MJP's design contained defects at coping level and at plinth level, which were liable to permit water ingress.
67. The question is therefore whether the result, in terms of crazing and cracking, was nevertheless an inevitable consequence of MJP's choice of the render or whether the cause lay in some breach of contract on the part of Birse.

Curing

68. If the render is not cured properly, it may shrink as it dries and thus crack. The relevant British Standard, BS 5262: 1991 gives advice. It says:

"28. Resistance to crazing

Crazing results from differential shrinkage of the surface of the rendering in relation to its interior. The cracks formed are narrow and generally do not extend far below the surface: they may, however, develop into shrinkage cracks (see clause 27).

Cement-rich steel-trowelled finishes are particularly liable to craze; on the other hand, leaner mixes with a scraped, textured, or other rough finish are highly resistant to this defect.

The risk of crazing should be minimised by:

…

(c) the avoidance of overworking, which causes an excess laitance to be drawn to the surface;

(d) the avoidance of too rapid drying out of the final coat."

45.2 Curing

It is essential that a newly rendered surface be prevented from drying out too rapidly, although protection from the sun and wind, or spraying with water, may only be necessary in hot and dry weather. In sunny weather, the work should be carried out in the shade whenever possible, following the sun round as the day passes.

…

Consideration should be given to more stringent than normal protection of coloured renderings and decorative finishes in hot weather, wind, rain, or other adverse conditions."

The contract specification (1/39/A) also has an express requirement: "Prevent excessively rapid or localised drying out". Curing enables the render to gain strength and thus resist the stresses so caused. The experts' joint statement said:

"It was agreed that the extent and adequacy of curing (or protection) will have been an important factor in the causation of crazing and cracking of the render, but there remains uncertainty over the measures that were intended or implemented."

Curing may take place by simply allowing render to dry out of its own accord. Dr Sims said that weather conditions often almost do the curing, by being wet or damp, but that for a render with a high potential for shrinkage it would be essential if there was heat and wind both being conducive to rapid drying. Dr Leek also took that view (see his report and evidence on Day 14). However more is required in order to prevent excessively rapid drying out. It was also agreed by the experts that, in this case, this would have to mean polythene sheeting hung close to the face of the render, shrouding it. In September 1992 there was some correspondence between Birse and its specialist subcontractor, Bellworth Plastering Ltd, about the subject. In it Birse made it clear (on 18 September 1992) the responsibility of Bellworth Plastering Ltd: "2. Cure protection and covering up own works on completion". That Bellworth was well aware of the need for protection for that and other purposes may be seen from its letter of 28 October 1992:

"We write with reference to the Pozament render on the external panels to the south elevation on Residence East and West and regret to inform you that the polythene protection has been removed by others resulting in damage to the Pozament panels and angle beads and also paint and wood preservative have been splashed onto the Pozament render.

As you will appreciate we cannot be held responsible for the damage that has occurred and accordingly will record any remedial measures that have to be taken."

In the absence of any other explanation these letters appear to me to be no more than routine exchanges about who was responsible for providing protection (almost certainly in terms of cost or financial liability) and the consequences in terms of liability to others if the absence of protection led to impact damage etc. It does not provide evidence that steps to promote proper curing were not being taken or were being hindered.

69. Time was spent poring over such photographs as there were to demonstrate that the polythene sheeting had not been provided or had been removed prematurely. Although the sheeting looked like and perhaps was ordinary monoflex covering (see Mr Barber's evidence) there was evidence from Birse's witnesses that the polythene sheeting was provided for curing. For example Mr Seamer recalled "plastic sheeting covering the scaffolding to aid curing and protect the render." The data sheets reinforced the obvious: "Cannot be used in temperatures below 4°C or when unprotected from frost". The render in both its applications was not thus not to be applied if not protected from frost. (I do not accept the view of Dr. Leek in his report that the manufacturer required for the XF render protection only from frost: the data sheets say otherwise.)
70. There was some debate as to whether adequate curing was to take place naturally or needed to be promoted. In my view the short answer to this lies in the absence of any evidence of complaint that Birse or its subcontractor was not "promoting" curing by intervening to control ordinary drying out. The latter would be normal practice, as Dr Leek said. Intervention would be required on "certain occasions", according to Dr Leek, which is obviously right. Dr Leek also said that it would have been unusual if a manufacturer did not require curing. Yet the manufacturer's instructions for the finishing coat of XF render did not call for curing protection, when in fact the render was more susceptible to shrinkage. The ordinary climate in this country evidently provides the basic conditions necessary for effective curing. Were it otherwise MJP would have drawn attention, at the least, in the specification and on their visits to the site, to the measures needed (as would BAP in 1991). Dr Leek's summary was: "I expected there to be a system which could comprise doing nothing, to applying some additional measures over and above natural exposure to the environment". The Pozament render, was remarkably, the "talk of the canteen" according to Mr Greenwood's statement and evidence. By this he included everybody. He said that the plasterers "had an awful lot of people standing around them while they were mixing it and putting it on. A lot of people meaning people out of the office". Thus Mr Cox said that, once it became apparent that crazing was occurring, he took steps to see that nothing was wrong with the workmanship. In view of Mr Cox's reliability, it is perhaps more relevant that he kept a record of the dates of the re-rendering work for Ms Tooth in order that she could see whether there any correlation between weather conditions and the damage. No such correlation was established. In these circumstances, in my judgment, the need for careful curing should it have been necessary was unlikely to be overlooked.
71. There was an attempt to discredit Dr Leek's views as it was said that the opinion that he had expressed for the trial differed from an opinion which he had given for the purposes of a mediation in 1997. I do not consider that the earlier opinion is inconsistent. In it Dr Leek did no more than point out the obvious. He considered the complaint by Birse to Bellworth about the removal of the polythene sheeting. He commented that if this "protection" formed part of the curing system, its removal would result in increased drying and crazing. In his later report he did not repeat that statement as such but in his evidence he made his position clear:

"Q. This is a product -- we will come on to this -- that you say, the Pozament render, has a high shrinkage factor?

A. Yes, it does.

Q. And it is going to have a surface coating which is going to be steel-trowelled?

A. Yes, it has.

Q. And everybody knows it is more liable to craze?

A. Yes, it will.

Q. And therefore, it would follow, would it not, Dr Leek, that the skilled applicator would have to be more careful to ensure that the product was properly cured, whether by natural conditions or covering?

A. He would have to have been aware of this particular product's idiosyncrasies in that respect, to treat it any differently from any other render that he would have been aware of."

72. In my judgment the sheeting served also to control curing. There was an absence of decisive evidence from Birse or anybody else about whether the sheeting had been used properly and whether humidity had been controlled.
73. As a general observation, a contractor's ordinary obligations to use workmanship and materials of good quality are not in some way diluted or come to a standstill when the works include special or specialist work and materials. A contractor who undertakes such works is expected to use the workmanship and materials commensurate with the specialist work or materials. To say "but I'm only a general contractor" may provide an explanation but it is not an excuse or defence. The contractor has contracted for the result which must therefore be delivered. If the contractor does not know how to achieve that result or is not prepared to accept the risks involved then he should not have tendered for or accepted the contract. This applies just as much and perhaps more so to work or materials which of their nature will be carried out by others. Indeed, given that on virtually all building contracts of any magnitude, the role of the main contractor is now that of a manager and is no longer the employer of skilled crafts and tradesmen, an employer, in using traditional language, is now to be taken as relying on the contractor to use his managerial know-how not only to procure the requisite skills but also to know whether and to what extent they are being provided adequately to met the requirements of the contract. This is well illustrated by Birse's action in suggesting that Pozament should be substituted for Fibrocem. Birse was here discharging its obligation to ET, but Birse remains as liable as if Fibrocem had been specified in the first place. Another aspect of the nature of a contractor's obligations was demonstrated in a discussion about a passage in Dr Sims' report where he said at paragraph 4.2.1:

"Dr Sims thus continues to believe that the GR18 render product exhibited a high potential for drying shrinkage, so that a competent contractor should have recognised the need for exemplary practices in its application, for the risks of crazing, cracking, curling and delamination to be minimised."

Dr Sims explained that, so far as he was concerned, exemplary and standard good practices were not necessarily different and that by ""exemplary" he meant doing it properly, to a high standard. He also said:

Q: You do not mean by "exemplary" that it has to be a very, very high standard.

A: I mean by "exemplary" that it has to be high standard.

Q: Higher than normal?

A: No. Render always has to be installed with high standards or it is likely to go wrong.

JUDGE HUMPHREY LLOYD: If by "normal", you mean the way some people do it and get away with it, then it is presumably higher than normal. But if by "normal" you mean doing it properly, then it is not really higher than normal.

A: Thank you, my Lord. I think that does clarify what I am trying to say. What normally happens is not necessarily what I am looking for. What should happen is what I am looking for and what should happen is a good standard of installation and that is all that is required in this case".

In my view Mr Leek's opinion did not materially differ. He was asked about Dr Sims' opinion:

Q. And you were in court and you heard him explaining what he meant by "exemplary", and he explained that he did not use that term in the sense of being higher than average standard or higher than normal but as meaning simply that the Pozament required ordinary good standards of workmanship.

A. Yes, I heard him say this.

Q. Now, I wonder if you could help us with your views in respect of that question of the standard of workmanship?

A. Yes, I believe that the Pozament render, because of its high shrinkage characteristics and the very high quality of surface that was required by the architect, if it was going to have any chance of being successful at all, which I still have my doubts about, it would need to be cured in a manner which was far better than one would normally expect to find on a building site in respect of render application.

I feel that normal standards for render curing would be really as set out in British Standard 5262 which is the standard for render which requires that the render surface does need to be protected or -- not protected -- needs to be prevented from drying too rapidly but only requires that that surface actually be protected under adverse climatic conditions of sun, wind, rain and frost, i.e. applying that under normal atmospheric conditions, then the render really does not require anything particularly being applied to the surface.

Consequently, in terms of this render's particular characteristics, it probably does require something additional in order to be able to meet any requirement of not realising its shrinkage potential."

Both therefore consider that the standard of "good workmanship" or "good practice" has to be set by reference to the work in question. Since the required finish of the render was clearly specified by MJP for the contract and since its characteristics were then ascertainable the means and risk of obtaining it were the responsibility of Birse.

74. Thus it was incumbent on Birse to see that the render was properly cured; this could be done by the use of the protective polythene sheeting. However, if Birse intended to let the render dry out naturally (whilst still preventing excessively rapid drying out), then it did not assess or monitor the conditions in which the render was drying out, as Mr Bowers said. However Dr. Leek, in his report and his oral evidence, said that there may be other causes of crazing, including "long term shrinkage" or carbonation. Dr Sims accepted that both are possibilities, but in reality the most likely reason would be the lack of control of curing, having regard to humidity, temperature and wind.
75. ET's case however assumes that cracking and crazing would not have happened. As will appear, I do not consider that to be correct. ET suggest that a cause may be a failure on the part of Birse to control curing. In my judgment Birse do not appear to have controlled curing but it is important that, even though it may not have taken overt positive and informed steps, nobody thought at the time that the provision of sheeting was insufficient for the purposes of curing (questions of protection against physical damage aside). Whatever conclusion may be reached about Mr Cox's recollection and attention after 1993, the evidence is that during the contract he kept an eye on the rendering, especially once crazing was found to be occurring, in order to make sure nothing was going wrong with the workmanship. The recurrence of the problem points away from lack of proper curing. In my judgment, even if Birse did not punctiliously comply with its obligations to control the curing, there is no evidence from which I can conclude that any such breach actually caused the crazing or cracking, given the other possible causes, notably the susceptibility of the render to behave that way when used externally. The Pozament render had a very high shrinkability, and an even higher shrinkability in the crisp surface zone than its data suggested, as Dr Sims acknowledged. It was to be applied with a steel trowel, against the advice of the British Standard, so as to produce laitance which is to be avoided as it is the main cause of crazing. Pozament is monolithic. Each coat is required to be applied much sooner after the previous coat than is usual. The manufacturer's instructions for the finishing coat of XF render did not call for curing protection, when in fact the render was more susceptible to shrinkage than normal sand-cement renders.

Overworking

76. The specification provided that the render was to be worked with a steel trowel to provide a smooth finish. Both involved bringing cement and water to the surface. In consequence, the surface is susceptible to drying out rapidly and crazing, as Dr Leek said in his report. ET therefore said that care should have been taken not to over work the render. In my judgment ET has not made out any case that Birse did not take such care. First, because of the susceptibility of the render to crazing, the very existence of crazing does not lead to the conclusion that there had been overworking. Secondly, there is no contemporaneous evidence of overworking. There is no evidence by petrographic examination or otherwise to demonstrate that Birse went over the boundary. At one point Dr Sims raised the possibility that as a result of using the wrong mix, a plasterer might have to overwork in order to achieve the smoothness required. There is no evidence to support this theory so I do not consider it a likely cause.
77. In October 1997 Mr G.C. Maden of Mott Mcdonald inspected a building at the John Moores University in Liverpool since it had been reported that crazing in the render (there "Duracote" had apparently been applied) had occurred. A report of the visit was included in revision B of the Mott McDonald report of November 1997. Dr Leek said that Mr Maden's instructions "were to look for areas of the type of defects that we had seen at the Cable & Wireless College; to look for cracking, any crazing or spalling, or incipient spalling, were it there…. to look at the condition of the render, identify defects, if any, that were present." It seems that, although the deterioration observed was similar, there was also cracking at the daywork joints between windows. I do not consider that any conclusions can be drawn from what little is known of that inspection or from its discussion by Dr Leek. It is a good illustration of the reasons why the probative value of hearsay evidence can be limited or non-existent. Although both reports from Dr Leek (the 1997 report and that produced for this case) said that crazing was "common", the documents relating to Mr. Maden's investigation not only do not support that conclusion but tend to show otherwise. The drawings record no incidence of crazing and there is only one photograph indicating crazing. ET also said that these records were consistent with what was recorded in one of Birse's internal memoranda dated 22 September 1995. That said:

"RE: SITE INSPECTION OF THE POZZAMENT FINISH AT THE LEARNING

RESOURCE CENTRE, LIVERPOOL POLYTECHNIC

Ken

Further to Stephen Cherry's continued comments regarding the quality of the Pozament Finish on the above mentioned building, please find enclosed photographs taken during my visit to inspect this quality, and my comments as follows:-

Front Elevation

With the exception of view 1, there was no sign of any cracking/crazing.

Side Elevation

Cracks appearing at views 2 to 5 inclusive, together with what appears to be some very poor remedial works to areas that have presumably cracked previously.

The remedial work looks like some kind of wash coat, which you can clearly see from the photographs, has not worked at all.

Again there is no sign of the fine crazing, that we are experiencing at C & W.

External Staircase Support Wall

One main fault line, running parallel with the lower landing level.

Rear Undercroft

- Soffit

1 No. fault central to soffit, running for approximately 6m. It looks in time, that this fault will continue along the whole of the rear elevation.

- Undercroft column

1 No. fault running full width of column.

- Undercroft end column

The render work is hollow, and extensive crazing has occurred.

The majority of the rear elevation is curtain walling, and access to the remaining side elevation was not possible at the time of my visit, although viewing from a distance the finish matched that of the front elevation.

Whilst the finish to the front and side elevations are an acceptable standard, problems similar to those experienced at Cable & Wireless are evident, (with the exception of the fine crazing).

To investigate this matter further, I feel we would need to undertake the following:

Obtain analyse and compare samples of Pozament from both the Cable & Wireless Buildings and the Resource Centre (causing disruption to both sets of clients).

Check on the background material at all locations, and check if any expamet back or similar was used on the Resources Centre.

Confirm that the curing time between the base and the topcoat was the same on both projects.

Confirm the condition at the time of the application at the Resources Centre (i.e. was the area fully sheeted, was there a controlled atmosphere during the application).

Obtain confirmation (if possible) from Pozament that the same material was used on both projects.

Please confirm if you want me to take this matter any further."

Ms Tooth told me about her visit to another building at the same university that was being repaired. She did not notice any cracks, and only inspected it from outside the site boundary. No conclusions could be drawn from that account. All this makes it clear to me that it would not be right to draw conclusions adverse to Birse without first-hand evidence which would demonstrate that the render in Liverpool has not crazed or cracked but was applied to the same specification as at the college and in the circumstances which require one to conclude that the college responsibility for a satisfactory finish lay not with the choice or the design but with the contractor.

Map cracking

78. The term "map cracking" or "cracking" describes deeper cracks with the appearance of "crazy paving". They indicated that the render has de-bonded. The experts agreed that the mechanism for map cracking is drying shrinkage "where the intended monolithic sequence of coating and bonding had not been successfully achieved". It was also agreed that the probable cause was inadequate bonding of layers of render. The samples tested show that Birse did not use a bonding agent. Birse accepted that map cracking would be caused by the long term shrinkage of the material, coupled with carbonation, but that, if crazing led to the cracking, then it was due to MJP's specification of an eggshell smooth steel-trowelled finish. I agree. I see no reason to hold Birse liable for the general crazing or cracking.

Cracking at stop and angle beads

79. ET maintain that the render applied over the stop and/or angle beads was not in full contact with the substrate so that the beads were not adequately restrained. This in turn would permit shrinkage movement of the render and would have caused cracking. As set out above, the specification (for the application of render generally) required render to be firmly bonded. The specification in relation to the fixing of the beads themselves (rather than the render applied to them) is at Specification page 1/37L and 1/38A. As seen by ET, the issue was whether both stop and angle beads were adequately bonded (as required by the specification and in accordance with good practice) and so as to restrain curling and cracking.
80. Birse's case was, essentially, that it was unlikely the render would adhere fully to a smooth metal surface which also has a different coefficient of expansion and there would be differential movement. There was "expamet" mesh on either side of each corner and next to a stop bead, the render had to go over the smooth nose of the bead. Dr Sims accepted that there was a possibility of cracking. Birse also maintained that the specification ought to have required mechanical fixings. ET accepted that Dr. Sims' view was that that would have been preferable. However ET also raised the adequacy of the bonding to the underlying blockwork (or other substrate). Mr Fleming explained the position in paragraph 7.10 of his report where he said that the reason the beads failed seemed to be because the render was not worked through the expanded metal mesh to make contact with the substrate so that the dabs alone were relied on to secure the beads. He explained that if render was worked through the expanded metal lathing on the beads, then it would adhere to the blockwork behind. Provided that the lathing was clear of the blockwork, there would be an homogeneous edge to the render. If however the beading was fixed only by dabs, then the only bond would be in their location.
81. Ms Tooth said there was cracking and curling at and parallel to stop beads and angle beads to render on blockwork. During the remedial works when the stop beads were removed, it was noted that they had not been fixed mechanically but had been located with dabs before application of the render. According to her, in all instances the beads had been inadequately installed in that the render was not in full contact with the blockwork behind. Cracking at and perpendicular to edge beads and angle beads to render on the blockwork was also associated with the stop beads not having been fixed mechanically but located with dabs. So too the render was not in full contact with the blockwork.
82. The materials experts however agreed;

"Whatever the required or permitted mode of fixing for the beads, it was agreed that, for success, either the bead detail needed to be securely anchored to the background, to provide independent support and restraint to the render, or the first coat of render needed to penetrate through the attached lathing and bond effectively with the background surface, or preferably both should be achieved".

In my judgment in so far as ET's criticism is of Birse's failure to fix the bead, this is in reality another aspect of MJP not seeing through the consequences of their choice of the render. Although contractors have a high degree of responsibility for applying the requisite standards of workmanship and good practice in order to implement a design successfully, I do not consider that Birse in this instance was obliged, in effect, to make good the absence of a detail that was necessary for the success of the application selected by MJP. Cracks at these locations are not, in my judgment, the product of poor workmanship in providing an inadequate bond to the substrate or the bead.

The application and use of Duracote

83. Dr Sims said, rightly, that this seemed to be a minor issue. Birse admits that it was in breach of contract is so far as it used (reformulated) Duracote with a high sulphate content which did not comply with the specification (as varied). However, Birse maintained that it was used in one place only, namely the west garden seat - total area of 9m². In addition Birse maintain that even if Duracote had not been used the render on the seat would have needed replacement because the design together with the use of stainless steel beads (see above) was defective as it was required to be taken over the top of the seat. It was therefore bound to crack. ET said that this type of Duracote was available from February 1995. It was therefore submitted that it was probably used by Birse in remedial works carried out after this date, not just on the west garden seat but also on the east garden seat and the gardener's store. Dr. Leek suggested that Birse would have used up stock before turning to the new Duracote – but, not surprisingly, there is no evidence one way or the other. Dr Sims had not conducted any analysis.
84. In my judgment I cannot infer that because this product was available from February Birse must have applied it to areas other than the west (or east) garden seat. In any event I am satisfied that even had Duracote not been used there would have been a further failure for which Birse is not liable. The claim for the west garden seat and the other areas fails.

The laundry wall

85. ET's pleaded case (paras. 28.24 to 28.26) alleges a failure by Birse in not taking steps to prevent shrinkage cracking caused by differential movement between different substrates. Ms Tooth said:

"There was severe cracking and associated debonding to the laundry wall, which I believed to be due to differential movement of the concrete and brickwork substrates I was advised by Gordon Cox were behind."

The experts' statement said:

"LT explained that, although the Laundry Room wall is a relatively small area, it had exhibited severe and extensive long straight cracking and debonding of the render. The background was a complex mixture of structural concrete, concrete blockwork and concrete wind-posts, with metal lathing over some of the concrete-blockwork junctions, which LT stated was fixed in too few places. She reported that the render cracking had mainly occurred over the areas of concrete wind posts. DL considered that, irrespective of fixity of the metal lathing, the complexity of construction and the large numbers of junctions between different substrates would lead to considerable differential movements between them, which would not be eliminated by the detail specified. MD noted from photographs that the concrete units appeared proud of the blockwork, so that added problems might be expected along the junctions, where the render would also need to change in thickness. The experts agreed that the rendering system in this location was the same as elsewhere (albeit covered), so that similar features might be expected, but with the highly varied background giving rise to the extensive cracking."

What was specified accords with BS 8000: 1989: Part 10, clause 2.2.2.3. Dr. Leek, however, considers that the detail would have been inadequate. ET's submission was that, given the quality of Birse's work elsewhere Birse must have been at fault. Birse's case was that Mr. Fleming had seen no evidence of fault; that the design was flawed; that joints in the render would probably have avoided this problem (as agreed by Dr Sims); and that remedial work was inevitable.

86. In my judgment Birse is right. There is no evidence of any breach of contract on the part of Birse. In view of the other possible causes (which are highly plausible) it is not enough to infer that because elsewhere Birse's workmanship was poor, it was here the reason for the cracking. This complaint fails.

Remedial works - Sto acrylic

87. In all cases, the replacement render applied was a Sto polymer modified cementitious render. In all cases, the cementitious render (whether as existing or replaced) was then overcoated with a Sto acrylic render.
88. During the trial Birse applied for and obtained permission to amend its Reply (at para. 26.4A) so as to make it clear that it claimed that "by using Sto acrylic ET has achieved a superior finish and avoided the maintenance burden of cementitious render". Birse's case was that the Sto render was different in nature from a cementitious render, and far superior. After hacking off and re-rendering damaged areas, ET could have applied a polymer modified cementitious (i.e. high build) coating. Birse did not say that ET acted unreasonably, as given its expectations, it was not unreasonable to opt for a higher quality material in order to obtain the performance that MJP envisaged of Pozament. However it was contended that Birse should not be required to pay to enable ET to achieve a finish that is superior to the steel trowelled cementitious render specified. I deal with this question on the basis that my earlier conclusions may be found to be wrong.
89. ET did not accept, first, that it ought to have applied some polymer modified cementitious coating and, secondly, that the Sto render provided a superior finish and that it did so because it avoided a "maintenance burden". The relevant agreed statement between Ms Tooth, Dr Sims and Dr Leek was as follows:

"LT described the remedial solution that had been adopted by CTP. On the main buildings, only debonded render (as detected by sounding) and stop beads had been removed and made good, with any severe cracks being chased out and made good. The whole area had then been over-coated, using a Sto system that is capable of covering a varied background without betraying any of the underlying topography on the finished surface. LT added that the Sto system was aesthetically pleasing and minimised future maintenance beyond periodic washing. MD supported CTP's approach and confirmed that the Sto system was preferable to any paint-like solution. IS accepted that CTP had needed to select a solution that could be reliably expected from experience and reputation to be fully effective at restoring the originally intended quality and nature of finish. DL noted that the materials were a significant departure from those originally specified, as were many of the details incorporated.

DL, on the other hand, whilst agreeing that all the preparatory work had needed doing, was not convinced that over-coating with the comparatively expensive Sto system was justified; he thought that a high-build coating would have been adequate and should have been the subject of trials. LT maintained that such trials would have needed too long to prove the effectiveness or otherwise of an untried system, because, whilst junctions, major discontinuities and unevenness might grin through an inadequate system fairly early, reflective cracking would take longer to develop. DL remained of the view that a less expensive over-coating system should have been investigated, as the materials considered were neither untried nor would a long period have been required for any unevenness or discontinuities to be visible. On the basis of the information known to him, IS supported the decisions that LT had made in respect of the remedial work.

The experts agreed that, on current evidence, the remedial rendering appeared to have been successful."

It is clear that when Ms Tooth first chose Sto she did so in order to provide a reliable solution to the predicament facing ET. The reasonableness of that judgment is not in question in itself and I dismiss the suggestion that before the work came to be done (in 1999) it ought to have been reconsidered and trials carried out. Birse did not persuade me that there was some alternative cementitious coating (not specified) would have proved satisfactory had it been subjected to trials and that therefore the cost and time involved in such trials would have justified. Ms Tooth described its advantages in her statement:

"The chosen overcoating render was a Sto acrylic system, which had the advantages of providing a consistently flush finish over areas of new and repair render, and of having flexibility so that many cracks that would with a thin coating have had to be repaired could simply be overcoated, and movement joints were not required."

Her decision was supported by Dr Sims. It was correct and not unreasonable.

90. Secondly, Ms Tooth also said in her statement that Sto "had the advantage of having a self-finish, i.e. washing would be the only required form of maintenance for a significant period of time." I do not see that as such the Sto render was superior or that it was better in terms of maintenance. It is essentially the same as Pozament which (like Sto acrylic) is a self-finished render (i.e. unpainted) but which would have required washing from time to time – which, according to Dr Leek, might have been about every ten years. The literature for each says nothing about maintenance. As Dr Leek observed, the need for cleaning "would hardly be a strong selling point for the material". Dr. Leek knew no more about the maintenance requirements of Sto acrylic render and from his evidence in cross-examination I conclude that the only possible difference might have been that Pozament might have weathered like stone, as that is what Pozament said to Bellworth on 28 September 1993:

"Further to your request for maintenance details with reference to GR Render, we confirm that the material will weather in a similar fashion to natural stone and as such may require a colour restoration from time to time.

Cleaning is best carried out using a mild metasilicate detergent applied by spray and washed down with clean water. In the case of mould or fungal growth a proprietary fungicide should be applied according to the manufacturer's instructions, this however may be less durable.

Major damage will require reinstatement using GR Render 1 as a backing coat and GR Render FC to finish according to the specification from Pozament Limited."

It is therefore clear that Pozament might have required cleaning to restore its colour. I do not consider that I can conclude, by inference only, that because Sto was acrylic it would not have required periodical cleaning also. It appears that in each case true maintenance would be required perhaps after 15 or 20 years (see Dr Leek's report in 1997). It is clear that when Ms Tooth first chose Sto she did so in order to provide a reliable solution to the predicament facing ET. The reasonableness of that judgment is not in question in itself and I dismiss the suggestion that before the work came to be done (in 1999) it ought to have been reconsidered and trials carried out. Looking at all the evidence I reject the case that Sto provided a superior finish or its use would have reduced the need for maintenance significantly.

4.B Armouralia

91. For certain external walls, MJP's design included the use of coloured sand/cement render (with a coloured finishing coat - red, blue and green) using a proprietary product called Armouralia. This was also referred to as "venetian" render. The use of Armouralia was instructed by MJP. ET complain about the following walls:

(1) the concrete wall defining the east boundary of the mast compound (green Armouralia);

(2) the concrete wing wall by the main north entrance doors to the administration block (green Armouralia);

(3) the concrete wall supporting the mast within the mast compound (red Armouralia).

The last had map cracking and the walls (1) and (2) had curling and cracking and render beads. Ms Tooth gave the following account in her statement:

"196. There are four external walls to which a specialist coloured render by Armouralia was applied. Three of these were 'mottled' green, being one of the walls defining the site entrance, and two walls defining the main entrance to the buildings, in the central area north of the administration block. The fourth wall with an Armouralia finish was the wall supporting the mast, adjacent to the comms. room, finished with a 'mottled' red.

197. When I first inspected the walls adjacent to the administration block, and the mast wall, they exhibited cracking and curling of render at angle beads similar to that observed with the Pozament white render.

198. The northern wall also had vertical cracks, which in my opinion were due to drying shrinkage of the render and/or concrete substrate. These cracks looked very unsightly because water had penetrated through the cracks from behind and had deposited a white leachate on the front face of the render.

199. The white render below main dpc level to the buildings on the site was vulnerable to debonding due to frost attack, and in the case of the green walls, a substantial section below what was the mock dpc level had fallen off the northern wall, inside the mast compound.

200. At the top of the northern end of the northern wall a portion of the render was missing. I was informed by Gordon Cox that this damage had been caused by Birse.

201. The remedial works to the north Armouralia green and red walls by Tellings comprised the following:

removal and replacement of curled cementitious render at angle beads;

replacement of cementitious render below mock dpc level;

overcoating of the render with Sto acrylic render;

provision of a new Armouralia finish."

92. Miss Jefford opened ET's case on the basis that MJP, as well as Birse, were liable for the state of the walls. The complaints in relation to matters of design were not pursued against Birse, namely that render was but ought not to have been applied below the level of the "mock DPC"; that no copings were provided to protect the heads of the Armouralia walls and that vertical cracking occurred in the northern wall because if the absence of movement joints. Against Birse there were two complaints: there was curling and cracking at the location of angle beads and stop beads and that there had been mechanical damage to part of an Armouralia wall during the course of the works which Birse had failed to remedy. The claim originally covered the cost of the works carried out and was for £34,225.09, including fees. It was claimed jointly against Birse and MJP. However, the settlement between ET and MJP included £4,654.61, i.e. the total cost of re-rendering the Armouralia below the plinth level. As noted above, the claim against Birse has therefore to be reduced to £29,570.48, inclusive of fees. Both complaints may be dealt quite shortly. Birse relied on the fact that ET apparently a 10 year guarantee from Armourcoat but had not had recourse to it so as to make it, in some way, "the author of its own loss" as Miss Jackson put it. The availability of another avenue of redress is irrelevant to Birse's potential liability to ET. It would be another matter if ET had clamed successfully under the guarantee and if Birse were now liable.

Angle Beads and Stop Beads

93. The first point relates to cracking and curling at the location of the angle beads occurred on the green Armouralia "northern" wall forming the eastern boundary of the mast compound, the wing wall by the entrance to the administration block, the site entrance wall and on the red Armouralia wall surrounding the mast itself and at the stop beads at low levels. ET's case on the first point is the same as its criticism of Birse in respect of the Pozament render, namely that the angle and stop beads were not properly bonded so as to restrain the render. However the Armouralia render was applied by Armourcoat , i.e. by specialists who manufacture and market Armouralia. Mr Bowers sent Mr Wiszniewski Armourcoat's specification and method statement:

"SPECIFICATION FOR RENDER BACKING COAT FOR ARMOURALIA (EXTERNAL) ON TO IN SITU CONCRETE BACKGROUND

(1) Shot blast/scale all concrete surfaces to remove any mould oil or contaminants and provide a mechanical key.

(2) Mix and apply bonding coat to the surface of the brickwork by brush stipple. Leave to dry for a minimum of 24 hours.

(3) Mix and apply slurry coat to the stipple finish ahead of the render application. The slurry coat must not be allowed to dry ahead of the render application, but applied wet on wet. Set stainless steel beads in position and apply scratch coat render at a thickness of 6/8mm. Lightly scratch surface and leave to cure overnight.

(4) Mix and apply slurry coat to the existing scratch coat ahead of the render application ensuring that the render is squeezed into the wet slurry coat. Rub up using a cross grained wooden float to provide a flat and true surface. Render must be kept wet for a minimum of 3 days by spraying with water and then allowed to cure for a further 28 days.

Material Specifications

Render 1 part O.P.C.

3 parts Medium Sharp Sand (Zone 2)

Gauged with S.B.R. 1 : 1 with clean water

Bonding/Stipple Coat 1 part O.P.C.

2 parts Medium Sharp Sand

1 part S.B.R. 1: 1 with water

Slurry Coat 1 part O.P.C.

1 part S.B.R. 1 : 1 with water

Coverage 10 m2 at 10 mm thickness will require 200 kg of dry mix

S.B.R. Usage 200 kg of dry mix will require 10 litres of S.B.R.

ARMOURALIA METHOD STATEMENT

INTERNAL

1) Shot blast/scabble all concrete surfaces to remove any mould oil or contaminants and provide a mechanical key.

2) Bridge over any dissimilar background materials using Expamet mesh prior to the application of Armourcoat Basecoat Plaster.

3) Apply bonding/stipple coat to all in situ or precast concrete substrates. Leave for 24 hours to dry.

4) Apply Armourcoat Basecoat Plaster to a thickness of 10 mm and leave with a flat and true surface finish.

5) Apply 1 mm layer of Ardex X7 plus Ardion 90 Acrylic Resin to all surfaces prior to the application of the Armouralia finish, and allow to dry overnight.

6) Mix and apply Armouralia plaster in 3 – 4 successive coats and polish the surface to achieve the appearance as agreed with the architects. Leave for 24 hours to dry.

7) Apply colourwash to the surface of the plaster and wipe off any excess. Leave overnight to dry.

8) Hand apply a wax polish to the surface and buff up using an electronic buffer.

EXTERNAL

Apply RZ.89805 Methacrylate coating to all surfaces and wipe off any excess. Leave to dry for 72 hours before exposure to elements.

ARMOURALIA WORKING CONDITIONS

Armouralia is not a conventional plaster relying on surface evaporation to firm up and enable the surface to be polished to a required level.

The unconventional working characteristics create the need for special working conditions in terms of temperature, humidity and draught control.

Temperature

The ideal working temperature for Armouralia is between 10 and 25 degrees C. At low temperatures there is insufficient surface evaporation to create a high polish on the surface. At high temperatures the rate of evaporation is too fast to enable anything but small areas to be worked effectively.

Humidity

High levels of humidity also result in insufficient surface evaporation to polish the surface of the plaster.

Draughts

Draughts or air flow can cause localised drying especially at the corners and edges. This will lead to problems with the uniformity of surface finish and therefore should be avoided if at all possible.

Lighting

In order to achieve high quality work to the desired surface finish it is vital that a high level of illumination is provided. The lighting should ideally be at an acute angle to the wall to highlight any unevenness or imperfections in the surface.

Scaffold

The scaffolding must provide clear and unimpeded access to all areas of the wall or surface to be plastered.

Comments on detailing to external concrete wall and tower

HORIZONTAL SURFACES

Rendering the top horizontal surfaces of the concrete wall and tower will provide poor weather protection. This may result in water penetration behind the vertical render faces with subsequent frost damage affecting the render adhesion. Further the absence of a well formed drip is likely to result in unsightly water runs and stains to the wall and tower faces. We would recommend the addition of a coping detail e.g. an aluminium clip over section to shed water off the top of the wall and tower. Such a detail should also include a well formed drip to throw water clear of the face (see attached copy).

JUNCTIONS WITH THE GROUND

Where it is proposed to take the render down to contact the finished ground level could also give problems. Permanent contact between the ground and render is likely to result in moisture penetrating behind the render. This will maintain these areas at a high moisture level. Subsequent frost action could then force the render off the wall in these areas. Further the render in the area close to the ground will be dirtied by splashes from the ground when it rains.

We would recommend that the render be terminated at a minimum 150mm above the finished ground level. The render should be taken to a stainless steel render bead. The 150mm strip will inevitably go earth grey in colour and we would suggest that the existing concrete be left exposed or be painted in a suitable "greyed" colour.

FIXINGS

It is quite common for renders to be damaged subsequent drilling to fix items such as ariels. Where possible these holes should be drilled in the substrate before rendering takes place. Removable plugs should be fitted to mask the holes during the rendering operation.

Armouralia's Working Conditions are important: "Armouralia is not a conventional plaster relying on surface evaporation to firm up and enable the surface to be polished to a required level. The unconventional working characteristics create the need for special working conditions in terms of temperature, humidity and draught control."

94. Dr. Sims and Dr. Leek agreed a statement which was less than decisive – it was the best that they could do. It included:

"Armouralia Product

It was agreed that the exterior version of the Armouralia product had been used at the College and that the system details were as described by DL in his report (paragraphs 22 to 24, 31 & 56) and summarised by IS in his supplementary report (sub-section 4.4.2).

The undercoats were similar to conventional rendering, apart from the addition of SBR (styrene-butadiene rubber) to the mix water, including a requirement for 3 days wet curing followed by 28 days drying. A proprietary, four-coat, polished and acrylic-sealed finish was the essential Armouralia product. Although the precise composition of the Armouralia material was not known, it was believed to comprise a mixture of Portland and high-alumina cements, plus crushed fine aggregates, pigments and chemical admixtures. Neither expert was aware of any allegations that the Armouralia product per se had been defective.

Nature of Alleged Defects

Both experts were uncertain as the full range of complaints concerning the Armouralia. It was understood that some of the defects alleged for the white render had generally not affected the Armouralia, including surface crazing, map cracking and delamination. It was thought that some surface crazing had occurred, but that this had been rectified to the architect's satisfaction at the time.

Complaints concerning Armouralia appeared to involve four issues:

- vertical cracking,

- cracking associated with beads,

- low level debonding, beneath the false damp-proof course (dpc),

- contemporaneous damage, allegedly improperly repaired (not here considered further).

Vertical Cracking

It was agreed that the Armouralia had been applied to cast-in-situ concrete walls and that such walls would be expected to develop vertical drying shrinkage cracking in the absence of any movement joints or cast-in crack-control features. The experts considered that the vertical cracking apparent in the Armouralia was probably reflecting such shrinkage cracks in the background concrete. The summary given in the IS supplementary report (sub-section 4.4.3) was endorsed by both experts at the meeting.

Cracking Associated with Beads

DL and IS re-confirmed the commentary on beads given in their first statement (section 8) and agreed that these principles applied similarly to the Armouralia render. DL stated that his agreement was based on a presumption, for this purpose, that the MacCormac Jamieson Prichard (MJP) specification applied to the Armouralia work.

Neither expert was sure of the mode of bead fixing that was required for the Armouralia by the specification, nor whether the MJP letter to Birse, dated 20/7/92 (calling for mechanical fixings), was intended to apply to Armouralia as well as the white Pozament render. It was thought to be possible that MJP regarded the details of the Armouralia installation to be wholly the responsibility of Armourcoat Limited (Armourcoat), who both manufacture and install the product.

It was agreed that all the available evidence indicates that the Armouralia beading was fixed with dabs, probably of cementitious undercoat render, rather than being mechanically fixed. It was noted that Roger Fletcher, in his witness statement for ET, states that some of the dabs in the Armouralia areas appeared to be pink plaster, but the experts agreed that some of the sands in the vicinity are reddish in colour (being derived from the New Red Sandstone formation) and can give rise to pink-coloured cementitious render.

It was also agreed that there is no evidence that any Armouralia dab fixings had failed, so that the use of mechanical fixings might not necessarily have prevented the occurrence of the cracking associated with beads.

Low Level Debonding

Both experts believed that some Armouralia render had detached from beneath the false dpc formed by two stop beads and assumed that, in these positions, the render had debonded from the lower bead detail with its associated metal lathing.

It was agreed that rendering beneath a dpc, or in this case beneath a false dpc but nevertheless still in close proximity to the ground, should not have been included in the design. Such detailing was in contravention of all authoritative advice, including the BS 5262 code of practice, and was also against the specific recommendations of Bickerdike Allen Partners, to whom MJP had turned for specialist advice."

95. Dr Sims could add nothing useful so his cross-examination was brief. ET's case was that because there had been curling and cracking the beads had not been adequately bedded or bonded. However the only evidence that ET could really obtain was Dr Leek's opinion. All that he said was that the cause of the cracking did not lie in the whether the beads had been fixed with dabs or mechanically fixed but in the extent to which there was a bond between the render and the lathing. Even so, all that he could say was that therein lay a possible cause of cracking.
96. Birse's case was that there was no evidence of bad workmanship. Miss Jackson accepted that there was evidence of parting at the stainless steel angle beads. However she submitted that there were only two conclusions: either that Birse had used another incompetent specialist subcontractor or that stainless steel beads were bound to cause these problems irrespective of workmanship. I agree with her both that the former is unlikely and that, as with the application of Pozament, the reason for the failure and for the need for remedial work lay in the details of MJP's design. Dr Leek was merely postulating a possible cause. There is no evidence of it. This criticism of Birse fails.

Damage to the northern wall

97. ET said that's case the northern part of the northern wall had been damaged during the course of the works, (i.e. before practical completion) possibly by an item of plant, and had not been repaired by Birse. As matter of common law and a contract with the JCT conditions form a contractor is responsible for such damage occurring before practical completion. Miss Jefford accepted that the evidence in support of this claim was slender. It was merely Mr Cox's account but via Ms Tooth. In paragraph 200 of her statement (see above) Ms Tooth set out what Mr Cox had told her. However Mr Cox in a statement said that he could "specifically recall this damage due to the period of time which has since elapsed, i.e. 10 years. However, I am confident that if Ms Tooth said that I had informed her of this at the time, then that was the case, and therefore that the damage would have existed". Birse adduced no evidence to the contrary so the issue is whether ET has proved its case.
98. Mr Cox's statement is typical of most of his evidence at the trial; he had very little recollection beyond reconstruction from the documents. In this instance although such an incident ought to have been recorded, it seems that neither he (nor any one else) did so. If so, and although a number of documents can no longer be traced, this could be an instance of the lack of attention that Mr Cox paid to keeping Birse up to the mark. However Miss Jefford drew attention to a letter from Armourcoat to Birse giving a price for repairing "the damage to the northern corner of the garden wall". In the circumstances Ms Tooth's account is in my view not only correct in itself but also correctly records Mr Cox's contemporaneous observation. I therefore find that the wall was damaged and that since it occurred before practical completion Birse is liable to make it good.

4.C Blanc de Bierges

99. MJP specified the use of Blanc de Bierges paving, both inside and outside the buildings. Blanc de Bierges paving is of Belgian origin. It is manufactured or distributed in this country by Milner Delvaux Ltd. ET's claim relates to its use outside. The paving is pre-cast concrete. The problems concern the paviours (slabs of 60cm x 60cm nominal dimensions), rather than the setts. I shall deal only with the slabs since the setts have not created problems and it appears that no material remedial works have been done to them. Blanc de Bierges products were chosen because the slabs have a hand applied textured surface and therefore look less "manufactured" in appearance. One side is smooth; the other is finished by hand; it is therefore rough and is the upper and visible side of the slab. (Professor Knapton for some time believed that the slabs had been laid upside down and this may have continued to have coloured his approach.) Blanc de Bierges products were used extensively throughout the college, not only paving slabs, setts and kerbs, but also for drainage channels and larger units to form steps and plinths. Blanc de Bierges paving slabs have been used for the central pedestrian areas at the 'eye', and for paths around and between the buildings. The teaching blocks are surrounded by a Blanc de Bierges 'plinth', level with the main paving. The leisure block has a Blanc de Bierges staircase approach, leading to a Blanc de Bierges paved terrace and the leisure block main entrance. There is also a Blanc de Bierges plinth at low level in front of the swimming pool, above the waterfall feature. Interspersed in the paving and the tarmac to the roads there are areas of Blanc de Bierge setts, which are approximately 150mm square x 80mm deep. Although there was a misunderstanding as to what documents were available when the work was done the following extracts from a guide issued by Blanc de Bierges for the purposes of maintenance contains some useful material:

"1. EXTERIOR

…..

Joints should be kept in good condition. Mortar joints installed correctly will normally stand up to intense mechanical cleaning operations. Cleaning with a high pressure jet is an ideal method of identifying faulty joints before proceeding with the repair.

…."

100. ET's case is that the joints between the slabs ought to have been filled, but were not properly filled, so the mortar between the joints has cracked and the slabs have lost adhesion. The issue is about Birse's workmanship in filling the joints.
101. The Blanc de Bierges paving slabs are unusual as unlike ordinary machine made slabs which are of uniform size, they are intentionally manufactured to differing dimensions. Although nominally they are 600mm square x 50mm thick, they may be + 10% from 600mm (but Professor Knapton said that such a degree of deviation was ridiculous). According to an interview with Mr Milner of Blanc de Bierges the actual average dimension of a slab is 59 cm and the deviation may be less – certainly Mr Milner said in March 1993: "providing that the units are laid within the Blanc de Bierges modulation which allows for a variety of joint widths, the question of product tolerances should not arise." The manufacturers use a variety of moulds so as to produce the effect that the slabs have been chosen at random. It might have been thought that, historically, stone cut for paving would have similar variations in dimensions, as Dr Sims did, but Professor Knapton, who had studied paving for over 30 years said that that was not so. He said:

"…….I have researched historic paving systems, principally because of the project that I had in Nîmes in southern France where we re-established the Roman square, and I found that the paviours there which the Romans used had very accurate dimensions, more so than ours. Pliny the Elder is the authority on this and he confirmed that the joint width is equal to or less than the width of the blade of a Roman soldier's dagger.

Throughout history, that has been maintained. If we go through the last two or three centuries the stone slabs which have commonly been used in, for example, churchyards were cut to very accurate dimensions, often to within one or two millimetres.

It is only within the 20th century that we have used concrete for slabs and indeed commonly since the Second World War and always they have been manufactured to accurate dimensions.

Also, in terms of the way that they have been installed, of course we did not have concrete at all until it was used commonly in paving in the last part of the 19th century, and we have always installed slabs on a dry material, whereas we have always installed setts on a wet material.

The reason for this is that setts replicate traditional bricklaying. Setts were introduced because they were brought back on ships as ballast, and they were roughly made because they were ballast and nothing else. For that reason, they could not be laid in the same way that slabs were laid and had to be laid in a more expensive wet mortar."

He later observed that irregularities at the edges that can be observed in ancient flags (such as in the Temple in London) is the result of the slabs being laid too close and spalling at the edges. That ought not to have happened as slabs should not butt each other.

102. Thus ordinary modern pre-cast concrete slabs, being of uniform size, can be and are laid so that the joints between them will be of uniform width. The dimensional variations of the Blanc de Bierges slabs mean that the joint widths are bound to vary along the lines of joints, and even along the length of a single slab. If the joint between two slabs is uniformly narrow it can be filled with sand. (Those less than 5mm can only be filled with sand, as a mortar will not physically go into a narrow joint.) If the joint is wide, it cannot have a dry filling (sand or dry mortar) because it will come out. Such joints must be filled with mortar, as Professor Knapton agreed. However a further feature of the Blanc de Bierges slabs is that the edge may not be at right angles so the joint may vary along its length. Herein lies some of the issue. Dr Sims suggested the joints should be made uniform by cutting and grinding them. That would of course remove the unevenness which is one of the reasons why they were chosen. On the other hand, one would have to start at the narrow end and, at some stage along the edge, to vary the pointing material. Professor Knapton did not dismiss this but he thought that it would be difficult and "experimental" in the sense that there was no established experience.
103. ET claims the cost of remedial work to the Blanc de Bierges paving. The total amount has been agreed: it is £166,046.30, including fees. Of that, £13,199.71 was claimed against MJP alone (including the fall to the Leisure Block steps) and £3,172.11 has not paid to the contractor for the remedial work, Tellings. The maximum claim against Birse is therefore £149,674.48.
104. MJP's requirements were not appropriate for Blanc de Bierges paving. The contract Bills had a standard specification for both bedding and jointing of the slabs, e.g.:

"Item 2/83/C: Sample of paving: Provide at site where directed a sample of the paving specified in Part 1 demonstrating the compacted subgrade, bedding and finished paving including any specified edging, of sufficient size to show the specified pattern and jointing.

Item 2/83/E: Laying pavings: Lay in dry bedding (e.g. sand) or mortar with joints and pattern as specified.

Item 2/83/F: Mortar bedding: For pavings bedded in mortar, mix the bedding to a consistency that brings no water to the surface, lay to correct levels and compact. Lay the pavings with uniform joints and firmly tamp into the bedding.

Item 2/84/B: Mortar and grout jointing generally: Comply with the following general requirements:

1 Do not leave joints open longer than necessary before jointing; if necessary cover open joints to exclude wind blown fines or other debris.

2 Dampen open joints immediately before jointing.

3 Keep jointing material off the face of pavings.

4 Clean off any accidental stains or, if not removable, replace the affected pavings.

5 As soon as jointing is completed cover the pavings with waterproof sheeting (e.g. polythene) and maintain in close contact and exclude all traffic until joints and bedding are fully cured.

Item 2/84/E: Jointing with mortar as the work proceeds: Fill and point the joints with the specified mortar as the pavings are laid.

Item 2/84/F: Jointing with mortar mixed dry: Lay the paving units with open joints of uniform specified width and pattern. When the bedding is set and the pavings and conditions are dry, mix the specified mortar materials without water and completely fill the joints by pouring or brushing in as appropriate. Brush away all surplus as soon as the joints are filled and moisten the joints with a fine spray.

Item 2/84/G: Grouting joints: Lay the paving units with open joints of uniform specified width and pattern. When the bedding is set fill the joints with the specified grout. Mix the grout with no more water than necessary to avoid subsequent joint shrinkage. Work the grout well into the joints until flush. Keep grout off the surface of the paving.

Item 2/85/A: Movement joints: Form clean dry rectangular movement joints by means of removable fillets. Seal as soon as fillets removed with jointing material as specified."

It seems to me that MJP and Northcrofts, in preparing the Bills, had used a standard specification (such as the NBS) although Mr Cherry could not recall if this was so. Dr Sims thought that the specification was vague but that may have been with the advantage of knowing how it should be drafted for Blanc de Bierges paving; it certainly did not conform to its requirements. Blanc de Bierges paving was described in the Bill of Quantities as, for example:

"Blanc de Bierges precast concrete paving components; laid on cement mortar (1:4) bed; flush surface jointing with 1:2/3 dry mix cement/sand brushed in 20mm deep all in accordance with manufacturer's specification."

MJP's drawings (CW 5040E (September 1991) and CW 5042A (July 1992)) specified a foundation of crushed rock but the Blanc de Bierges recommendations required concrete. Paragraph 2/84/F of the Bills (see above) called for placing on a compacted dry mortar whereas the Blanc de Bierges recommendations (of 1992) required a bedding mortar of plastic consistency and that each slab should be placed and pressed into the bedding material so that the material causes the material to fill the joints to 75% of the paviour thickness. They read, in part:

"2. BEDDING

The best retention will be achieved by using a "wet" mortar for bedding Blanc de Bierges components. The mortar should have good workability and adhesion, similar to that for laying bricks.

…

The components must be bedded in "wet" mortar for at least 75% of their thickness to achieve the best adhesion."

The remaining 25% was to be filled from the surface. Requiring the bedding mortar to rise 75% of the depth of the joint assists in filling of the joint. The manufacturer's instructions also provided for mortar and "dry mix" jointing. It seems however that the requirement for 75% came from the laying of setts and was transposed to slabs.

105. Although the precise chronology is unclear, at some stage MJP decided that the Blanc de Bierges recommendations should be followed so that the slabs were to be laid into a wet bedding mortar and embedded to 75% of their depth, with a semi-dry pointing mix applied to fill the joint. Contractually there appears now to be no issue. The description in the Bills for Blanc de Bierges paving said that Birse had to follow the manufacturer's specification. The general specification (as already noted above) said in item 2/5F:

"F Conflicting instructions: Should a manufacturer's instructions and this specification be a variance, seek instructions from the Architect/CA. Manufacturers' instructions for Workmanship shall take precedence over BS 8000."

Birse may or may not have sought instructions. If it did, it did not get them. ET accept that MJP did not give any formal Architect's Instruction varying the specification, or rather directing that its provisions had to cede place to the manufacturer's requirements. It is accepted that there is no such instruction in those terms or to that effect. It is however clear that MJP expected (at the least) Birse to adhere to the recommendations of Blanc de Bierges. (For what it was worth, that was also the understanding of both experts.) That was indeed its obligation under the contract. The opening to the Specification concerning the order of precedence of Parts 1 and 2 makes it clear the requirement in the description of the paving materials to follow the manufacturer's recommendations displaced the general specification. That is borne out what happened.

106. Birse used a sub-contractor, Eagle Paving (represented by Mr Stephen Jukes). (According to Mr Greenwood Birse later used its bricklayers to lay the paving.) It seems that in January 1993 Birse (or, rather, Eagle) had found difficulties with the paving slabs. The minutes of a Quality Meeting on 19 January 1993 record:

"BCL confirmed that the variation in Blanc de Bierge could be up to 60mm in some areas and an accumulative variation making the slabs 'un-linable' but as detailed and specified. MJP to instruct BCL on alternative method of setting out."

On 16 February 1993 Mr Whitehead on a CVI form said:

"Following the conversation between M. Greenwood and D. Whitehead 26.01.93 and 02.02.93 we confirm the following:-

1. The sample of "brushed in" pointing to Blanc de Bierge paving slabs and setts is acceptable. The pointing is to be kept flush with the top surface of the slab/sett.

2. The irregularities in the width of the joints between Blanc de Bierge paving setts due to the "hand- made" nature of the units are acceptable and will not require extra weeks to create regular joints."

Mr Cherry said that at this stage the reference to pointing being "brushed in" meant that Birse was pointing with a dry mix. The occasion was, unfortunately, not taken up with Mr Greenwood, or even Mr Cox, when they were called. The former (but not, I think, the latter) might possibly have been able to throw some light on an additional and not very legible note apparently related to the CVI. It seemed to me to show that MJP was not convinced that Birse would achieve an effective joint although a brushed in mix might be acceptable visually.

107. As set out above, paragraphs 2/84/F and 2/84/G of the project specification required that the joints should be of uniform thickness and that they should be filled by brushing or pouring the jointing material so that the whole joint depth is filled from the top and provided for two types of joint filling material - a mortar comprising the dry constituents to which water is added once the dry mixture has been introduced into the joints and a grout. Birse apparently used a grout with the approval of MJP (or lack of disapproval).
108. Mr Cherry sought advice from the Managing Director of Blanc de Bierges, Mr Kevin Milner. On 1 March 1993 he wrote to Mr Milner (the letter reveals that MJP did not know as much as one would have expected about the product that they had specified):

"Dear Kevin,

CABLE AND WIRELESS TELECOMMUNICATIONS COLLEGE

Works are well under way on site and before BCL commence the large and complicated areas, I would like to have a much clearer understanding of some of the issues from yourself, that are arising from site.

Tolerances: Kerbs

I have discussed on the telephone kerb tolerances with John Flowers, and he has visited site and spoken to our resident Architect, David Whitehead, to witness my concerns at first hand. It appears that the majority of kerbs are over 600mm (due to 'bellying' on the table) therefore causing continuous dimensional creep in setting out long runs. This is being compensated (unfortunately) by cutting kerbs in the selected areas. I understand John thought they were on the long side and I would appreciate your findings and explanation of this issue. As you may guess this will be the subject of a claim by Birse.

Tolerance: Slabs

On my visit to site last week I became aware of excessively large joints (45mm) due to two adjoining 590 x 590 slabs being undersized by 20 and 15mm respectively in one direction. We do appreciate that Blanc de Bierge is a natural product and the irregularities are part of its charm. However, seeing joint widths of 10mm and 45mm adjacent to each other causes me concern and we will be instructing BCL to remove these excessively undersized slabs to be cut or returned to Blanc de Bierge.

Tolerances: Generally

Your specification states tolerances of + 10%. I have been assured by Claire that that only applied to your smaller components (7 x 7cm and 14 x 14 cm setts). I am sure tolerances were discussed with BCL during one of our meetings on site and minutes, do you have copies of these? If so, I would be grateful for a copy or any relevant sections of your notes. As you must appreciate advertising your products as a modular system is not compatible with a 59 x 59 cm slab arriving + 59mm. This could result in a potential joint width of 128mm!

I need to set much firmer tolerance guidelines with BCL on site and I propose a maximum joint of 25mm and minimum 5mm with an optimum of 10mm for paving.

Would you please set out, to me, what your expected range of tolerances are for 59 x 59 cm slabs, bearing in mind that BCL are currently not size batching and therefore are adjoining a large slab with a small one i.e. are laying them in order of delivery and as they come off the pallet.

Tolerances: Plinth

With regard to all the above my concern is at its peak when contemplating excessive tolerances on the leisure and teaching plinths. As you know we have 89 x 44 cm sized components meeting three faced corner specials. The permutations of errors are enormous. We must ensure your products can be delivered within acceptable limits and therefore be able to cope with the concrete tolerances and fixing components. I do not believe we can accept component tolerances generally above + 5mm which was what I understood was discussed and agreed on site (hence rounding dimensions to the nearest 5mm). I understand that you are endeavouring to construct your moulds to achieve tighter tolerances. Please comment.

……..

Your prompt reply on all of the above would be appreciated, as I would like to establish a clearer understanding between ourselves before instructing BCL."

109. There was a meeting on 4 March 1993, attended by representatives of Blanc de Bierges, including its Managing Director, Mr Kevin Milner who gave advice. Mr Flower's note of the meeting was as follows:

"Site: Cable & Wireless Telecommunications College

Date: Thursday, 4 March 1993

Present: Mr David Whitehead – MacCormac Jamieson Prichard (Architect)

Mr Mike Greenwood – Birse Construction (Site Foreman)

Mr Steve Jukes – Eagle Paving (Sub-contractor)

Mr Kevin Milner – Blanc de Bierges (Managing Director)

Mr John Flower – Blanc de Bierges (National Sales Manager)

Later all (except Steve Jukes) reported to Mr Paddy Barber – Birse Construction (Site Agent)

Car Park – Walkways – Paving

An area featuring unjointed 60 x 60 x 5 cm slabs and 14 x 14 x 4 cm setts was inspected and the joints across the path width were found to vary between 1 cm and 4 cm at the extreme.

Joints running in the longitudinal direction of the path were nil to 1 cm.

The inside dimension between the kerbs measured 151 cm whereas the system modulation should be 150 cm. This accounted for 1 cm of the 3 cm oversize joints. The general opinion of all present was that this area of paving was not satisfactory due to the irregularity and excessive width of the central joint.

It was noted that the texture of all paving setts and slabs ran in the direction of the path.

Blanc de Bierges pointed out that they do not recommend the laying of materials with the surface texture running in one direction. Mr Jukes claimed that the direction of texture had been agreed at the outset. In retrospect, this had prevented him turning slabs in order to regulate the joint width and achieve a better fit.

The principles of Blanc de Bierges modulation were discussed in details then 3 offending slabs were lifted and turned through 90° with the result that the joints in both direction became acceptable.

It was agreed that this area of paving could be easily relayed to an acceptable standard providing the direction of surface texture was random. It was further agreed that future areas would be controlled within the Blanc de Bierges modulation without any selection or cutting of components.

Blanc de Bierges commented on the ease with which slabs were removed, demonstrating that they were not firmly bedded to 75% depth in accordance with our printed recommendations.

Mr Greenwood and Mr Jukes agreed to resolve this as it is important in a car park situation.

Kerbs

The kerb runs varied slightly from the Blanc de Bierges 60 cm modulation and could have been made tighter.

Mr Jukes agreed that from his previous experience of laying Blanc de Bierges components that it was quite normal for him to quickly tidy up the odd joint with a disc cutter prior to final jointing of kerbs, slabs, etc.

Jointing

An area of slabs was found to have 'white' mortar stains. Joints were said to be a mix of 3 parts sand to 1 of white cement. This material was brushed in but had not hardened even though they were said to be three weeks old.

Various jointing options and methods were discussed in detail with Blanc de Bierges emphasising that a compacted semi-dry joint is the preferred option for a slab as it is stronger and easiest to control during installation.

Mr Whitehead confirmed that he specifically wanted a dry stabilised sand joint of a similar colour to the paving. It was therefore agreed that the defective joints would be renewed with similar material at which time any prior stains would be removed with a brick cleaner.

An area of mortar jointing was seen to have 'marginal' staining around the faces of the setts due to the mortar being too wet. A sample of dryer mortar was proposed to Blanc de Bierges who agreed it would be more successful.

In order to help improve the jointing Blanc de Bierges promised to provide some 1cm pointing irons.

Setts into concrete

Mr Whitehead showed some setts which were bedded directly into concrete. This method was demonstrated to be totally unsatisfactory and contrary to Blanc de Bierges printed laying recommendations.

Blanc de Bierges stated that the problems encountered so far on site could have been prevented had trial panels been constructed prior to commencing the work.

Everyone present agreed that they had received copies of the Blanc de Bierges laying recommendations.

…….

At the close of this stage of the meeting, site personnel stated that they had a much better understanding of Blanc de Bierges and agreed that a hand-crafted system of products must not to be confused with general precast materials.

All except Steve Jukes then met Paddy Barber in his office and related to him the main points discussed on site. He agreed that a simple solution had been found to their problems and it was concluded that on occasions, joints may need trimming for aesthetic reasons although this should be unnecessary if they continually work to a controlled modulated grid.

In passing, Blanc de Bierges stated that, at the request of Birse, some panels were intentionally indented on the back to highlight the actual reinforcement position in order to assist fixing.

The meeting concluded with all present agreeing to collaborate more openly in order to achieve better quality and productivity.

Signed: J K Milner"

Mr Milner sent a copy of his report to Mr Barber of Birse. In his letter he said that he was "pleased that work was now proceeding according to plan" and that he hoped that Birse would be able to resolve their problems now that it had "a better understanding of the Blanc de Bierges system and its principles of modulation".

110. The note of the meeting accompanied a letter to Mr Cherry:

"Thank you for your letter of 1 March which arrived whilst I was in Brussels. Knowing your deep concern, I visited the site immediately on my return yesterday with John Fowler who had already tried to resolve the problem last week. I enclose the report of the meeting.

You can be assured that the difficulties experienced on site are due to a lack of understanding of Blanc de Bierges by those by those responsible for installing it.

Once we had demonstrated the principles of Blanc de Bierges modulation, your David Whitehead and Bierges foreman, Mike Greenwood, together with Steve Jukes (Eagle Paving) were all of the opinion that the material could be simply laid to everyone's satisfaction.

In particular, an area of paving using 14 x 14 x 4cm setts and 60 x 60 x 5cm slabs was presented to us with excessive joints in one direction and almost nil in the other.

The main cause of the problem was the contractor attempting to lay the materials with the 'grain' going in one direction which is contrary to the spirit of the material.

It was demonstrated that by simply rotating 3 slabs through 90º, the wide joints disappeared. The sub- contractor agreed that this is normal laying procedure for Blanc de Bierges but stated that he had been directed otherwise.

The kerb runs could have been laid tighter within the modulation and the contractor agreed that on occasions, it was usual to give the odd kerb a quick trim after bedding for aesthetic purposes.

All in all, providing that the units are laid within the Blanc de Bierges modulation which allows for a variety of joint widths, the question of product tolerances should not arise.

…..

Tolerances on Plinths

Please be assured that I have personally carried out a sample check on these units and confirm that they are within + ½ cm.

On the general question of tolerances for Blanc de Bierges special products, you rationalised your original mm dimension to the nearest cm. The re-dimensioned production drawings were checked by Birse and yourselves and we have worked accordingly. You should therefore expect few problems although even special components will still demonstrate the intentional variation associated with Blanc de Bierges manufacture but to a less degree.

Stephen, please be assured of our total commitment to helping you achieve your goal. Yesterday's meeting on site was most encouraging because all those present were willing to collaborate with us which is quite a change from past experiences with Birse.

Please let me know how I can be of further assistance. Meanwhile, John Flower promises to liaise closely with David on site."

The upshot of the meeting with Mr Milner was plainly that Birse were to follow the manufacturer's instructions. The exchange was subsequently treated by Birse as setting out the agreement as to method (see Birse's letter to MJP of 17 May 1993). It is also clear that those present for Birse and Eagle Paving accepted that what Mr Milner had said and showed was feasible. A Clerk of Works Direction dated 21 July 1993 and a later letter from Blanc de Bierges of 6 October 1993 (see below) both refer to Blanc de Bierges' specification and led to no reaction from Birse as inapplicable.

111. The notes refer to a jointing tool being provided to press the mortar mix into the joint. I was shown it (or one like it). It was about 10mm wide. I was invited to compare it with one bought from an ironmonger (for about £1.50) which was a little wider. Birse's witnesses, Mr Greenwood and Mr Bowers, gave evidence that the pointing tool could not be used satisfactorily, but only, obviously, where the joint was narrower than the tool. However Mr Greenwood also said that Eagle Paving tried to lay the slabs in accordance with the instructions of Blanc de Bierges but abandoned laying on a wet mortar and reverted to a semi-dry mortar bed. However he had no recollection of the meeting of 4 March so I do not think that much reliance should be placed on his evidence. Mr Barber said that, when he asked for instructions, he was told by Mr Cherry to "just get on with it", but when cross-examined about this, it was clear to me that any such remark (in itself quite understandable) either pre-dated the meeting of 4 March 1993 (when Mr Barber agreed that instructions were given) or, if later, was quite justified as by that time Birse knew what to do.
112. On 21 July 1993 Mr Cox issued the following Clerks of Works Direction:

"Blanc de Bierge. Pointing.

The standard of pointing for the Blanc de Bierge paving is totally unacceptable and is to be removed and replaced in the following areas:-

…."

This must have been prompted by one of the visits that Mr Flower of Blanc de Bierges had paid that day. His note of it is as follows (parts are not relevant):

"Re: Site Visit – 21 July 1993 – Cable and Wireless

This site visit was requested by Steven Cherry of MJP as he was concerned not only with the general workmanship of those laying Blanc de Bierges products but in particular, the vertical cladding to either side of the waterfall on the leisure building.

This area was inspected and it was discovered that the slab used for the vertical cladding had been constructed with the fine finish hidden. This obviously meant that the textured face of the slabs were sitting alongside the special 'L' plinth modules which all have a fine finish.

It was noted that the joints around the slabs in this area were not pointed using a special tool and the joint appeared very wide and fussy as the mortar had, in most cases, been spread over the edges of the slabs on either side.

This was also the case on the vertical cladding and the steps down to a plant room at the far end of the teaching wing.

This gave the effect that the very wide joints were due to irregular and undersized products which was certainly not the case and this could be seen when mortar was scraped from the surface to reveal the edge of the slabs.

Also in this area, the special component on the corner of the leisure building was seen to have a deep curved groove in it which had obviously been damaged after arrival on site.

During the tour of the site, I was approached by David Whitehead of MJP saying that Birse had insufficient jointing tools for their requirements. This was surprising considering at least 10 have been left on site and later I handed 3 more to Kevin Pepper of Birse.

During this meeting with Kevin and his replacement in 2-3 weeks, I reiterated the discussions with Steven Cherry as to the quality of the jointing and that jointing tools were not being used. It was also pointed out to Birse and Steven Cherry that the 14 x 14 x 8 cm setts being laid either side of the channel were being laid into a damp sand/ cement mix which was certainly not rising up 75% to the side of the setts but only 10-15mm! This would leave only the jointing mixture to keep the setts in place and was completely contrary to our laying recommendations.

It was also noted that the jointing mixture was a very wet bricklayers mortar with the effect that it was being spread over the surface of the setts again giving a fussy, unsightly joint. It was felt that if the two mixes were swapped, the contractors would be then laying closer to our recommendations and they would not only find it easier and quicker to lay but this would result in a much stronger construction which would also look acceptable.

Mr Barber agreed that by this stage the quality of the work was not dependent on an instruction from MJP. Mr Cherry also said that MJP was then concerned about the pointing because that was visible.

113. A month later, there was still no improvement. On 16 August Mr Prichard, when writing to Northcrofts to list defects said:

"Please add to your notes the following areas of recent remedial work:-

….

Externals: ….

• terrible pointing in Blanc de Bierge – remedial work required (see CofW direction)"

On 19 August, Mr Cherry carried out an inspection. His notes recorded his concern:

"Pointing of Blanc de Bierges has not improved one jot since CofW D No 125? Or 127. Not one correct tool was spotted throughout site inspection."

Mr Cherry said in evidence to me that he was there referring to the tool used to compact the pointing mix or the mortar mix between the slabs, and the fact that there was not one tool there meant that Birse was using probably the wrong mix or using a trowel or the edge of a trowel.

114. On 5 October Mr Milner visited the site again. He was not pleased with what he found. He sent a number of letters thereafter (having submitted drafts to Mr Cherry of MJP). To Mr Cherry he said (in the draft form dated 6 October):

"It was a pleasure meeting you again yesterday and congratulations to you and your team on the deign of such a magnificent piece of architecture. I am particularly impressed with your attention to detail which makes it the finest scheme I have ever had the pleasure of working on. It is therefore such a great shame that the installation of the Blanc de Bierges products is so poor that your client presently cannot enjoy the true crisp nature of this unique material.

It particularly worries me that the contractor blames their short comings on the material rather than the lack of skill and site control yet there are areas such as the area west of the dining room which demonstrates that a quality job can be done when they set their mind to it.

We at Blanc de Bierges although extremely proud that you should have chosen our products are horrified to see that much of the work is still badly laid in spite of our numerous site meetings with various members of Birse management and their sub- contractors.

I list below some of the typical faults that I found on site yesterday which demonstrates that Blanc de Bierges products have not been installed in accordance with our printed laying recommendations and numerous site demonstrations.

Bedding

Slabs and setts are either hollow underneath, rocking or in some cases, they can be lifted cleanly from their mortar bed. On detailed inspection, the bedding mortar can be easily flaked of the base of both the components and concrete sub- base. It was further evident that the bedding mortar failed to rise up the sides of the components to achieve a 75% penetration as detailed in our laying documents. In our experience, the three most common reasons for the lack of adhesion of mortar to concrete is:

the mix is too dry

the cement content is too low

dirt on the mating surfaces

or a combination of all three factors.

We believe that the structural success of any rigid paving scheme, whether Blanc de Bierges or any other paviour is used, lies on the ability of the wet bedding mortar to create a strong permanent bond between the sub- base, the component and its neighbours.

The fact that many areas of paving on this site fails to meet these standards worries me particularly as the Birse foreman, Steve Hellowell, stated that the mortar was often mixed in large batches which he said "sometimes dried out before being used".

Levels

In many cases, slabs and setts are poorly laid with considerable discrepancies from paviour to paviour giving an uneven surface which is abnormal. A flight of steps adjacent to the leisure building do not allow rain water to drain off the treads. In icy weather, this will present a considerable danger if left unresolved. It is worthy of note that workmen carrying out remedial work in the leisure building bar area were using a "bent" wooden level to control their work. They and Birse foreman, Tony Hudson, were strongly advised to throw it away and use an aluminium type which does not wrap.

Jointing

Most components are very badly jointed with mortar. The methods and results varied considerably with some joints being well below surface level but most being above and spreading over the edge leaving an extremely messy appearance. I have to say that this is about the worst I have ever seen.

It was very sad to see wet mortar being applied with a trowel in spite of our numerous instructions to insert a slightly damp mix with the aid of the pointing irons we have supplied for this purpose. On this point, Steve Hellowell agreed that whilst they had been received by his company he could not say why they were not being used.

Some joints are already beginning to crumble and in other cases, the jointing mix was only applied as a veneer to the top of the joint aperture leaving a cavity underneath. When challenged about these faults, Steve Hellowell stated that these were solely due to the irregular nature of Blanc de Bierges components. I then showed him the area west of the dining area which was laid by a bricklayer said to have had no previous experience of Blanc de Bierges. Steve agreed that the products displayed the same irregular characteristics as those presently being installed yet they were in the main well laid with neat crisp joints.

By comparison, adjacent slabs newly jointed that day looked an absolute mess thus demonstrating a lack of skill rather than product quality.

In my opinion, all these faults must be rectified but the methods to be adopted will very much depend on the nature and severity of the problem. I will of course be pleased to assist you but of course the responsibility lies solely with the people responsible for installation.

Site Control

There was various instances of cement stains which were separate to the jointing mortar stains previously referred to. In certain areas, I noticed small amounts of cement dust which I deemed to have been spilled onto the floor. It goes without saying that this material will solidify and adhere to the floor if not removed immediately.

Damaged Components

I noted a variety of chipped items which I believe has happened either during or after installation. I doubt if repairs are a practical solution.

…..

I have also written to Cable and Wireless director John Roberts, Bureau 4 director Tom Taylor, Sutcliffe general manager Julia Cardy and Birse contracts manager Ken
Hall who regrettably left the site early thus not benefiting from the afternoon's demonstrations.

I trust that my recommendations can be followed and I will personally maintain a close link with all those concerned in order to ensure that the Blanc de Bierges materials are maintained in good condition. However, given the large expanse of areas which have numerous faults, I will have to leave it to you and your client as to the extent of the remedial work. Until all this is done, you nor I will have achieved a top quality Blanc de Bierges installation of which we are both equally proud to present to Cable and Wireless.

Please let me know how I can help you further."

To Birse Mr Milner wrote on 6 October

"It was a pleasure meeting you yesterday and I trust that the cleaning demonstration for the hard landscaping proved that Blanc de Bierges is easily maintainable.

I wish to make it perfectly clear to you that the quality of installation of our products is far from satisfactory. It is sad that this should have happened in spite of our numerous meetings and demonstrations to various members of your management team and sub- contractors. A small area to the west of the dining room demonstrates quite clearly that a good job could be achieved and it is horrifying to see messy work currently being placed beside it. In short, the difference between chalk and cheese. Messy jointing is inexcusable especially as we have supplied numerous pointing irons to site. Steve Hellowell admits that he knows of their existence yet no- one was seen using them.

I had hoped that we could have met later in the day to discuss some of the finer points of installation, particularly bedding and jointing of slabs, together with cleaning procedures for internal surfaces. However, as you left the site before gaining the benefit of the afternoon's cleaning and pointing demonstration. I detail below some of the items which I am sure you would wish to have brought to your personal attention.

Bedding

Slabs and setts are either hollow underneath, rocking or in some cases, they can be lifted cleanly from their mortar bed. On detailed inspection, the bedding mortar can be easily flaked of the base of both the components and concrete sub- base. It was further evident that the bedding mortar failed to rise up the sides of the components to achieve a 75% penetration as detailed in our laying documents. In our experience, the three most common reasons for the lack of adhesion of mortar to concrete is:

the mix is too dry

the cement content is too low

dirt on the mating surfaces

or a combination of all three factors.

We believe that the structural success of any rigid paving scheme, whether Blanc de Bierges or any other paviour is used, lies on the ability of the wet bedding mortar to create a strong permanent bond between the sub- base, the component and its neighbours.

The fact that many areas of paving on this site fails to meet these standards gives me great concern, particularly as your foreman, Steve Hellowell has stated to me that the mortar was often mixed in large batches which dried out before being used.

Levels

In many cases, slabs are poorly laid with considerable discrepancies from slab to slab giving an uneven surface which is highly abnormal. Flight of steps adjacent to the leisure building have been badly installed as rain water fails to run off the treads. In icy weather, this will present a considerable danger if not resolved. It is worthy of note that your workmen carrying out remedial work in the leisure building bar area were using a "bent" wooden level to control their work. They and your foreman, Tony Hudson, were strongly advised to throw it away and use an aluminium type which does not warp.

Jointing

Joint widths vary considerably in some cases as the system's modulation has not been controlled. Most components are very badly jointed with mortar. The methods and results varied considerably with some joints being well below surface level but most being above and spreading over the edge leaving an extremely messy appearance. I have to say that this is about the worst I have ever seen. For examples, see the lake edge path, around the pool, teaching east plinth and numerous other areas.

It was very sad to see wet mortar being applied with a trowel in spite of our numerous requests to insert a slightly damp mix with the aid of the pointing irons we have supplied for this purpose. This is all clearly set out in our specification sheets supplied to yourselves with the first ten or so deliveries until we were instructed to stop sending them. On this point, Steve Hellowell agreed that whilst they had been received by his company he could not say why they were not being used.

Some joints are already beginning to crumble and in other cases, the jointing mix was only applied as a veneer to the top of the joint aperture leaving a cavity underneath. When challenged about these faults, Steve Hellowell stated that these were solely due to the irregular nature of Blanc de Bierges components. I then showed him the area west of the dining area which was laid by a bricklayer said to have had no previous experience of Blanc de Bierges. Steve agreed that the products displayed the same irregular characteristics as those presently being installed yet they were in the main well laid with neat crisp joints.

By comparison, adjacent slabs newly jointed that day looked an absolute mess thus demonstrating a lack of skill rather than product quality.

In my opinion, all these faults must be rectified but the methods to be adopted will very much depend on the nature and severity of the problem. I will be pleased to assist you but of course the responsibility lies solely with yourselves.

Site Control

There were various instances of cement stains which were separate to the jointing mortar stains previously referred to. In certain areas, I noticed small amounts of cement dust which I deemed to have been spilled onto the floor. It goes without saying that this material will solidify and adhere to the floor if not removed immediately.

Damaged Components

I noted a variety of chipped items which I believe has happened either during or after installation. I doubt if repairs are a practical solution.

…..

I recommend that extensive remedial work is implemented and the respective areas given a thorough cleaning using the methods detailed above (depending on internal or external locations) to ensure that your client does not experience areas of paving failing over the forth coming winter season. I am sure that the installation faults outlined will surface even though they are not readily visible at this time.

Should you wish to discuss this with me, please give me a call and I will be pleased to visit you on site at a mutually convenient time.

115. Yet, two days later, on 8 October, Mr Cherry acknowledged the letters (probably the drafts) by fax thus:

"Kevin,

Thank you for your letters and your kind comments. For your information BdB was laid yesterday with all the old problems. In this case I intercepted directly. None of the men laying the slabs had a clue how to do it correctly! – What will it take short of us doing it together.

…"

He then wrote himself to Birse on 11 October:

"For the attention of Mr Kevin Gander, Senior Project Manager

Dear Sirs

CABLE AND WIRELESS COLLEGE

Blanc de Bierges

On Tuesday 5 October 1993 Kevin Milner, Managing Director of Blanc de Bierges visited the site at B4's and our request to discuss the condition and maintenance of the Blanc de Bierges paving. In attendance at various times during the day were K Hall, R Green and T Hudson and S Hellowell of BCL, representatives from B4, Cable and Wireless and the writer from MJP. Discussions were also held with the paving sub- contractors (later confirmed as Parsons) and observed various operatives in action and laying slabs. Nothing stated or discussed was new or had not been previously communicated on numerous previous occasions on site. A summary of the discussions and conclusions were as follows:-

1. Bedding: The mortar used for laying slabs is too dry and possibly too weak. The consistency should be as for laying bricks. The result is a lack of adhesion and a full bedding is not achieved causing the slab to sound 'hollow' or setts to break free. On inspection of the removed mortar it was brittle and came cleanly away from the paviours and concrete flooring.

A number of points were raised:

• The mortar was too dry.

• Adjacent surfaces too dirty.

• Too large a batch of mortar was made up for an area to be laid and thus has dried out after a short period.

• The mix did not rise up the joints created.

• The slabs were not laid level.

• Bent wooden spirit levels were being used.

• The slabs have not been covered after laying allowing the mix to dry out too quickly in internal areas.

2. Pointing: The mortar used for pointing slabs is too wet. The consistency should be as 'damp soil'. The result is a 'buttered' effect smearing mortar over the adjoining slabs creating an unholy mess. There are many examples of bad workmanship across the whole site but only one area (laid by Dave Perkin formerly of KC) that is acceptable (west of dining room). A number of points were raised:

• The mortar was too wet

• The mortar was not being compacted fully into the joint (too deep because the laying mix did not rise up – see above)

• The correct pointing tools have not been used despite numerous free issues to site. The tool is nothing special, it allows full compaction of the joint and leaves a flat level surface or just below the level of the slab.

• Numerous areas of pointing has now cracked and is falling out of its joint.

• Areas have not been covered after pointing allowing the pointing to dry out too quickly especially in internal areas.

3. Cleanliness: Numerous slabs are covered in stains ranging from coffee to mastic. Although most organic stains can be washed clean other stains will require further treatment as follows:

• Mastic: Replace the slab.

• Mortar: hydrochloric acid based mortar cleaner – take care not to affect joints.

• Paint: Solvent, but may spread. Suggest slab is replaced.

• Rust: Oxalic acid – but may affect joints. Suggest slab is replaced.

• Mortar: Overspill from poor pointing is unknown. Attempts in the bar have been fairly successful but are more prominent was the slab has been cleaned. Trial areas should be attempted before addressing the external paving.

• As work proceeds: Very little attention is paid by your subcontractors to cleaning the slabs as laying and pointing proceeds. Simple washing of overspills as slabs are laid is a lot easier than remedial works later.

4. Damage: A number of paviours have been damaged or purposely cut to fit or resolve a setting out error. Making good to these slabs will not be acceptable. Systematic replacement of slabs should commence.

Despite the above being communicated to your colleagues, on Thursday 7 October we witnessed slabs being laid incorrectly. Since February this year we have repeatedly told all BCL supervisory staff our concerns regarding the poor quality of laying the paving and the fact that it did not comply with Blanc de Bierges' specification. Since February numerous BCL personnel have been directly or indirectly responsible for the hard paving on the external works i.e. Matt Brooks, Kevin Hudson, Mike Greenwood, Paddy Barber, Kevin Pepper (on and off), Steve Hellowell (most areas), Paul Seamer and not forgetting Steve Dukes of Eagle Paving. All except Paul Seamer at one stage or another have been explained and shown points 1 and 2 above. The installed results on site is a direct result of complete breakdown in BCL management and communications, lack of control of your subcontractors, failure to apply consistent and suitable labourers and the persistent excuses of blaming the product.

This issue is being taken seriously and if not adequately and promptly addressed by BCL we shall issue an AI instructing any outstanding work and all necessary remedial work to be completed by others.

We await your response."

I draw particular attention to the fact that Mr Cherry said of the meeting in October: "Nothing stated or discussed was new or had not been previously communicated on numerous previous occasions on site". In my view there is nothing in that letter which was not correct.

116. Mr Cherry included Blanc de Bierge as one of the key items that had to be sorted out in October 1993 before the Practical Completion Certificate could be issued. He did so by reference to the earlier documents:

"Blanc de Bierge - See MJP letter 11 Oct 93

See CofW direction No 125 dated 21 Jul 93 (confirmed on AI)

See BdB letter 6 Oct 93 (sent to John Beatson and John Roberts)"

It was not a long list, but as Mr Cherry said, "It was what we would call a snapshot of issues. We could not give a client a full list, because he could not comprehend it." It was some of the major points of concern in October 1993. Mr Cherry sent another note to Birse on 27 October:

"To BCL: Attn K Gander/N Wilson 27/10/93

Issues outstanding for the issue of a Practical Completion Certificate

….

2. Numerous subcontractor issues: for each item BCL to provide method statement for rectification (where all areas to be dealt with and prog and resources for the following items:

……

BdeB: Conversation with 2 Parsons men old KC men (here since last Jan 92) and 2 Eagle men (here since May 93) have not been told how to lay and point the BdeB!! Area east of east common room is good so is west of dining room (bar one or two areas). Mortar bed is still too dry and mortar for pointing too wet. Remedial where (i.e. areas to be addressed) bedding, how, by whom, how long will it all take.

…

4. Snagging: Definite statement from BCL required

- Which snag items are BCL NOT going to do and why etc

- All o/s items that are being addressed will be cleared by when?

- The clear up rate is not good enough and if areas are re-offered as complete they must be – otherwise Co/W will leave."

The purpose was to get as many items dealt with before the issue of the certificate of practical completion. However items relating to the Blanc de Bierges paving remained on the snagging lists.

117. In her principal witness statement Ms Tooth described at some length what she found in 1997. Her first observations were reinforced by her later investigations and observations. She said:

"I first examined the Blanc de Bierges defects in 1997 when I reviewed the miscellaneous defects on MJP's snagging lists. At that time I noticed that jointing mortar was cracking, and coming out of the joints in places. I removed some of the mortar from joints. I observed that it extended down no more that 20mm at most into the joint, and that there was a void in the joint below it. I was of the view that jointing mortar installed in this manner would not last long before it broke up and came out of the joints. I was also of the view that the installation of the jointing mortar in this way was defective workmanship."

Her observations were supported in detail by Mr Sanderson of her firm, as set out in his statement and in court when he said that he had removed pieces randomly across the site on about 20-30 occasions. The joints were generally largely found to be filled inadequately, i.e. they were not filled to their full depth and often only a small portion of the depth of the joint was filled. In 1999 Mr Priestley made a video recording of the areas of failure. The actual state of the joints as left by Birse cannot be challenged. The experts agreed in their statements first, that the joints should have been filled to their full open depth, making contact with any bedding in the bottom of the joint, but in actuality had commonly been incompletely filled with mortar; secondly, often, only the upper part of the open joint was filled with mortar and then only the upper part of that mortar was compacted; thirdly, however, it was also agreed that BdB is regarded by Architects as an aesthetically enhanced product, for which the deterioration of jointing material would be considered unacceptable; and, fourthly, that in at least some areas and possibly in most cases, the mortar material used for jointing did not completely fill the joint depth, leaving some void between the jointing material and the bedding material at the base of the joint; such jointing would be "more vulnerable to lateral movement of the slabs … than would have been a fully filled joint."

118. Remedial works was carried out to a specification of Ms Tooth. The mixes were specified because they were recommended by Blanc de Bierges. The bedding mix was specified as 1:4 cement: sand "Mixed to a plastic consistency close to that of masonry mortar. When squeezed in the palm of the hand, the ball should spread out without flowing". The jointing mortar was specified as 1:3 cement:sand: "Mixed to a damp earth consistency, considerably drier than that of the bedding mortar. When shaped in the palm of the hand, a ball of mortar should just hold firm". The remedial works contract was awarded to Tellings and commenced in November 1999. On 14 December 1999 Mr Priestley filmed the Blanc de Bierges materials throughout the site to provide a record of its condition at the commencement of the remedial works contract. The paving remedial works contract was completed in October 2000.
119. ET's case was that Birse ought to have followed the Blanc de Bierges specification of the jointing material (or bedding material) but that in any event it is good practice to fill the joints between paving slabs to their full depth. ET relied on what the experts agreed and what Dr Sims had said:

"Q. Indeed. All right, now a semi-dry bedding can be used to fill joints but if your bedding has only come a little way up the joint, say it has come up 20/25 per cent, you are then left with the other 75 per cent to fill from the top with this semi-dry pointing, are you not?

A. Yes.

Q. And that is a very tall order for the paving gang?

A. No, I would have said it is a fairly common order.

Q. The sides of the Blanc de Bierges are rough and are going to cause some friction?

A. You mean these sides?

Q. There is going to be some friction to force the semi-dry pointing down?

A. Yes, but nothing exceptional.

Q. And to try to fill a gap of that sort of depth is very difficult for a paver to achieve?

A. It is hard work but it is not unusual.

Q. What is more, he is not even going to know whether he has achieved it because it is very difficult to tell if you have got a semi-dry bedding down 5 centimetres.

A. I do not understand why. You ram to refusal.

Q. As best you can, there is the suction of the paving slabs?

A. I do not think that would have any effect.

Q. Right. As I understand it, any criticism that you make is of the method that was used by Birse and not of the craftsmanship of the individuals who were employed.

A. I was not on site at the time the work was done so I cannot criticise individuals. I can only criticise the result.

Q. All right. But the evidence that you have seen, leaving aside the splashes on the top surface which are not what was being addressed by the remedial works in general, the evidence that you have seen does not lead you to believe that the craftsmanship used by the men involved was in any way of bad quality?

A. If we are talking about jointing, it cannot be good quality if the joint filling does not reach to the bottom of the space, however deep the space. It is like filling a tooth. You expect your dentist to put the amalgam to the bottom of the hole in your tooth and not just skim across the top. He pushes it in until he cannot push any more in. It is exactly the same as what you do in filling the gaps between paving slabs, whatever paving slabs we are talking about.

Q. I have suggested to you the difficulty, particularly in narrow joints, of doing that with a semi-dry mortar.

A. It is not easy but it is done all the time."

Professor Knapton agreed that a fully-filled joint was the best thing to achieve. Mr Fleming's opinion was the same:

"Q. Then, at 7.6 over the page, you say: "Birse's work was less than satisfactory where they did not lay the paving units so the bedding material rose to 75 per cent of the units."

It is the words "less than satisfactory". What you mean there is that Birse's work did not comply with the Blanc de Bierges recommendations, do you?

A. I do. I also mean "and", the "and" being completely fill the joints between paving units with mortar because really that is the essence of it. This 75 per cent business is a figure. It is a nice figure to bandy about but the key to the performance of the units is the filling of the joint. If the mortar does not rise 75 per cent -- put it this way, the 75 per cent is pretty important if the units are at all in relation to their plan area because they can topple about. But if you are doing it with large flat slabs it is somewhat less important to hold them in place, to my mind, and I am speaking as an architect who has seen paving laid rather than as a materials scientist. It is really filling the joint that counts.

Q. So despite the wording here?

A. I have said "less than satisfactory". That is based on trying to follow the Blanc de Bierges recommendations.

Q. Yes. Because what this paragraph is really saying is if it is not 75 per cent up in the joint, it is not satisfactory.

A. No, if it is not filled it is the "and" that is quite important.

Q. Yes, it has to be both?

A. Well, no. If it is only 50 per cent rising up the side then the joint has to be that much deeper. If it is only 25 per cent up the side the joint has to be deeper still.

JUDGE HUMPHREY LLOYD: If it was 50 per cent but they filled the joint.

A. It would not be that important, they had still filled the joint.

JUDGE HUMPHREY LLOYD: Would you say the work was satisfactory or less than satisfactory?

A. I would say it was satisfactory if the joint was filled.

MISS JACKSON: All right, satisfactory, but it is not what Blanc de Bierges say you were supposed to be doing.

A. The 75 per cent figure is the figure from Blanc de Bierges that is not being achieved.

Q. So the satisfactory way of doing it is unlikely to be the way that Blanc de Bierges say to do it?

A. Well, Blanc de Bierges do not talk about leaving a gap between the surface pointing and the 75 per cent or less than 75 per cent.

Q. I am not talking about a gap either. You are saying it will be satisfactory as long as the bedding and the jointing meet, but if that is not 75 per cent of the way up, then it is not what Blanc de Bierges are telling the contractor to do.

A. If the bedding mortar has not squeezed 75 per cent of the way up, then it does not follow the recommendations which have been issued by Blanc de Bierges in their published technical literature. But that is not really the point; it is filling the joint that matters."

120. Birse's case was, essentially that the cause of the cracking lay in the bond (or the lack of it) between the bedding mortar and the slabs. The latter were laid with their smooth side down which was an unsatisfactory way of achieving a good bond. The slabs therefore did not have the necessary vertical or lateral stability and moved around. Furthermore, on Professor Knapton's evidence, it was impossible to have laid the slabs on a bed of fluid or wet mortar so that a mortar of the appropriate consistency could not have risen 75% of the way up the joint. The fact that the joints were not completely filled was a contributory factor. They could not have been so filled, given "the dimensional instability" of the slabs so there would always be differing joint sizes. Blanc de Bierges paving slabs are no better than any other paving slabs. Cracked pointing is normal but it looks worse with Blanc de Bierges slabs. That was not Birse's fault.
121. I do not accept that the effective cause of the cracks in the joints was the bedding material. That is not to say that the bedding is irrelevant. As Dr Sims said, the bedding and the jointing have to be considered as a system. I do accept that it is unlikely that it would have possible to have achieved the Blanc de Bierges recommendations such that the bedding mortar rose to 75% of the depth of the joint. However slabs were not laid on wet mortar but a dry or semi-dry mix. The result was that, as Professor Knapton said, had there been a bond between the wet mortar and the smooth side of the paving slabs, there would have been no mechanism that could have contributed to the premature failure of the joints between the slabs as it was only if the slabs were laid on a drier bed that there would be less lateral support to the slab. Thus whilst the effect of the slabs having a smooth underside cannot technically be discounted as a factor, Dr Sims must also have been right, in my judgment, when he said that it was a small factor. The fact remains that Blanc de Bierges required the joints to be filled. In my judgment that was feasible, as Dr Sims said, whose opinion I accept. It could well have been difficult, time-consuming and costly to Birse. That was Birse's risk and responsibility under the contract. It had to comply with its obligations and did not do so.
122. Professor Knapton set out a plausible analysis of the actual failures that occurred. He said, as he had agreed in the experts' statement, that the joints should have been filled to their full open depth but as they had not been, only the upper part of the open joint was filled with mortar and compacted. Thus the jointing material spanned the joint and thus could more easily deteriorate. Professor Knapton said if the joints were filled they would not deteriorate at the same rate or in the same manner as joints inadequately filled. If the slabs had been well bonded to the underlying material, then the jointing material would not have been as highly stressed, and the deterioration would have been much less severe, but in my view was liable to crack and did so because of the inadequate jointing material. If the bedding material had been the cause then there ought to have been further and persistent joint failure, following the remedial works. The slabs have been pressure washed twice since the remedial works. On the first occasion, some 18 months later, Mr Burns said that only 5% of the pointing failed and that the same occurred on the second occasion.
123. In my judgment Birse did not lay the slabs in accordance with the requirements of the contract, viz. in accordance with the manufacturer's instructions or with good workmanship and skill, either in not following the recommendations of Blanc de Bierge or in accordance with good practice. In particular it did not make sufficient effort to fill the joints properly to their full depth. It would have been apparent that the use of a mortar other than wet mortar would have meant that the jointing material would have had to have been placed to a greater depth, yet that was not done. It is not as if Birse did not know what was required. It was given clear and detailed instruction by Blanc de Bierges on more than one occasion and sought no further alternative instructions from MJP had it thought that it could not have done what was required. It was provided with suitable tools by Blanc de Bierges; yet it did not use them. It failed to appreciate the implications, in terms of the degree of care and application, of MJP's choice of materials. During the contract and the trial it maintained, wrongly, that it need only use the standards of workmanship and practice applicable to ordinary paving slabs (as Mr Whitehead said), whereas its obligation was to use the standard of skill required by the materials specified in the contract. It is a typical case of under-estimating the requirements of a contract. Birse is liable for the cost of the remedial work, i.e. £149,674.88.
124. The case against Birse was opened on the basis that it had failed to provide a fall of 2% to 3% on the steps of the Leisure Block and adjacent to the west entrance of the Administration Block so there was ponding. However this was not pursued against Birse (the amount claimed having been included in the settlement with MJP), rightly, since there was insufficient evidence for me to draw the conclusion sought by ET.

4.D External Walling And Coping Defects

125. This part of the case relates to ET's claim that there were defects in the construction of copings, cavity trays and dpcs in a number of places, including the outbuildings (gardener's store, the transformer room, the communications room and the sub-station/gas meter room) at high level; the residences at and above window heads; the leisure block at high and low level. As a result, water entered. ET's claim includes the cost of work to hack off and over-render external vertical rendered surfaces. The claim for dpc and cavity tray defects started in 1999 when Ms Tooth opened up the walls. It is not therefore about defects on the snagging lists in 1993 or 1996. However ET's claim in the action also includes some items in Schedule 4, as the cost of the remedial works covers items described only in Schedule 4, viz. 1/1, 1/2, 1/3, 6/3, 6/18, 6/25, 6/58, 6/59, 14/43, 14/49 and 14/50.
126. Birse make the point, with reason, that the remedial work is extensive. It accepts that it was reasonable, but only to make good deficiencies in the design or to improve it, aesthetically. In addition to work in the Leisure Block, at low level, and for the plant room bridge, work at coping level was carried out to 12 windows on the south elevation of the residences. The existing copings were removed, a layer of Kemperol membrane was placed over the top of the wall or cavity, and the copings were of course replaced. In the outbuildings the concrete copings were replaced by aluminium for aesthetic reasons. Elsewhere extension pieces were added to the aluminium copings to extend the drips. The cost has been agreed at £156,150.29 including fees, of which £38,372 was claimed against MJP alone, £10,025.36 against MJP and Birse jointly and £107,752.92 against Birse alone. As a result of the settlement with MJP, the total claimed against Birse is now £117,778.28, including fees.
127. I shall consider each affected structure in the order of ET's final submissions. Before doing so, I shall deal with some general points raised by Birse since Miss Jackson says that the complaints "need to be taken with a pinch of salt". First, Birse had carried out remedial work to eliminate all problems of damp found before it left the site in 1996. She says that with the exception of some leaks at 12 windows in the south face of the residences (which were not even mentioned in the original Defence and Counterclaim but which, in my view, should not have been present at all), water was not entering any of the buildings in 1999. Secondly, if the dpcs and trays were riddled with defects as alleged, Miss Jackson said it defied belief both that such widespread and numerous defects could have escaped the vigilance of the Clerks of Works (and of the architects on their visits during construction) and that they had not led to water penetration. (Nevertheless Birse admits that it was in breach of contract as regards bad workmanship at low level in the leisure block, although Birse referred to "some unusually difficult detailing" by MJP, a reservation that is irrelevant.)
128. In my view, the second submission is relevant only to a decision as to whether, as a matter of fact, Birse's work is or was so extensively defective as to justify the remedial work. Whether a clerk of works did his job well or badly does not affect a contractor's obligations under the building contract (unless of course the obligation was to do the work to the satisfaction of the clerk of works, which is not the case here, as clause 12 of the JCT conditions says that he acts only as an inspector). Birse relied on the fact that dpc and cavity defects had been picked up and that any inspector would have been ever more vigilant. Indeed Mr Cox took progress photographs and he ought not to have failed to notice if he was photographing dpcs with discontinuities. Mr Cox also said that he insisted that no dpcs or trays be covered up without his inspection.
129. The first submission requires consideration. Birse's case is that if there had been extensive defects by way of inadequate dpcs, dirty cavities etc, then signs of failure would appear early in the life of a building. Mr Woodward said in his Report and in cross-examination that if damage had not occurred within the first six years or after it has gone through a reasonable weather cycle of several years then it was unlikely that it would then occur. Mr Fleming provided another perspective. He said, in relation to pre-cast concrete copings, that with good workmanship the copings would keep water out for longer than with poor workmanship, and in relation to the aluminium copings at the college, although with copious amounts of sealant in the joints, water penetration would be prevented for a while, movement in the metal would shear the sealant and that would happen, irrespective of the workmanship, with the pre-cast concrete copings used on this job. Mr Fleming expected conventional details to behave in the normal way. Thus water penetration might be expected in due course even if the workmanship was good. In 1999 Mr Priestley found that the undersides of aluminium and concrete copings were wet, and referred to photographs which showed that the dpc had been punctured by Birse but the question remained as to whether any water had got into the building as indicated on Exhibit D1. Mr Woodward formed the opinion that there were only "isolated vagaries of construction", as set out in his Report:

"Thus, I have concluded that the main causation of any water penetration into the inner parts of the building would have primarily been due to the inadequate design requirements compounded, in places, by isolated vagaries of construction in the laps of the flexible dpcs and cavity trays."

130. A building owner is not entitled to expect perfection and has to accept work that does not comply with the contract where such work does not materially detract from the intended use and occupation of the building. An owner has to expect and accept unwanted "presents" from the builder, provided that they are not visible and not deleterious. What the eye does not see the heart should not grieve. To that extent only, what Mr Woodward termed "vagaries" are not significant breaches of contract. However it is clear that such instances were not isolated. Too many examples of poor and shoddy workmanship were found, some of which were known at the time to Birse's management. Mr Bowers was confronted with a striking example in the form of a letter of 16 July 1992 from KC Construction (London) Ltd, one of Birse's sub-contractors:

"We confirm your instruction to attend site at the weekend in order to carry out repairs to the Blockwork/Dpc to the openings of the Residence West North Elevation.

This clandestine operation is necessary as you do not wish to bring to the attention of the Client the extensive damage caused by your tradesmen to our dpc's at cill level.

Presumably you have in mind the consequences we encountered when we had a problem in adjusting a dpc 10mm, let alone rectifying damage.

Please note that to make good these walls to the quality required by the Specification we will be required to carry out extensive demolition of both Common & Forticrete Blockwork. This at low level in a confined area. As this work is far more involved than what we believe you envisage is necessary to be a good repair can you please issue to us a method statement of how you require the repairs to be executed.

Yours faithfully

K. Curran"

131. Birse also maintained that the results of the investigations carried out for ET did not justify the conclusion that Birse's work was suspect, tested by whether there was evidence of damage by water.
132. In her statement Ms Tooth described how the investigations came to be carried out. As regards the copings to the top of walls, and cavity trays and dpcs under the copings to the residences, leisure block, green armouralia walls and outbuildings, she said that during the course of preparing the specification for the render remedial works, she noted that MJP's detail for the aluminium copings to the top of rendered walls was inadequate as the overhang over the face of the render was only 25mm (whereas it should have been at least 40mm, to comply with good practice including BS 5628: Part 3: 1985 BS Code of Practice for Use of Masonry); the dpc under the coping was detailed to terminate at the junction between the blockwork and the render (instead of extending beyond the outer face of the render), as a result of which any water penetrating the joints between individual lengths of coping, and any water penetrating any defects in the sealant seal between the coping and the render, could get round and cause debonding and cracking of the render as a result of frost action; the design required the gap between the underside of the coping and the top of the walling to be sealed with a gun applied sealant, the maintenance of which would have required costly access scaffolding or other equipment that would not have been needed for any other maintenance required of similar frequency. She therefore specified for the removal of the copings, for the dpcs to be modified so as to extend beyond the outer face of the render, and for the copings to be replaced with matching polyester powder coated extension sections fitted to provide a proper overhang over the walling.
133. She said that that in the first area where the aluminium copings were removed (the nine rendered/glazed protruding pods to the south elevation of the residences) she saw that the joints between individual lengths of coping were not watertight, and that water was penetrating them. However she said that she did not necessarily consider this to be a defect, as a dpc or cavity tray is normally installed below copings to deal with such water penetration and that although MJP's drawing 8365 showed a cavity tray, the fact that a dpc had been installed was in itself of no consequence. What she was concerned about was that not only did the dpc under the coping not extend over the render (an inadequacy in the design) but the dpc was rucked, and discontinuous in places and rainwater could penetrate down into the walling below the dpc. She therefore instructed Kemperol to be applied. However she did not consider it necessary either to ask anybody at ET if there were signs of water penetration internally or even to look for them herself. At an experts' meeting it had been recorded:

"LT and Martin Dykes advised that there was little or no damage beneath the aluminium copings. However, it was agreed that the render system would be potentially vulnerable to any water penetration beneath the aluminium copings, in the same way as the concrete copings."

Whilst Ms Tooth had reservations about the phrasing she agreed she did not see any damage. Mr Fletcher said that the only damage was in the form of vertical cracks in the top of rendered panels in the pods, where MJP's design required the dpc to be penetrated by fixings for the angles to which the coping was attached. Ms Tooth was unaware of this cracking.

134. It was also not at all clear to me whether the south elevation of the residences was opened up because of the results of investigations of other copings or vice versa or whether that led to the opening up of the leisure block (as Mr Priestley said the opposite).
135. In my judgment there was no evidence of damage caused by any workmanship defects or any manifestations of any workmanship defects at coping level to justify the decision to open up and to investigate. The same applies to the administration block. Thus there was no support for the consequence pleaded against Birse:

"The installation of the dpc under the copings by Birse was defective and the dpc did not cover the head of the walling and the extent of the outer face of the blockwork. This allowed water to penetrate to the walling."

However on the evidence of those who carried a thorough investigation of the buildings (Ms Tooth, Mr Priestley, Mr Fleming and others) there is equally little doubt that the investigations did reveal an amount of shoddy work by Birse, which were not "isolated vagaries of construction", as Mr Woodward put it. In my view it does not matter since the question is not whether there were numerous and seemingly reprehensible instances of poor workmanship, but whether they caused water to enter the building and the need for remedial work. They clearly did not. Remedial work initiated by Ms Tooth to put right design defects was not caused by defects in workmanship even though the latter were put right as part of that scheme. Mr Woodward, in cross-examination, fairly recognised this. He said that once a coping had been taken off, where there had been a leak, if a number of construction defects were then found, then it would be prudent to carry out remedial works to make them good, irrespective of whether there had been water ingress or not. Ordinarily I would also conclude that the building owner might recover any additional cost necessitated by that further work where there were reasonable grounds for thinking that damage might result. In this case however I consider that Mr Woodward is right in his view that if no water damage had appeared in six to seven years it was not likely to occur and the very fact that Birse had had to be picked up during the contract for slovenly workmanship again cannot itself be a justification for the remedial work if there was no water damage or likelihood of it.

The Residences

The Pods/South Elevation

136. This part of ET's case has already been covered. There are nine pods protruding from the main walls to the south elevation of the residencies. Each has two side walls that are rendered. The rendered walls have aluminium copings with a dpc underneath the copings. The dpc had to protrude beyond the face of the render but MJP's design did not require the dpc or tray to be seen. It had to finish short of the external face, and a mastic seal was to be inserted. Miss Jefford accepted in her final submissions for ET that ET could not succeed in its original allegation that the dpc did not cover the head of the wall, since the dpc had only been designed to extend to the outer face of the blockwork and not to the render. However she maintained that the remainder of ET's case had been proved, namely that in many places the dpc was rucked or not properly bonded or was punctured "allowing water to penetrate the walling".
137. I do not consider that it was seriously challenged by Birse – nor could it be in the light of the evidence about Birse's workmanship overall, in particular the detailed photographs that were taken – that ET's investigators found many gaps; many holes and cuts; occasions where the dpc was not taken to reach the outer face of the blockwork; the application of the sealant was variable and often did not reach the dpc. For example, Mr Fletcher said that 90% of the dpc on the pods failed to reach the external face of the blockwork, and according to Ms Tooth the gap was up to 25mm . She said that the overall impression of the work on the dpcs was "appalling". Mr Fletcher said that it was "a disgrace". It was also reasonable, in my judgment, to assume that what was found would be present elsewhere.
138. I also accept ET's submission that these instances constituted a breach of contract by Birse. However I do not accept that they were the cause of the remedial work which was needed and being carried out to put right faults in MJP's design details. MJP's design led to the dpcs being punctured. The only damage (to the render) was where dpcs were punctured by screw fixings. No water penetration or render damage occurred due to any defects in workmanship deficiencies. That was the primary cause of the remedial work.
139. Furthermore, as already set out, I reject the submission that there was a sufficient risk of water penetration to justify remedial work. If the residences had not by then been damaged as a result of these defects they were not likely to be. Obviously once work was being carried out it was reasonable to cover Birse's defects, but the general application or overcoating of Kemperol would then be required also to put right MJP's deficiencies. In my view Mr Woodward rightly said that making good the deficiencies in the design of the dpc means overcoating to the back of the mastic on either side. Water may still penetrate the coping and will then be trapped by the mastic. As put to him, he saw that it would be necessary to cope with MJP's poor design of a mastic seal along the head of the wall under the coping to take the overcoating over the head of the wall. It seems to me that that analysis shows that the remedial work as executed was needed to replace the function of mastic seal (the physical replacement of which had its own difficulties) with "something that does not suffer from … design inadequacies", by taking the overcoating over the top of the render and turning it down the face of the render and thus avoiding the mastic seal and protecting the head of the render. Thus the problem of water coming through the head of the coping was dealt with. If I am wrong in this conclusion then it must follow that Birse would be liable for cost of applying Kemperol to cover the render at the head of the wall rather than that needed to put right MJP's negligence. Otherwise in my judgment Birse is not liable for the costs of the work to the pods.

The South Elevation Blockwork Walls

140. The south walls of the Residences were fair-faced blockwork. They are situated at a higher level than the pod copings and form a continuous parapet. At their heads there were cavity trays with aluminium copings installed. ET contend that the cavity trays were defective in that there were gaps in them adjacent to the movement joints, gaps between adjacent lengths of cavity tray, and no stop ends. Birse has made no admissions. Its earlier defence that part, at least, of the defects were attributable to MJP's requirements is apparently not pursued, rightly, as it was not tenable. I understood it to take the position that any defects were "isolated vagaries". Its main defence was that there was no damage.
141. In my judgment Birse is right in that defence. I accept that there were defects as contended by ET. Ms Tooth described the position in her statement. Investigations were carried out because of reports of water penetration of the residences' south elevation walling and because scaffolding was up for remedial works to the render, and because of what she said was the "the appalling workmanship observed with regard to the dpcs to the pods". On her inspection she found no stop ends at the movement joints. Mr Fletcher said in court that the only problem on the south elevation was gaps in the movement joints in the cavity trays. However the investigations did not reveal anything further. There was no water penetration through the inner leaves of the building. ET's claim in this respect fails.

The window openings in the south blockwork walls

142. Cavity trays were installed above and below the window openings to prevent water penetration. A concrete lintel is inset into the inner blockwork wall and protrudes across the cavity, above which there is the continuous cavity tray attached to the inner wall. The tray was to cross the cavity to the external blockwork wall, then pass through the wall to allow water to run off it to the exterior. The cavities were filled with blown mineral wool as insulation so it was important to avoid water affecting this insulating material. ET's case was that these cavity trays were defective as they were not of adequate length and did not extend across the whole length of the lintel, stop ends were inadequately formed or missing, and there was mortar and debris within the cavities.
143. In 2000 Ms Tooth investigated reports that, despite remedial works to the copings, water penetration was still occurring to south facing bedrooms on the first floor of the residences, at the sides of windows, beside the window head. Her investigations revealed that trays stopped short and had no stop ends and accounted for the water penetration. She said:

"The cavity tray should have extended the full width of the lintel, and should have been provided with stop ends. However, at the end inspected it stopped approximately 70mm short of the end of the concrete lintel, and it had no stop end."

Later:

"In Room 615 the lintel/cavity arrangement was the same. Here the cavity tray stopped approximately 130mm short of the end of the lintel inspected, and had no stop end."

There were similar reports for Rooms 415 and 312. (The middle digit denotes the floor; room 615 is block 6, floor 1, room 5.) Ties were also found covered with mortar bridging the cavity. Cavity trays below the window frames also did not have stop ends, so that water could make its way through the cavity but their absence does not appear have caused any damage.

144. Birse did not really challenge the evidence about mortar and debris in the cavity trays. Its case was that stop ends were not required until July 1992. On 17 June 1992 Mr Stephen Cherry of MJP carried out an inspection and made notes which included:

"1.6 All D.P.C's to be cleaned off, rubbish removed from underneath and damaged replaced.

….

1.20 No stop ends on D.P.C. trays under cills or at window heads. This is unacceptable and must be rectified. Refer to drawings for exact locations of end stops where required.

….

1.22 Cavities are full of rubbish i.e. plastic taps, rebar spacers, timber wedges, and timber cut ends. This is all impossible to pull through coring blocks. Resolve immediately."

That was followed on 18 June 1992 with a CVI of Mr Whitehead "Yes I said this":

"Stop ends to cill cavity trays.

All cill cavity trays to windows not yet built are to have stop ends on."

Mr Prichard had to write very firmly to Mr Cox on 26 June:

"Dear Gordon

…

Stephen and Dorian have been spending far more time on site inspecting work than I had ever expected. I hope you and Ken will be able to manage once the work becomes repetitive. You must let John or I know in advance if you need other help.

Please investigate the recent concern over the cavity wall work that Stephen discovered. This must be explained and you must keep a very close eye on it. Evidently Birse are using you and us as their Quality Control and this will happen on future trades no doubt.

You must pursue Birse to act on your instructions, if they don't, alert us immediately and chronicle this in your site reports. If you see particular, rectifiable, deficiencies in Birse's staff or subbies by number or quality please inform us immediately and chronicle them in your site reports.

Regrettably, the situation is not getting better so your constant efforts are needed and appreciated."

Then on 3 July 1992 AI 31 was issued:

"31.0 Works not in accordance with the Contract

In accordance with clause 8.4.1 and 8.5 of the contract we instruct you to remove the following work not in accordance with the contract and carry out all necessary reinstatement works and notify ourselves of your intended programme and method statements to carry out these works by return.

…

31.1.6 Cavity Debris – House types A-J inclusive

Cavities and expansion joints are full of debris, including concrete, timber, scaffolding accessories. All cavities are to be cleaned out and cleanliness maintained.

As recorded on site visit notes dated 10.3.92, 31.3.92, 22.4.92

31.1.7 Cavity Trays at cills – House types A-J inclusive

Trays have not been installed correctly at the jambs. The tray profile should penetrate jamb blockwork at least the width of the vertical d.p.c. with which it is lapped. This does not happen – it is not clear what does happen but trays have been cut and paths for damp are visible in some instances, as shown on drawing numbers MJP CW 8520, 8521, 8522, 8523, 8524, 8525

As recorded on site visit notes dated 20.5.92

31.1.8 Cavity widths – House type A – Check B-J inclusive

Cavity widths measure 95mm along large lengths of wall.

Cavity widths re to be 85mm. On some walls in excess of 100mm has been recorded (This may be due to defective blockwork). All cavities are to be surveyed and confirmed with CoW that they are in accordance with the specification and good workmanship. OAP have confirmed that cavities can be corrected in small increments over a whole storey height. No steps in cavities are permitted. All areas requiring corrective measures are to be noted and agreed with the CoW.

As recorded on site visit notes dated 17.6.92

31.1.9 Cavity Trays stop ends at head and cill – Houses types A-J inclusive

Cavity tray stop ends have been omitted at both heads and cills (they were specified at all head and some cills). They must be installed on all heads and on all second floor cills on south elevation. As shown on drawing numbers MJP CW 8520, 8521, 8522, 8523, 8524, 8525 and 8526

As recorded on site visit notes dated 17.6.92

….."

145. That AI led to correspondence with Birse. On 14 July 1992 Birse wrote to MJP

"We reply to your AI 31 with regard to the Residences, as follows:

…..

31.1.6 Cavities are being cleaned out and inspected prior to re-commencement of second floor facework, by your CoW as is good building practice.

…

31.1.7 The vertical section of the tray did in all cases penetrate the jamb blockwork at least the width of the d.p.c. with which it was lapped (as shown to your CoW, Mr Wilson.

The horizontal cill section of the d.p.c. was not being built into the blockwork, and it is not called up on your drawings for this to happen. However, we are undertaking these works in the spirit of co-operation with the design intent, and will request payment through the correct channels.

….

31.1.9 Where cavity tray stop ends have been omitted, they are being replaced. We note with some concern, however, their apparent lack of importance where they have been omitted by BCL from drawings in critical locations and await your instruction."

MJP replied on 21 July 1992:

"Further to your letter dated 14 July 1992 dealing with AI 31, we comment as follows to your responses.

…

31.1.4, 31.1.5 Your replies are noted.

& 31.1.6

31.1.7 The relevant drawings all note that the cavity tray is to pass into the jamb at least the width of the vertical d.p.c. This can only mean the whole of the cavity tray and not just some parts at your discretion. Consequently no extra payments will be made for your making good works. We would like to record that your original practice resulted in at least two visible instances where the cill closing block and the outer leaf were in contact with no d.p.c. between. We are confident therefore, that our instruction to remedy the work was wholly reasonable.

31.1.8 Your reply is noted.

31.1.9 In our judgement, cavity tray stop ends are absolutely essential at the head of all windows and are indicated on all drawings.

Cavity trays stop ends at cills, while desirable and depending on the detailing, are not absolutely essential in areas where we have not shown them. You have also been asked to install them on all second floor windows as yet unmade. Do you have any objections to this?"

Birse then wrote on 30 July 1992

"Further to your letter dated 21 July 1992 ref SC/EC/djw/5.1 we respond as follows:-

…

31.1.2 Your CoW has examined these DPC's with us and is of the opinion that they are satisfactory. Whilst we are not averse to performing all necessary works, we are loath to waste our resources performing work for the sake of it.

…

31.1.7 As previously stated, this work is being undertaken in a spirit of co-operation.

31.1.9 Cavity tray stop ends have always been put in at the head of all windows. These were, as you state, indicated on all drawings on elevations and sections.

Cill DPC stop ends, where indicated, are shown only on elevation, not in section. This change of notation is what caused some confusion.

We have no objection to installing all DPC's to windows not built prior to the receipt of all the instructions to do so."

Birse's case was that it was clear that cavity tray stop ends were only required by MJP for second floor window heads constructed after 21 July 1992 and that since the windows affected were on the first floor it was not liable. I do not read the correspondence in that way. In my judgment MJP's attitude and its AI (item 31.1.9) was clear: stop ends ought to have been provided; they were on the drawings; and, I add, might obviously be needed as a matter of good practice even if not shown on a drawing. However Birse's letter of 14 July endeavours to make the best of a bad job: Birse is plainly in no position generally to question what MJP has found but suggests that MJP has not shown them on their drawings and that they were therefore omitted. MJP's answer draws a distinction between those required for heads ("absolutely essential") and those required for cills 9"desirable and depending on detailing"). Birse's replies accept that stop ends at heads were required on the drawings and according to it, had been provided. But the evidence from the investigations is that they were not and that Birse's statement was not correct. Mr Bowers, although the nominal signatory of Birse's letters, could throw no light on what was almost certainly a routine exchange and not likely to be recalled by some one in his position. I have to conclude that stop ends were not provided at the cills and they ought to have been, either because they were shown on the drawings or because they were plainly needed (which I believe is also implicit in Birse's letters, as one would like to expect from a contractor of its standing). Birse is therefore liable to ET in this respect. Although there was no evidence of damage, I consider the absence of stop ends justifies remedial work

The north elevation blockwork walls

146. Although on this elevation, defects were found which were the same or similar to those found on the south elevation, i.e. missing and badly formed stop ends; splits; debonded laps and trays; and inadequately formed stop ends, falling short of cavity ends, there was no evidence of water penetration. Although it may therefore have been reasonable for an investigation to have been carried out the remedial work was not caused by any breach of contract on the part of Birse. The claim made under this head is not sustainable.

The Administration Block

147. ET's case in respect of the blockwork walls to the Administration Block was the same as its case in respect of the north elevations of the residences, as was Birse's. Ms Tooth said that she saw no evidence of any water penetration. There was no damage caused by Birse's work not being in accordance with the contract. The remedial work was not occasioned by anything done or not done by Birse.

Outbuildings

148. MJP's design for the outbuildings (the gardener's store, the transformer room, the communications room and the sub-station) was for single skin blockwork walls, rendered on the external face, topped with a dpc and a concrete coping. They thus differed from the other buildings. In addition on MJP's detail the dpc terminated at the outer face of the blockwork with a silicone seal at that point. It is common ground that MJP's design fails to protect the head of the rendered wall (see their drawing 5054A). Even if the joints between the copings were pointed properly water would eventually get through (as Mr Fleming said).
149. ET's case was based on the discovery by Ms Tooth that in the gas meter room in the sub-station the dpcs as installed by Birse did not extend to the outer face of the blockwork and that laps were insufficient and unsealed. These defects were found elsewhere. Mr Fletcher said that they were found in gardeners' store, the transformer room, and the communications room. Water had penetrated at the top under the copings at the juncture with the render. Ms Tooth said that cracking and debonding was particularly bad in the outbuildings at a high level, and, in her view, was probably caused by water penetration due to defects in the dpcs in the copings. Kemperol capping was installed throughout the lengths of walls and a new wider aluminium coping placed above.
150. ET's case is that the defects in Birse's workmanship entitled ET to carry out the remedial work. It is the same as its case on the pods on the residences. ET was entitled to carry out the remedial work and Birse is liable for the remedial work that involved taking the Kemperol over the head of the wall. Birse's case is also much the same. Miss Jackson submitted that MJP's design was bound to lead to water penetrating behind the render. Mr Prichard accepted that had Birse complied with the drawing then a dam would have been created which would trap all water penetrating the coping joints until the accumulated water found a path through and into the blockwork behind the render. If Birse had not achieved a seal throughout then water would pass between the blockwork and the render.
151. In my judgment Birse is correct. The defects discovered in the construction of the dpc might have been the occasion, as it were, for the remedial work but they were not its cause. It is clear that there had been or would have been damage to the render even if Birse had done its work correctly. The remedial work was required to correct an inherent defect of design. ET's claim against Birse fails

The leisure block

The leisure block at low level

152. Birse admit liability in respect of the remedial works carried out to (a) the west walls squash court/gymnasium; (b) the north and south re-entrant walls squash court/gymnasium; (c) the north wall squash court/south wall gymnasium. Birse do not otherwise admit liability.
153. ET's case relied on the evidence of Mr Priestley which could hardly be challenged. He painted a picture of sporadic defects (there were some places where there were none) but of quite extensive damage (insulation wet; laps not bonded in places; cavity trays bridged; wall ties bridged with mortar; water was dripping from the insulation material; cavity tray did not tie in with the tanking; problems with the tanking). In my judgment the nature and extent of the defects to the leisure block were such that it would have been reasonable to carry out remedial work beyond that required only to put right the identifiable defects. I shall however consider each section.

The leisure block spine wall

154. Mr Priestley set out what he found in his statement. He said:

"53. At the beginning of February I carried out an investigation to identify the cause of water penetration into the pool and snooker rooms to the north and south sides of the bar along a central, north/south spine wall. I instructed Roger Fletcher, and through him Tellings, to remove blocks for inspection externally from the cavity walling above the flat roof area. I noted that a cavity tray had been installed and there was evidence that a significant quantity of water had penetrated through the external walling down on to it. Once again, as with the lower level cavity tray, laps to this cavity tray were unbonded so that water on the top of the tray could penetrate through the laps and into the cavity below. I also observed a large split in the cavity tray where we had opened up.

54. Below the tray, the blockwork construction was saturated through both sides of the cavity and the steelwork within the cavity had water lying on top of it. There appeared to be a white fungus growing on top of the steel. I have made the comment above that I observed that the blockwork material chosen was particularly porous. This was very noticeable to me when we investigated this area. The spine wall was particularly exposed on both faces and at the time of my investigation there was a large amount of efflorescence on the external faces of this wall. I would not have expected to see such a large degree of efflorescence on blockwork of that age."

Ms Tooth said in her statement that the remedial work carried out was the same as that for the residences, i.e. Kemperol over the cavity trays, and the installation of extension drips to the copings.

155. Birse do not materially challenge what was found. Miss Jackson submitted that Birse could not say what isolated repairs might be appropriate and the question was whether the whole of the Kemperol installation to the spine wall was necessitated by design, workmanship or both. She contended that the dominant reason for the application of Kemperol was a defect in the design without which isolated repairs would have sufficed. She equated the position with the residences.
156. The same solution may have been applied but I am unable to see what defect in design was the cause of the remedial works. The evidence is clear: the remedial work in this instance was required to put right defects of workmanship. The evidence also points to the conclusion that they were not isolated, or certainly not so "isolated" as to warrant some other treatment. Birse is liable to ET for the relevant costs.

The blockwork walls to the squash court and fitness room

157. The issue is about whether ET can recover the cost of the general repairs. Birse has admitted liability for what it describes as "isolated repairs" to "isolated defects" (some seven or so missing or badly formed stop ends and an unsealed lap). I do not accept that the defects found could here be called "isolated defects". Mr Priestley said that every stop end was wrong. In my view the evidence justifies the conclusion that general repairs were here warranted as a result of Birse's errors.
158. Otherwise I accept Birse's case that it should here only be liable only for the costs of the remedial work which it has admitted. I cannot determine whether there were any more items. In view of the criticisms made of the design of the squash court it would not be right to hold Birse liable for any greater cost.

The plant room bridges

159. It is agreed that MJP's design was defective as it ought to have incorporated a waterproof membrane, instead of RIW, which was of no use there. Mr Fleming said that the stalactites underneath the bridges and the damage to the decorative coatings were the result of the lack of a waterproof membrane and of the way the bridge had been detailed. ET accept that MJP's negligent design was a cause of the carrying out of remedial works. Birse admits it is in breach of contract in the construction of the cavity tray and lead gutter. However it contends that the water penetration and damage and the remedial works to the bridge were due to MJP's design. In my judgment Mr Woodward and Mr Fleming were in agreement that the reason why the remedial works were carried out was to put right MJP's error and not Birse's mistakes. Accordingly, ET's claim against Birse fails.
160. It is convenient here to record a related claim for unremedied defects. In Schedule 5, Appendix 14 to the Counterclaim ET claim (item 1.4.1) for opening up and rectifying of 12 unidentified locations of water penetration and (items 1.4.3 and 1.4.4) for opening up and remedying further areas where water penetration has occurred and for providing Kemperol and extension drips (although extension drips were only the subject of a claim against MJP). These claims were not the subject of any expert agreement or evidence. They have never been investigated by ET, although they were found in June 2002. Mr Burns' statement says that there are 44 areas where he found water penetration. He said that he had recently carried out an extensive visual examination of the residences and found 28 residences with damp patches showing on the internal south wall; 1 residence with damp patches showing on the internal north wall; 15 residences with damp patches on the internal east wall. However when Mr Burns was called he had to correct his statement as the latter 15 were caused by leaks from heating pipes. Mr Fleming gave no opinion on them. Mr Woodward was surprised about the new leaks as Mr Burns had not mentioned then to him even they had been around the buildings together in order to show Mr Woodward any areas of damp. The evidence does not support any claim against Birse.

4.E Ceilings

Kitchen Grid

161. ET claim that in the kitchen the grid supporting the ceiling tiles ("Ecophon" tiles) was not properly installed by Birse. It was left bent. Prior to partial possession of the ground floor of the administration block (which took place on 6 September 1993) the defect was included in a snagging list of 24 August 1993 as "Perimeter frame to suspended ceiling buckled"; it was listed again in 1994; and in 1996. It does not appear that Birse questioned its inclusion, at least initially. It therefore appeared in Schedule 4 Item 2/9 (as it was based on the 1993 snagging lists). It is a mystery why it was then described as "Framing (PVC) to suspended ceiling has buckled, apparently due to heat". ET has replaced the ceiling. The amount is agreed: £4,200 plus fees = £5,544.00. ET found it difficult to clean the tiles so they had to go, according to Mr Burns. They could not be washed as they were loose under pressure It does not reflect well on ET that it did nothing for so long about the dirty tiles.
162. Birse deny liability, maintaining that the defect appeared after installation and was not attributable to a cause for which it was liable. However it did not advance any positive evidence beyond Mr Allsop saying that if there had been a defect on installation Birse would have put it right and to rely on Mr Cox's supposed vigilance to pick up a fault (which I discount heavily). Mr Woodward concluded that there was no evidence of distortion.
163. The space above the kitchen ceiling contains services. The tiles are likely to have been removed to get to the services. The tiles got dirty; they had to be removed; the grid might therefore have been damaged. When Ms Tooth first inspected the ceiling she found the grid to be bent and saw gaps between the tiles. Similar evidence was given by Mr Fletcher and Mr Priestley. Neither Ms Tooth nor Mr Priestley could say if the tiles had ever been clipped. If the grid had buckled, tiles would be displaced, but Mr Woodward accepted that there could be other reasons for the tiles being displaced. Mr Allsop recalled that the Altro flooring in the kitchen had to be replaced. In his statement he said that this in turn meant that Mr Cox would have checked every aspect of the area, as requested by him. Mr Allsop's statement was put to Mr Cox who agreed with it, but that carries little weight with me.
164. In my judgment the answer lies in the lists of 1993 and 1994. I conclude that Birse did not put right in 1993 the ceiling grid which was then bent. It was left like so many other defects and thus re-appeared on subsequent lists. Given that the tiles were not cleaned as it was difficult to do so, there is no reason why the ceiling grid would have been distorted from September 1993 and 1997. In 1995 Mr Cox was concerned with the floor and not the ceiling. The item had not been cleared from the snagging lists. There is no evidence that access to services bent the grid. Birse has simply used subsequent events (all of which could have accounted for the state of the grid had it first been observed in 1999) as a smoke screen to distract attention from its primary failure. It is liable for ET's remedial work.

Dampa ceilings

165. The Dampa ceilings are manufactured by Dampa Ltd. They are metal plank suspended ceilings formed from perforated planks with an acoustic inlay. They were used throughout the teaching and administration rooms in the teaching blocks. They are beneath the timber waveform roofs and accordingly share the waveform profile. As such, they are, as Mr Woodward said in his report, subject to movement. ET claims that they were not installed correctly in 13 places in the East and West Teaching Blocks. It has listed the damaged planks, based on a list prepared by Ms tooth who first inspected the ceilings in 1997. Remedial work was carried out between January and June 2000 by Deelys Ltd. The cost is agreed at £2,699.00 plus fees, i.e. £3,562.68. The defects appear in a specification, accompanied by photographs.
166. There are also 10 unrectified defects in Schedule 4 relating to Dampa ceilings which ET intends to put right (nos A1/ 169, A1/170, A1/180, A1/221, A1/235, A1/236, A1/238). In addition ET has listed 15 further unrectified defects in Schedule 4 relating to Dampa ceilings which it either does not intend to put right (nos A1/186, A1/188, A1/192, A1/220, A1/227, A1/238), or may not do so (nos A1/181, A1/187, A1/189, A1/190). At this stage I shall assume that ET will not put them right. (My earlier conclusions therefore apply.)
167. Birse deny liability primarily on the grounds that ET has not established that the ceilings were left by Birse in a damaged state. It pointed out that they are subject to movement and to damage: lights are recessed into the Dampa and damage can occur if there is alteration. In addition Birse say that because the panels are clipped into place, it would have been obvious if they had been inadequately fixed or misaligned. Some of the defects were first noticed in 1997, and some in 1999. Birse remedied all panels that were damaged or misplaced at the time of snagging. In respect of the west wing, a Certificate of Making Good Defects was issued. There was some general evidence from Mr Allsop about how he had pointed out and disclaimed liability for damage to the ceiling caused by the fit-out contractor (but this would have been in 1993/1994). He also said that maintenance staff had damaged panels changing light bulbs. Mr Cox agreed with these statements but that does not carry Birse anywhere. Mr Allsop remembered buying replacement planks on his own credit card. This evidence establishes that Birse did put right some of the defects; it does not however establish that the defects that then existed were put right.
168. I deal first with the defects that were put right. I shall refer to them simply by the numbers taken from the lists or schedules. Defect 1 is defect 4-23 in ET's Schedule 4. It was not noted in 1993 but apparently existed in 1994 (schedule 4; A1/174). Mr Woodward said that layout drawings indicated that a wall divider had been added between those rooms, and that the Dampa planks might have then been damaged. I have to say that this is a plausible reason for the damage which ought to have been noted when the original snagging lists were prepared. I do not consider that Birse is liable for this defect.
169. Defects 2 and 3 are said to have existed in 1993. ET relied on a snagging sheet to show that the former existed in 1993 but otherwise neither defect appeared on the 1993 or 1996 defects lists. There are further problems in reconciling the defect listed in 1993 with Defect 2 as the room numbers are not coincident. In my judgment these defects are typical in that if they had been outstanding they would have been listed at the time by MJP or a Clerk of Works. I am not satisfied that Birse left the planks in a damaged state and accordingly it is not liable.
170. Defect 4 was listed in 1993 and is no 5-169 in ET's Schedule 4. It was noted also in 1996. Birse suggest that looking at a photograph the defect could be the result of deflection causing the edge to spring free as it is a single plank which is said to have been inadequately fixed. That may be, but it was listed in 1993 and there is nothing to show it was then put right. In my judgment Birse is liable.
171. Defect 5 was not listed in Schedule 4. I do not consider that Birse left these planks misaligned. Had it done so, they would have been spotted before 1997. Birse is not liable.
172. Defects 6 and 7 are similar to defect 5; they were not noticed until 1999. However they were in Room W36 where Defect 8 was. That was noted in 1993 and 1996. Birse is not liable for the same reasons.
173. Defect 8 is not in the same category as Defect 4. Birse suggest that, since the defect is displacement due to the weight of a light, it might have recurred, as the light was cleaned or a bulb changed. The photographs show that the lights are located across two planks so changing the lights might affect them. In cross-examination Mr Woodward said that there was no other reason for access. As the defect was noted in 1993 but not thereafter until much later ET has failed to establish that it was not then put right by Birse. In my judgment Birse is not liable for it.
174. Defect 9 was not shown on any 1993 or 1996 defects list, other than one which was said to relate to a defect in Room E.34 of the East Teaching Wing. Defect 9 is in Rooms W2, W3 and W4 of the West Teaching Wing. In my judgment the answer is the same as that for Defect 3. Birse is not liable.
175. Defects 10 and 11 are in the same category as Defect 5 and for the same reason ET's claims are rejected. Birse is not liable.
176. Defect 12. According to Ms Tooth, the defect was that the edge to the Dampa plank was not straight adjacent to the south suspended ceiling fitting to Room W106. I find it difficult to accept that this could be caused by anything after 1993. As Ms Tooth said in cross-examination (albeit about the defects generally) there was no sign of work having been done in the rooms which might have accounted for the plank being bashed. In my judgment it existed, was not spotted and Birse is liable.

4.F Curtain Walling, Windows, Doors and Glass

177. ET suggested that its case could be set out in the schedule which Birse provided as part of its submissions. I have therefore done so [in a Schedule 1 which is not reproduced]. I have also adopted Birse's suggestions that I could include my decisions in it. The items do not warrant more extensive treatment. ET's claim relates to work carried out in 1999 and 2000 and its amount has been agreed at £2,359.00 plus fees = £3,113.88.

4.G Felt Roof Work, Lead Flashings and Zinc Roofing and Flashings

178. Birse made these items also the subject of a Schedule. There were 15 defects in the installation of felt roof work, lead flashings and zinc roofing and flashings. Birse accepts that 12 of these 15 defects are workmanship inadequacies. The items in issue are items 3, 11 and 12. They are the subject of a Schedule 2 [which is not reproduced] in which I reject Birse's case that its liability might be quantified on a pro rata basis at 12/15ths of the total remedial cost for the 15 items in that schedule which is agreed at £13,391.93 including fees. Birse is liable for every item.

4.H Paintwork to External Metalwork

179. ET claims that paintwork on external gates, balustrades and handrails (which were polyester powder coated steelwork) was damaged by Birse and was not adequately repaired. Ms. Tooth identified four major areas of defects. (They are also set out in the report of Mr. Fleming.) They comprise

Handrail upstands and fixings to all the terrace balustrades on the leisure block: a poor finish with paint flaking off balustrades, handrails and fixings;

Gates, ironmongery fixings, and the gate and enclosure panels to the compound at the east end of the East Teaching block: damage to gates and ironmongery with rust appearing throughout the metalwork;

The west end of Teaching West: paint finish to the balustrades and handrails to the balustrades at the top of the cable intake room chipped and flaking and the handrail rusting;

Handrail to the cable intake room means of escape to the west end of Teaching West: paint finish to the handrail badly chipped and flaking.

The costs of the remedial work was agreed at £5,405.00 plus fees, i.e. £8,070.37. However ET never paid retention of £135.13 so its loss cannot exceed £7,934.87.

180. ET (and its expert, Mr Fleming) suggested that the paint had not been properly applied but that was not the case which Birse had to meet. That case was that Birse was liable to make good damage to the paintwork which had been listed at the end of the defects liability period and which was still listed in 1996. Birse relied on the facts that all the areas affected are heavily trafficked and susceptible to damage. Mr Fletcher's description of the damage put right in 1999 was of paintwork that had been chipped by impact. The coating specified (polyester powder) was susceptible to damage, as Mr Fleming recognised, but he also drew attention to the description of flaking which he thought signified a lack of proper making good. There was evidence from Mr Woodward of the recurrence of damage even though ET had carried out repairs. Birse did not adduce any cogent evidence by way of defence but only general evidence of the making good from Mr Seamer and Mr Allsop.
181. But for the lists with paintwork requiring attention there would be no basis for concluding that Birse had failed to put the paintwork right. As paintwork is obviously liable to damage a contractor in the position of Birse would be entitled to jib at making good paintwork unless it was liable or unless it otherwise thought it sensible to do so, and would do so. That did not happen. My conclusion is that the paintwork was damaged within the period for which Birse was responsible and was not put right. Birse is therefore liable.

4.I Light Fittings and Other Electrical Items

182. ET complained about:

(1) Water penetration around external recessed light fittings, in particular, to the leisure block and green Armouralia walls.

(2) Fixings to the base of "Dalek" light fittings which were not secure.

(3) The armoured cable in the transformer room was not connected to the gland.

However the last was admitted by Birse. The cost of reconnection was said to be £50.

External Recessed Light Fittings

183. This claim relates to external recessed light fittings in the leisure block and the Armouralia walls which have been affected by water. Those in the leisure block have not been put right; those in the Armouralia walls have been remedied, according to Ms Tooth. The issue is what Birse did that was wrong. When cross-examined all Ms Tooth could say was:

"Q. The problem with them was that they were found to have water in them sometimes?

A. Yes.

Q. What was the remedial work done for those?

A. I do not know because that item was dealt with by an electrical contractor who investigated them, worked out what was wrong and rectified them.

Q. And he did not tell you?

A. No.

Q. So you do not know what was wrong with them?

A. No, I did not need to know.

Q. Because you were only interested in the easiest things to be remedied, is that right?

A. That is right. And these were things where we had to do it in an economic way."

In relation to the Armouralia walls, Mr Burns made a vague reference to the use of the wrong type of gland – Birse supposedly having not used one for external applications. However that was only what Mr Burns had heard. In my view if the defect had been put right there ought then to have been evidence of the defect in workmanship, or perhaps, of materials. Mr Woodward rightly observed if remedial works had been carried out, then there would have been an investigation as to the cause of any defect. In the absence of such evidence Birse cannot be held liable. The same applies to the fittings in the leisure block where there has not been any investigation at all.

The "Dalek" fittings

184. The "Dalek" fittings were lights in the car parking area. Not surprisingly they could be damaged by vehicles. The foundations of 16 or so were found to be loose in 1996 (there were many more of them). The question is what Birse did wrong. Ms Tooth could not say when asked; she had carried out no investigation into the work that Birse was to do under the contract. She said that Mr Priestley should be asked. He was, but he too knew nothing about the work that Birse was to have done. The fittings had been found to have moved so "a new concrete base of a sufficient size, weight and stability to prevent rocking of the light fitting was to be installed, and the light base was to be re-fixed using new stainless steel fixings" . It is not enough, in my judgment, in this instance to infer that because only some are affected and not the whole series they must have been badly installed in some way so as to become loose. ET has to show what the fault was. It could have done so but has not. Birse is not liable.

4.J Miscellaneous Remedied Defects

185. My decisions are set out in Schedule 3 [which is not reproduced]. The nature of the items do not justify further discussion or consideration.

4.K Lake Aeration System

186. The lake was not amongst the Contractor Designed items. Its construction was the subject of a provisional sum item 2/11/B in the Bills: "Construct water features as follows: - Lake to south side of west Teaching wing Channel 1 between Central Court and East Courtyard Waterfall to west end of Leisure building: £85,000." On 27 November 1991 MJP issued an AI:

"Confirmation of the following information regarding the water specialist installation issued on 25 November 1991

1. FOR INFORMATION a copy of Water Techniques revised quotation dated 18 November 1991.

2. FOR INFORMATION one copy of Water Techniques drawing number WT-PH-1476. (This drawing requires revising to exclude services and fittings relating to the filtration and chemical dosing system which is not included in the quotation dated 18 November 1991.)

3. The scope of Works will include the design of following water features.

(a) Waterfall adjacent to the leisure building curved 19m long and dropping approximately 1.5m.

(b) Water Channel 85m long along the Y axis in the east garden running under the main access road and linking via a "chaddar" to the central pool.

(c) Central courtyard pool, irregular in shape but with turbulent water features.

(d) Lake in front of west teaching wing containing approximately 1200m³ water to be aerated and circulated.

(e) Water display feature in a font within the Leisure Building.

The quotation includes the design of each water feature together with all relevant working drawings, supply of all necessary mechanical and electrical equipment, complete installation of all fittings including all connections to infrastructure and services."

(The relevant drawing was ultimately WT-PH-1476D.) There was also a later instruction of 28 February 1992 by which MJP formally instructed Birse to place an order with Water Techniques "for the design, supply and installation of the Water Features" in accordance with the quotation of 18 November 1991 for £73,650 in accordance with clause 19.2C.1 of the Conditions of Contract. The work was instructed against the bill item and under clause 13.3.1 of the contract conditions.

187. ET complains that there are no venturi pipes to complete the aeration system; that the pump is situated at the bottom of the lake; and that the electrical supply to the pump used an earth wire as a phase live wire (contrary to IEE Regulations). The remedial works cost £12,230.43 including fees. The point on the wiring appears not to be challenged. Mr Woodward said in paragraph 5.248 of his report and in court that he understood that Birse had conceded liability in respect of the incorrect wiring. Although there was no such concession I can see no reason why the point should not have been conceded so, in my judgment, Birse is liable for this item.
188. As to the other items, Birse submitted that it had complied with the instruction or instructions. The scheme the subject of MJP's instruction did not make provision for venturi intake pipes and provided for the pump to be bolted to a plinth at the bottom of the lake in accordance with that quotation and drawing. Water Techniques' quotation did not include for aerators. As they said to Birse on 9 August 1993, the water is simply re-circulated. Mr Woodward and Mr Fleming agreed that the Water Technologies scheme did not require venturi intakes.
189. This contract makes a clear distinction between items for which the contractor is responsible for the design (Contractor Designed items) and other works. The dividing line between design and mere installation or "workmanship" is not fixed and it can sometimes be thought to be fine. Matters of good practice and workmanship involve decisions as to what is best suited in all the circumstances and if, as I think, "design" is really little more than the result of one or more decisions to select or to specify then good practice and workmanship includes decisions which are part of a design in that they implement the basic design or the design so far. Accordingly it is not in my judgment an answer to say that design is not the responsibility of the contractor or that an instruction to give effect to a provisional sum could not render a contractor responsible for a matter of design. PC and provisional sums are used where the design has not been completely settled; they are frequently used for specialist work; such work very commonly includes elements of design; the effect of an instruction to use a provisional sum so that the subject of the instruction (work, materials, services, etc) becomes part of the Works under the contract whereby the contractor is as liable for it as if it had been specified and priced and included in the Contract Sum.
190. Thus I see no reason why in principle Birse should not be liable for the design of the work instructed by MJP. However in the context of this contract, with its clear distinction, that can only mean that Birse is liable for those elements of the design needed to achieve the work chosen by MJP. That did not include the provision of venturi intake pipes. Such pipes are desirable but they are not thought to be essential. They would therefore have had to be required. In this instance I can see no reason whatsoever to extend or import into the work the necessity to provide them as part of the design. Birse is not liable for the pumps.
191. Similarly, as agreed by the architectural experts, the work instructed by MJP was clear: the pump was to be fixed to the bed of the lake. In my judgment neither Birse nor Water Technologies was obliged to give ET anything else. There may well have been difficulties in inspecting and maintaining the pump without putting on a diver's suit or draining the lake, but that does mean that the pump was wrongly designed. Birse is not liable.

4.L Clerk of Works Costs

192. Included in the final page of Schedule 5, Appendix 14 to ET's Counterclaim, amongst other miscellaneous claims, was an enigmatic claim for £168,170.63 for "other costs". Birse took no point on how it was pleaded. ET's quantum expert, Mr Chelmick considered that the part was attributable to the costs of the Clerk of Works, of his secretary and of his portakabin, between October 1993 and August 1996. The following was agreed between Mr Chelmick and Mr Jewell, Birse's quantum expert on 17 October 2003:

"The figure for the Clerk of Works Team has been revised by Mr Chelmick to cover a period of 1^st April 1994 to 31^st July 1996, which is capable of arithmetical agreement. It should be noted however that the claimed cost is generally only supported by Cable and Wireless listing of figures, as shown in Appendix B to Mr Chelmick's supplementary report. Whilst the experts have no reason to believe that these costs may not have been incurred, there are no original documents disclosed to evidence the costs."

(ET said that supporting documentation had been disclosed but it required a special viewer and had not yet been inspected by Mr. Jewell.) The figure was now £98,836.44. Is Birse liable? If so, the actual amount will have to be agreed or decided.

193. ET's case was that by 31 March 1994, under the terms of the agreement of March 1994, Birse ought to have made good all the "snagging defects". It did not do so. Mr. Cox (the remaining Clerk of Works) had to stay on the site. Miss Jefford maintained that it was a foreseeable consequence of Birse's breach, and reasonable, that ET should retain Mr Cox to monitor progress and keep records (although I doubt whether ET got value for money). Alternatively, Birse is liable for failing to make good defects in accordance with clause 17.2 of the contract and the instructions to deal with defects listed between September and December 1994.
194. Miss Jackson submitted that Birse was not liable since it had substantially completed the snagging defects. ET released from the £470,00 held back: £435,000 on 25 April 1994, £8,130 on 30 August 1994 and a further £13,145 on 12 September 1994. It was not necessary to keep a Clerk of Works on site full time with a secretary and a portakabin until July 1996 because of defects valued by ET and their professionals as only £13,725. The defects liability period lists that were prepared and addressed towards the end of 1994 and into 1995 formed part of the agreement of March 1994 and did not arise from any breach of it, but from ET's own decision. In addition in a witness statement Mr Cox said that he was not authorised to sign off defects when they had been made good. ET had Mr Burns there to arrange access for remedial works. In addition Mr Cox also said that he was kept on full time for a year after Birse left site in 1996.
195. As a matter of principle a building contractor may be liable for the costs of supervising the execution of making good of defects. As a matter both of law and good construction practice, "practical completion" signifies completion of all but very minor items that do not affect in any material way the use or occupation for the works for the purposes for which they were intended. When an employer regains possession the contractor leaves. The employer does not expect nor is it obliged to allow the contractor to remain to finish what ought to have been completed. The risk and responsibility for the works transfers to the employer; obligations as to insurance change. Generally therefore an employer does not expect to retain a clerk of works (or resident architect). Should thereafter defects appear which require the employer to retain people to supervise or inspect the execution of remedial works by the contractor, or in its default, by others then the costs of such untoward supervision or inspection is in principle recoverable as damages from the contractor as a consequence of its breach of the building contract.
196. However on the facts of this case the costs of and associated with the retention of Mr Cox are not recoverable from Birse. First, it is clear that Mr Cox's duties included oversight of the fit out works so he was retained until March 1994, at least, for the convenience of ET. Secondly, it was clear prior to the agreement of March 1994 that Birse would have to carry out a significant amount of remedial work. The agreement was a settlement of the parties' claims at that date. If ET thought then that Mr Cox would continue to be required then it must be taken to have taken his costs in account. Thirdly, although Birse were then putting right work not in accordance with the contract, much of it resulted from ET's prior wish to have possession and was ordinary snagging that would not have justified the retention of the clerk of works. Fourthly, ET evidently thought that Birse had discharged its obligations reasonably promptly since it released £435,000 to be paid on 25 April 1994, £8,130 on 30 August 1994 and £13,145 on 12 September 1994. I accept Birse's submission that if the costs of putting right the defects outstanding after April were thought to be around £20,000 then ET was not justified in keeping a clerk of works on site full time with a secretary and a portakabin thereafter until July 1996. Fifthly, in so far as Mr Cox was required to produce defects liability period lists in that period then that was work that had to be done under the building contract and was not due to a breach of it for which Birse would ordinarily be liable and, in any event, if it had been thought to be Birse's liability then the costs formed part of the settlement. Sixthly, Mr Cox's presence does not appear to be caused by any breach by Birse but was for ET's own purposes: he was not authorised to sign off defects; Mr Burns could have arranged access; above all, he remained for a year after Birse left site.

5. Unremedied Defects

197. In view of my decision about the recoverability of damages for unremedied defects (whether ET intends to rectify them) it would not right or sensible to go through every one of the defects in Schedule 4 in these categories, even above £1000. I deal only with some of the items about which there was evidence and submissions. The parties should be able to forecast my likely conclusions on the remaining items. (I have already considered aspects of the Dampa ceilings.)

5.A Mosaic Tiling

198. MJP specified mosaic tiling for the showers and bathrooms. The mosaic tiles were small pieces of glass that came stuck to sheets of paper. The paper was on the front of the tiles and was to be removed after application. On 31 July 1992 Birse submitted a method statement to MJP

"WALL CLADDING

PREPARATION OF THE WALLS

The wall surface must be perfectly flat; even and dry.

SETTING PAPER MOUNTED MOSAIC

1. Make sure that room temperature is not below 10 degrees C or above 40 degrees C.

In the case of exteriors, avoid setting the mosaic at elevated temperatures and/or in winds that could dry the setting materials too quickly.

2. Using a 3 mm toothed trowel, evenly spread a layer of adhesive obtained by the mixing of a special white cement (Kerabond white or similar) with a latex rubber compound (Isolastic or similar) in the proportions of 3 Kg of cement with 1 Kg of latex for every 1 Sq. Mt of surface to be covered.

CAUTION!!! Never apply adhesive to the back of the sheet to avoid having it slide downward due to an excess of adhesive.

3. Do not apply the adhesive to an area greater than 5 or 6 sheets, about ½ Sq. Mt to avoid the formation of a film on the surface that could hamper the proper setting of the tiles.

4. Apply the sheets of mosaic with the paper facing toward you, securing them to the surface with a pad of wood or rubber, taking care to keep the print on the sheets facing in the same direction.

5. Next, level any possible differences in thickness caused by the adhesive with the same pad, controlling that the distance between sheet is equal to that between the tiles to avoid having the sheets show themselves, disrupting the uniformity of the covering.

CAUTION!!! Do not use the toothed trowel for this operation as it could damage the tiles.

6. As the first 5 or 6 sheets are set, before continuing to apply more adhesive and other sheets, begin to dampen the paper on the first sheets with a wet sponge. Continuing with the setting, return to dampen the paper on the sheets already set until it can be seen that the paper can be removed with ease.

7. Carefully begin the removal of the paper from the mosaic, parting from a high corner of the sheet and pulling downward in a diagonal motion to facilitate the removal.

It is possible from this point to adjust any tiles that present themselves irregularly with the edge of the trowel.

The time available for eventual adjustments of the sheets, at normal temperatures, is approximately one hour from the moment the adhesive was applied. At higher temperatures the available time is reduced by relative amounts.

SEALING THE JOINTS

The grouting of the joints must not be undertaken until 24 hours from the time the tiles were set.

CAUTION!!! Before beginning to grout the joints, clean and tile surfaces to remove any residue of glue that may have been left from the paper.

8. For the grouting a specific product is suggested (Keracolor or similar), in the measure of 1 Kg per square meter of surface to be covered to be spread with a rubber trowel vertically and horizontally to fill the joints and diagonally to remove the excess until all joints are filled.

For the grouting of the joints in swimming pools, baths, fountains etc, the product specified above (Keracolor or similar) should be mixed with a compound of a synthetic rubber base, (Fugolastic or similar) to the ratio of 2/3 of the former with 1/3 of the latter.

CAUTION!!! Clean the joints with a damp sponge removing the excess filler remaining on the tile surfaces, to avoid having the grout entrench itself in any microscopic crevices on the mosaic surface. Take care however not to remove the grout from the joints.

9. Once dry, clean the mosaic with water. If cement dust still remains on the tiles use water slightly acidified (with 5-10% commercial chloridic acid) then rinsing immediately.

CAUTION!!! "Vetricolor" or "Gemme" glass mosaic series are not suitable for heavy traffic floors, for which "Silex Mat" series is suggested.

IMPORTANT

For the setting and filling of the joints of the lighter colours, and particularly for the white mosaic, it is suggested that white cements or adhesives be used to avoid having the grey colour of the normal cement disturb or alter the colour of the mosaic."

Birse said that the paper obscured the work and that it made the job all the more difficult for Birse which had considerable difficulty in producing a result which satisfied MJP. Dr Fleming described the tiles as being appearing "very clean cut to the eye" and although "sharp" they were "not dangerously sharp in that they cut you if you lent against a shower wall", unlike other glass mosaics. However as the mosaic tiles were to be applied to a rounded shower enclosure then they could only be used as a surfacing ceramic material. In addition Dr Fleming pointed out that it would be difficult to get every single joint the exact same size at curves. He said that the essence of good mosaic work was to be able to adjust mosaics with a spatula as they were pulled off the paper. There were two ways of carrying out mosaic work: one was to grout the tile before putting it on the wall (which was quite often done with paper-faced mosaic); the other was to put the tile on the wall and to insert a spatula between the mosaics and move them around a bit to get them even. However careful adjustments were required to avoid sudden changes in the width of the joint, rather than a gradual change in width. All this is, in my judgment, of little or no consequence. Birse undertook to carry out this work; there was no reservation accompanying the method statement. Some of the mosaic work was good. Mr Cherry said that a specialist tradesman achieved an excellent finish in the swimming pool but was not seen again "and …the rest of the mosaics were of much lower quality and in some areas quite appalling".

199. Mr Cherry's view is borne out by a survey carried out by Mr Wilson, Clerk of Works on 21 July 1993. It stated:

"DEFECTS

RESIDENCE WEST – SURVEY OF MOSAICS

HOUSE A

GROUND FLOOR:

Flat 0.1 – surface of mosaics is not flat in some areas. Some joints are too wide. Grout gap top of mosaics near ceiling is approximately 8mm and too wide. In places to top of mosaics near ceiling is approximately 8 mm and too wide. In places joints are not correctly grouted or only half full. Mosaics to back of kerb are bulging and gap between kerb, tiles and mosaic is almost 10 cm wide and totally unacceptable. Mosaics curve out towards stop bead. Some tiles are broken. Area around wash hand basin – Mosaics are incomplete around shaver point and unacceptable. Mosaics near stop bead curve out. Mosaics near floor level are stepped and wavy and unacceptable. There are some joints too wide and there is a general problem with pieces broken off corners. Mosaics to right of wash hand basin where they go around on two faces, the courses do not match and are approximately half a course out.

Flat 0.2: - Left hand side of shower, the turquoise tiles are approximately half a course out to the grey tiles adjacent. Mosaic is missing near end of shower kerb. Some mosaics are not fitted correctly and stick out. Tiles are also out of course to right hand side of shower. Mosaic joints are too wide in places and tiles are uneven. Mosaics curve out near ceiling. Mosaics behind mixer valve plate are uneven and need re-setting around shower hose connector and back plate appears to be missing. Joints to mosaics are not fully grouted in places. In other places they are rough where grout is left on the face. Mosaic below kerb is uneven around radius and is curved from top to bottom. Tiling to kerb – grouting is dirty, untidy. Joints at ends, there is excessive grout in corners. Mosaics to wash hand basin area, joints at high level on left near mirror are wavy. Single line of mosaics to left hand side of high level cupboard, there is half a mosaic fixed and this throws all the mosaics have a course out. Generally perimeter joints are too wide for grout. This also applies all the way around the bottom of the mosaics against the vanity top. Face of mosaics is filthy with grout that has not been removed.

Flat 0.3 – Mosaic joints in places are too wide. Some joints are not correctly filled. Joints near ceilings are not grouted with epoxy grout, the adhesive is showing through the joints. Generally the top joint against the ceiling is ragged and dirty. The panel to the side of the cupboard at low level is very poorly fixed, is stepped and wavy and needs to be replaced. Some tiles are again broken and require replacement. Corners are over filled and not cleaned off and in a dirty condition. This also applies to a joint between mosaics and bath. Mosaic bath panel: The joints are stepped and uneven and this panel has not been grouted but there is grout smeared over the face and left in a filthy state.

Flat 0.4 – Bath panel: Mosaics have not been correctly and some joints are still empty. The mosaics are again stepped and grouted uneven. Joints around perimeter of panel have not been correctly filled. Panel to side of cupboard at the bottom, mosaics are wavy. Size of room is too large for bath so that mosaics do not fit over roll which could cause water and smell problems at a later date. Joint between mosaics and ceiling is uneven and large in places. Grout has not been cleared off face of mosaics at ceiling level and is ragged. Some mosaics are full of adhesive instead of grout. Mosaics curve out towards stop bead – again there are broken tiles in places. Some odd mosaics stick out.

Flat 0.5 – Shower kerb: Face of the mosaics is in a filthy state with grout smeared over the faces not cleaned off. A lot of the joints are filled with adhesive and not grout. Tiles to kerb are not correctly grouted, some tiles are stepped. Mosaics to back of kerb: mosaics are slanting out to the bottom, are curved and around the radius are curved and stepping. They are incorrectly grouted, there are gaps and a lot of joints filled with tile adhesive. The end of the partition to the side of the toilet is approximately 8mm out of square. Some joints are too wide. Mosaics around radial corners are wavy, joint between mosaic and ceiling is too large and mosaics uneven in corners in this area. Some mosaics are stepped, mosaics are curving out at bottom around shower tray. Mosaics to inside of side partitions are poorly fixed with joints too large, wavy lines uneven, mosaics with odd ones stepping out. Some mosaics are again broken, the top 200mm appears to have been completed at a later date and appears different to the rest, is uneven and steps out. Mosaics around wash hand basin area – grout not cleaned off properly and joints ragged. Mosaics to bath are out of course at left hand end. Again some mosaics are broken, joint too wide at top near cupboard.

Flat 0.6 – Mosaics to shower: Joints around radial corner are too wide and wavy. Mosaic around shower hose connector are uneven. Some joints are not filled with grout correctly and around bottom, joints are filled with adhesive. Mosaics are poorly fitted around overflow pipe. Mosaics at bottom next to shower tray curve out. Mosaics behind kerb – joints filled with adhesive not grout and are out of course at one end and left in a filthy state. Tiles to kerb, joints untidy, some tiles stepped. Area around wash basins, some mosaics are broken at corners, grout has not been cleaned off face of tiles properly. Joint to top of mosaics under cupboard too wide, joint near mirror is messy. This also applies to joints to corners under mirror.

First Floor:

Flat 1.1 – Mosaics curve out towards stop beads. Some mosaics are stepped and uneven, joints around radial corners are wavy and too wide and mosaics curve out at bottom near shower tray and also at ceiling level. Some joints are not filled with grout correctly, others left rough. Area around hose connector, there is adhesive filled joint and rough epoxy grout around this area. Vertical joint near left hand radial corner, the vertical line is stepped. Mosaics behind kerb, slope out at bottom, are curved to left hand end and joint to corner is half filled and ragged. This also applies to the right hand corner where grout has not been removed. Kerb tiles are filthy and joints to ends rough. Tiles are uneven and joints not filled correctly.

Flat 1.2 – Mosaics to kerb are uneven, left in a filthy state with paint and grout over the face, some joints are filled at the bottom with adhesive not grout. Courses to mosaics to both side partitions are out of course as they go around the corners. Tiling to kerb is filthy and joints incorrectly filled. Mosaics to back of kerb, slant outwards towards bottom, curve the wrong way in the radial corner, the joint between the mosaics and the shower tray in this area is too wide, the mosaics are not grouted correctly, the mosaics in the radial corner stick out beyond the line of the tiles to the kerb, mosaic joints are too wide in places, corners broken off of mosaics, wall area at back of mixer back plate is not flat and there is a gap of approximately 4-5mm behind the back plate which will allow water to run down the back. Mosaics around hose connector, there is a mosaic missing. Mosaic to right- hand partition on the inside are in a shocking state, they are bobbling all over the place, at the bottom there are ¼ of a mosaic out of line, stepped and require replacement. Grouting has not been carried out correctly around shower tray to mosaics, joints on mosaics to ceiling are ragged. In places, joints are filled with adhesive not grout. Mosaics missing around overflow pipe and this area not grouted. Mosaics around wash basin area, broken, poorly grouted in corners and stepped in places.

Flat 1.3 – Bath panel: Mosaics stepping, grout on face and generally in a filthy condition and are finished short of the cupboard at the right hand end leaving a gap showing timber. Panel to side of cupboard, mosaics are wavy at bottom, corner joints are too thick and poorly finished. Grouting not cleaned off and left ragged in mosaic joints. Jointing between mosaics and bath not correctly filled. Mosaics down to bath at tap end do not cover roll of bath. Finish of mosaics to aluminium trim is uneven and poorly grouted. Some joints are not grouted correctly, there are broken mosaics in places.

Flat 1.4 – Mosaics to bath panel are stepped and bobbled. Grouting is rough and unfinished around edges. Mosaics are finished short of cupboard panel showing bare timber. Mosaics are stepped in places, some joints are too wide. Panel to side of cupboard: Grout left on face of mosaics, bottom edges wavy, mosaics and grouting not finished off around overflow pipe, joints to corners are untidy. Grouting joint from mosaic to bath untidy. Joints next to bath filled with adhesive not grout.

Flat 1.5 – Mosaics around wash basin, grout not cleaned off face of tiles, broken mosaic to right hand corner of wash basin near front, above this area mosaics to edge of partition appear to be a different colour and are uneven. Some mosaic joints are too wide, mosaics broken, mosaics to radial corners are wavy and stepped and lines wavy. Mosaics below mixer tap are stepped and uneven at the bottom and require replacement. Mosaics to bottom of radial corners curve out at the bottom, some joints are incorrectly filled. The gap between the ceiling and mosaics is too wide at on end and where it has been filled, has been left ragged. There is a mosaic missing next to hose connector, there is a gap behind the back plate to the mixer tap because the partition is not flat to receive this. There is epoxy grout on he mixer unit, mosaics behind shower kerb bulge at the right hand, grouting is untidy to bottom joint next to shower tray. Tiles to kerb near radial corner are poorly fitted, there are steps in the tiles and he joints are not correctly grouted, mosaics in the front of the shower tray kerb are stepped, poorly grouted, joints at bottom are filled with adhesive not grout and joint next to floor is being left in a filthy state. The shower tray in this flat is damaged, there are chips in the enamel.

Flat 1.6 – Mosaics are stepped, wavy and curve out near stop beads. Some joints are too wide, particularly at radial corners, joints are incorrectly grouted, tiles curve out at bottom of shower tray, some tiles are broken, mosaics around hose connector poorly fitted and poorly grouted, the gap at the top of the mosaics next to the ceiling is too large and poorly finished. The gap between mosaics and shower tray is poorly finished, has not been raked out and grouted correctly. Mosaics behind kerb, joints are filled with adhesive not grout, are fixed wavy and slope out at bottom at left end. Corners poorly finished, tiles to kerb again grouting unsatisfactory.

Flat 1.7 – Mosaics carrying on around corners of partitions are out of course. Partition next to toilet, mosaics are poorly finished around hose connector, and mixer tap. There are humps and hollows, joints not grouted correctly in some areas. Grouting not carried out around overflow pipe. Mosaics to radial corners are wavy and undulating. Mosaics curve out at bottom near shower tray. Joints at bottom filled with adhesive not grout, joint between mosaics and shower tray not grouted correctly. Mosaics at back of kerb are out of course with other mosaics and slope out at bottom. Kerb tiles are filthy. Mosaics to face of kerb are stepped and uneven. The grouting is of a poor standard and left in a filthy condition with grout on the face. Tiles and mosaics do not co- ordinate around radius.

Second Floor:

Flat 2.1 – Mosaics are stepped and uneven, some joints too wide. Gap to bottom of mosaics near shower tray is too wide, not raked out and not grouted, mosaics poorly finished around hose connector and not grouted. Mosaic has been drilled above mixer tap, some areas of mosaics have obviously been replaced and these stand out as such and require attention. Some joints are not grouted correctly, grout has been left on the front of mosaics, mosaics behind kerb are sloping out at bottom and wavy. Kerb tiles are dirty, some mosaics are broken.

Flat 2.2 – Mosaics to front of shower tray have been left filthy with grout smeared over the face, there are broken mosaics which require replacement, near the top left hand corner, a mosaic appears to be missing at end of kerb tiles. Mosaics to face to front of shower tray curve out in corner near cupboard, there are also hollows in the face at the bottom. Joints are filled with adhesive instead of grout, joints to kerb tiles are too wide in places and surfaces uneven, tiles are either stepped back or overhanging face of mosaics around radius. Mosaics are out of course around corners of partitions at back of kerb, joints are filled with adhesive instead of grout, some joints are empty, mosaics around radius are uneven and slope out at the bottom, in some areas mosaics are stepped, joints in radial corners are wavy, some joints are too wide, some mosaics are broken. Mosaics and grouting not finished off correctly around hose connector, some joints not filled with grout, grout left on face of mosaics, joint between mosaics and shower tray not raked out and correctly grouted, mosaics to rear of kerb are out of course with adjacent mosaics, around bottom of shower tray joints are filled with adhesive not grout. Shower tray is chipped, mosaics around wash hand basin – some mosaics are broken particularly on corners, joints at corners are messy and require tidying up. Mirror is broken at left hand corner, either mirror is out of square or walls are out of plumb, opening appears tapered.

Flat 2.3 – Bath panel: Joints at bottom of bath panel are filled with adhesive not grout, there is a line step in the panel, generally face of the panel is filthy and grout left on, there is a problem with the joint to the floor on the bottom of the bath panel, grouting is unsatisfactory to this area and appears to be over the floor covering at the right hand end. Panel to side of cupboards, the mosaics are wavy at the bottom, mosaics to the top of this panel – joints are filled with adhesive not grout, two mosaics are cut short, generally joints of mosaics to ceiling are messy, grout has been left on face of mosaics, bottom joints around bath are filled with adhesive not grout, mosaics behind taps do not cover the roll of the bath and water can get down behind the bath in this area, mosaics are scratched, some grout joints – the grout appears to be filthy and does not colour match, requires to be changed.

Flat 2.4 – Bath panel: some mosaics are stepped, some joints are too wide, mosaics are stepped short at the left end and grouting is messy and unacceptable. Face of the panel is filthy with grout left smeared on the face, the bath is chipped, panel to side of cupboard, mosaics to the bottom are cocked, uneven, wavy and incorrectly grouted, work has not been finished off satisfactorily around the overflow pipe, some vertical lines are wavy, in particular over the taps, grout has been left on the face of the mosaics, corner joints are untidy, again mosaics do not cover onto roll of bath which will allow water penetration down back of bath. Mosaics in areas are stepped and wavy and incorrectly grouted, joint between bath and mosaics has not been raked out and correctly grouted,, some mosaics are broken particularly on corners.

Flat 2.5 – Mosaics around wash - basin area at high level are stepped, corners are broken off some mosaics, joints are not correctly filled, finishing to corners is untidy. Shower: Mosaics to front of shower tray – some are stepped, joints at floor level are filled with adhesive not grout, some joints are wide, some tiles are broken, some mosaics are wavy, tiles to kerb are dirty, mosaics to back of kerb are uneven and covered in grout, jointing to tiles is poor. Joint between mosaics and shower tray has not been raked out and correctly grouted, some mosaic joints are too wide, mosaic missing near hose connector and grouting in this area is poor.

Partition behind mixer.

Unit back plate is not flat and there is approximately 4mm gap to allow water to run down the back. Mosaics in places are stepped and uneven. Vertical line to radial corners are wavy, grout has been left on the face of the mosaics and joints in some places are rough, mosaics sloping out to stop beads, joint from mosaics to ceiling is rough.

Flat 2.6 – Tiles to kerb are filthy and appear to be covered in floor adhesive, mosaics to rear of kerb are wavy and sloping in at the bottom right hand end, corners at ends are too large and poorly finished, mosaics slope out near stop beads, some areas of tiles are stepped, some areas are wavy, joints are not filled correctly, vertical lines around radial corners are wavy down the vertical line and the mosaics themselves going up the corners are undulating, joint between mosaics and ceiling is messy. Mosaics and grout not finished off correctly near hose connector and grout left on face of mosaics, wall is not flat at back of mixer tap back plate. There is approximately 4mm gap. Blue grout has been allowed to set on mixer tap unit and in attempting to remove this, the unit is now scratched and covered in grout. Vertical tile lines at lower levels on all three elevations are wavy, bobbling about and joints not filled correctly. Joints between mosaics and shower tray has not been raked out and correctly grouted. Shower tray is chipped and scratched.

Flat 2.7 – Tiles to front of shower kerb, some are broken, some are wavy and stepped. Tiles slope out at bottom, grouting is filthy, mosaics at side of sink alongside Formica panels down to top of shower tray kerb are set back approximately 6mm from the face of corner bead, tiles to kerb have been left in a filthy state, mosaics uneven around hose connector. Some joints too wide, vertical line of joints down radial corners is again wavy and undulating down the corner, some joints are incorrectly filled, grout has been left on face of mosaics. Some mosaics are stepped, some are wavy, joint around shower tray has not been raked out and correctly jointed. One mosaic next to shower tray has slipped down leaving a 6mm gap, joint around ceiling line is untidy, some mosaics are broken, mosaics behind kerb – joint at top next to kerb tiles is almost 10 mm wide and unacceptable. Mosaics step and are also wavy, grout has been left on face of mosaics, vertical line around radial corner is sloping, some mosaics at top stick out beyond face of edge of tile. Kerb tiles are filthy, incorrectly grouted and grooves appear to be filled with grout. "

200. After 1993 substantial remedial works were carried out. Room 106 (originally room A 06) was given as an instance of Birse's inability to do the work properly but it is typical of the work under this head. The mosaics in it were listed as a snagging item in 1993 but not in 1994 although on 12 September 1994 a Clerk of Work's direction (no 158) was issued:

"Room 106 Mosaics

The standard of mosaic work in the above shower room is appalling and not up to the required standard. The whole of the mosaics are to be removed, re rendered where necessary and new mosaics fixed to the walls. This work is to be completed by Friday 16 September 1994 at the very latest to enable the room to be let. The College has already lost two weeks remuneration due to the long delay in completing the work."

It appears that work was done since tilers arrived in October 1994 to attend to Room 106. In June 1995 Mr Cox and Mr Prichard inspected the mosaics and produced this note:

"CABLE & WIRELESS COLLEGE

Residences: Notes on shower/bath mosaics inspected on 26 June 1995

Present: G Cox and D Prichard

101. Backing irregular so edges splayed.

Acceptable

106. Done three times, LHC joints very wide, irregular backing.

Epoxy grout on tray.

Retain if grout removed (if kept, no payment)

112. Grout colour changes but best left.

Tiny middle corner at bottom

Remove grout from tray

116. Irregular edges, wide joints. Colour of grout changes.

Regrout when washed out.

126. One wide horizontal joint all round where retiled; top half OK.

Bottom corner poor. Needs work.

203. Grout colour irregular, wide joints in corners.

Leave since unlikely to better.

216. Bottom RHC, lip on LH edge, first impression OK but gets worse.

Work needed.

217. Three wide joints, grout on tray. Patches of grout washed out.

(If kept, no payment)

302. Sheet joints visible. Replace, not wanted even if free!

305. Retain if black joints cleaned.

317. Bottom middle corner very wobbly.

RH bottom corner and dark joints (retain if given it?)"

201. Whatever work was done evidently satisfied MJP generally. On 15 March 1995 MJP issued Certificate of Completion of Making Good Defects No.1. In the covering letter MJP said only:

"We have determined that the damp egress into some of the bedrooms is an external fabric problem, but once resolved, further making good will be required to the interiors of these affected rooms listed on this certificate. This decision has been made with your understanding and co-operation.

A specific issue of further remedial works to the mosaics in a few of the showers still has to be resolved."

On 23 June 1995 a site meeting was held "to record progress on defects and to establish which sections were complete to release stage completion Certificate of Making Good Defects. …. BCL did not issue a status report so the minutes below record the current situation, actions and agreements.". The items included:

"Mosaics:

Inspection being carried out by CofW. Some mosaics still not acceptable. Defects range from missing grout, missing or badly aligned tiles, to extremely poor setting out. BCL agree remedials have not been wholly successful. BCL wish to reach a compromise on how to resolve the issue without keeping BCL on site and major disruption to College. MJP/BCL/CofW to inspect after meeting to agree degree of acceptability and set up systematic inspecting procedure with MJP (DP), CofW and BCL."

202. A few days later, on 26 June 1995 Mr Cox and Mr Prichard inspected the mosaics and produced this note:
"CABLE & WIRELESS COLLEGE

Residences: Notes on shower/bath mosaics inspected on 26 June 1995

Present: G Cox and D Prichard

203. Backing irregular so edges splayed.

Acceptable

204. Done three times, LHC joints very wide,, irregular backing.

Epoxy grout on tray.

Retain if grout removed (if kept, no payment)

205. Grout colour changes but best left.

Tiny middle corner at bottom

Remove grout from tray

206. Irregular edges, wide joints. Colour of grout changes.

Regrout when washed out.

207. One wide horizontal joint all round where retiled; top half OK.

Bottom corner poor. Needs work.

208. Grout colour irregular, wide joints in corners.

Leave since unlikely to better.

209. Bottom RHC, lip on LH edge, first impression OK but gets worse.

Work needed.

210. Three wide joints, grout on tray. Patches of grout washed out.

(If kept, no payment)

211. Sheet joints visible. Replace, not wanted even if free!
212. Retain if black joints cleaned.
213. Bottom middle corner very wobbly.

RH bottom corner and dark joints (retain if given it?)"

214. It appears that some work was put in hand. On 7 August 1995 Mr Cox wrote to Birse:

"DEFECTS

There are a number of items on the defects list that are still outstanding and are causing us some concern. They are as follows:

Mosaics to the Residences and Leisure Building.

Replacement doors to the Residences.

…..

We are also still awaiting your revised remedial works programme which was promised after the last site meeting on 23 June 1995.

Please will you give your earliest attention to these matters."

A reply was sent to Mr Cox on 22 September 1995:

"Dear Sir,

Thank you for your letter ref. FGC/CAB dated 13 September 1995. We apologise for the delay in replaying.

In response to your letter of 7 August 1995, regarding outstanding defects we use your numerical sequence.

Mosaics to Residences and Leisure Building

The shower/bathrooms have been inspected by yourself, the Architect and Birse and have been categorised as follows:

Type 'C' are condemned and room nos. 127, 222, 226, 302, 517, 525, 601, 605, 916, and 922 fall into this category.

Type 'B' require remedial works to the grouting and mastic and room nos. 116, 126, 201, 205, 216, 217, 224, 225, 227, 301, 305, 317, 325, 326, 401, 415, 416, 417, 421, 422, 512, 516, 521, 522, 523, 524, 604, 606, 611, 612, 614, 627, 702, 704, 711, 713, 714, 716, 717, 722, 723, 801, 802, 804, 811, 823, 901, 902, 904, 905, 912 and 913 fall into this category.

In the majority of these you claim there are differing colours of grout, this effect is caused by the staining of soap scum and mould growth on the surface.

Type 'A' are acceptable and all other rooms fall into this category.

The shower/bathrooms in category 'C', if re-tiled will be required for two weeks each whilst remedial works are carried out and materials allowed to cure. The completion date of works will depend on how many rooms are made available to us at a time.

Category 'B' rooms will be required for 1 day each.

The male and female toilets in the Leisure Building can be re-tiled; each would be required for a two week period and all fixtures and fittings would be removed.

The mosaics to changing rooms and swimming pool area will be checked for grouting and mastic remedials.

We are at present negotiating with a specialist mosaic tiling company to carry out these works.

……"

Therefore Room 106 did not feature in that list. It is in my judgment clear that there was an agreement and that its purposes was to attain finality as to what was and was not acceptable to MJP. It was suggested to Mr Allsop (but not accepted by him), that in some instances MJP accepted work although not satisfactory but because they could live with it. I do not believe that this was correct. MJP was as interested as Birse in bringing matters to a conclusion. Birse continued with remedial work as best it could, given that the College was heavily occupied.

215. Later in the year (November) a status report recorded that

"Mosaic Tiling

1.8.1 Remedial works to Residential blocks where re-tiling is not required has been on-going for some weeks. Completion of these will be by end November 1995.

Re-tiling to Residential and Leisure blocks will be carried out by a specialist mosaic contractor and will be commenced by 27 November 1995 when the rooms are made available to us.

Duration of works is expected to be 3-4 weeks.

2.0 Snags

2.1 Snagging has been carried out in all areas. Inspection sheets have been offered to Clerk of Works for acceptance/comments. Return visit is being carried out now, commencing in Residential Blocks, following C.O.W's comments. This system will be followed throughout all areas with completion anticipated by end November 1995."

On 30 January 1996 Birse reported that it had taken advice from two specialist mosaic companies and that as regards the residences there were a few showers which could be remedied by re-tiling but that "none of the other rooms would benefit from attempts at remedial works other than cleaning and regrouting – which has been done". Similarly Mr Prichard's notes of a meeting on 1 February 1996 stated:

"All showers have been upgraded to be usable. Due to heavy college bookings Birse have not arranged to start stripping and replacing the ten condemned showers, many to be out of action for a week and possibly three at a time. Mr Beatson agreed the work need doing but at a convenient time for the college, say over Christmas 1996. The way forward is to remove this work from Birse."

Mr Prichard agreed that there were only ten showers left outstanding and apparently there was no complaint about the rest. As he said:

"….there was an acceptance by then that there were ten which the client wanted something done about, we wanted something done about because it was becoming progressively impractical to keep going back over mosaic and trying to improve it. It is very difficult to do piecemeal remedials to mosaic."

216. Room 106 appeared in the 1996 list of outstanding snagging items as part of the re-tiling still remained to be done in the residences. In the leisure block, all the work was carried out except to the male toilet, which Birse agreed needed to be hacked off and re-tiled, but as the leisure block was in constant use, the work was not done. No work was done to Room 106 thereafter. Mr Woodward nevertheless treated the mosaics in Room 106 as acceptable. He inspected 44 rooms and found only three had defects in them that made them unacceptable. He therefore did not regard the conditions recorded by Mr Prichard as unacceptable.
217. Birse had a number of defences. As I understood it, Birse first wished to rely on the unreality of being able to meet MJP's requirements in the real world. This in turn raised the running issues of the standard of workmanship expected of Birse and of the satisfaction of the architect, MJP. The former also led to discussion about tolerances and the like. Miss Jackson used the term "workman's tolerances" in order to excuse (or explain) the result that Dr Fleming cautioned against. Birse also relied on MJP's acceptance of some of the work. It also of course maintained that ET had suffered no loss or was not now entitled to recover the full cost claimed.
218. In my judgment it is necessary to consider, once again, Birse's primary obligations. There are no specific standards of workmanship applicable to the mosaic tiling. Clause 8.1.2 says that the workmanship has to be of a standard "appropriate to the Works, provided that workmanship shall be to the reasonable satisfaction of the Architect where and to the extent that this is required in accordance with clause 2.1.4". Accordingly the standard is that of MJP's satisfaction. This is not an instance where achievement of the architect's satisfaction was impossible, impracticable or unreasonable. It is clear that Mr Prichard and others of MJP took a close interest in the quality of the work and from time to time expressed satisfaction or otherwise. An expression of satisfaction is not binding, for example, it does not prevent a later complaint about something that was not apparent or reasonably apparent. It is possible for either Contractor or Employer to be dissatisfied with the standard and thus to question the Architect's opinion.
219. In this instance it is my view clear that MJP was satisfied with the standard of much of the mosaic tiling. For all practical purposes the status of the work was established in 1995 when there was agreement about categories A, B and C. In my judgment Birse attained its contractual obligations in relation to category A. MJP's acceptance with the work constituted satisfaction for the purposes of the contract. I also consider that MJP's opinion was reasonable and should not be opened up, reviewed or revised. The fact that MJP were not dissatisfied with some aspects of the works does not deprive the certificate of effect. It affected ET's use of the premises and Birse relied on it. It would not be right now to say that it should not have been issued and that the defects liability period is still running. Birse did not attain its contractual obligations in respect of the work to be done under categories B and C in so far that work remained outstanding in 1996 lists. To that extent Birse is liable to ET. I do not accept that Birse could obtain the satisfaction of MJP; the work was plainly feasible in that some of it could be done, and was done, satisfactorily. It may have been expensive to do so but that is not an answer.
220. However it is equally clear that it would be pointless now to complete the outstanding work. It has performed well; it has plainly deteriorated as a result of use; there is no evidence of any real intention on the part of ET to make good the work; the work now required would be to make the tiling presentable after years of use and would not now be to rectify anything done or not done by Birse.

5.B Joinery

221. This claim is for £219,286.56 (including fees) which is the total of the costs agreed for doors etc to be replaced. The claim is essentially about doors that have warped. They include big, solid doors with three hinges. The problem began in 1993, even before completion. For example on 18 May 1993 a clerk of works reported to Mr Gander:

"Internal Areas Only

The following items were revealed on a Routine Spot Check System on the Residential Blocks.

The Items should not be considered a Final Check List but a guide to the standards required by the architect Mr David Whitehead and Mr K Wilson, Clerk of Works.

The consensus of opinion was that the Entrance to the Building and the Circulation and Landing Areas were prime observation points, finished Workmanship Standards must be treated as such.

Therefore all horizontal and vertical proposed shadow lines to plaster and skirting areas must be met with a far greater accuracy.

General Plastering Standards have been remedied but still Door linings and plastered walls are misalign.

The residential Quarters, standard of workmanship has still to be improved but apparently great strides have been made to reach the standards now achieved.

Standards

GENERAL NOTES all internal

…..

Doors are being fitted into linings that appear to be either in wind or out of true – several linings appear to be oversize and create fixing problems by far the biggest headache appears to be in the Double wardrobe door arrangements. Several areas in block 'B' are totally unacceptable."

222. On 22 July 1993 MJP sent Mr Gander a series of reports from the clerks of works:

"We attach a series of reports made by the Clerk of Works over the last two weeks. These sheets should be read as helpful notes to assist your works leading up to you offering areas for snagging. Naturally the comments on houses D and E will be incorporated into our snagging sheets where appropriate."

One dated 12 July 1993 read:

"RESIDENCES – INTERNAL AREAS

E Area

Internal door hanging standards still give considerable concern. Many doors are hung with excessive gaps between meeting edges. Typical situations appear in E0.1 and E0.2, irregular gaps at meeting rebate and on some occasions at hanging edge.

E1 area: Several doors are beginning to twist in their present format leaving the rebate area at central locking positions. Typical situation E1.1 bathroom – excessive gap south wardrobe double door units again E1, 2&3.

Main Entrance Doors showing slight signs twist in E1.1 area. Decorators working.

Very bad meeting edges in E2.4. Twisted doors being decorated in E2.12 wardrobe arrangement.

E2.1 is of a similar nature – doors being taken out of twist in their rebate by moving hanging edge out of at the top or bottom accordingly. Typical example is E2.1, single wardrobe arrangement. Double wardrobe door in E2.3 beginning to twist.

Doors binding entrance door to E2.3 no access decorators are operating on final completion stages.

D Area

D0.4 – Excessive gaps on wardrobe doors.

D0.1 – Excessive gaps to main entrance door, bathroom door in showing signs of twist, main wardrobe doors are in twist.

D1.5 – Wardrobe doors beginning to twist at edge.

D1.6 – Excessive gaps at main entrance door.

D1.6 – Bathroom door in twist at head.

D1.6 – Wardrobe door in twist badly hung.

D1.12 – Main entrance door excessive gaps, wardrobe doors taken out of twist by moving hanging edge out of top.

D1.1 – Wardrobe doors in twist at head.

D1.1 – Mosaic worker operating in bathroom, shower tray illuminated appears to indicate damaged areas.

D2 area – Painters operating – unable to gain access."

The twisting and warping of the doors featured on the snagging lists. Birse acknowledge that it carried out extensive work. Mr Allsop said, for example, that 53 new doors were ordered (but he was there only repeating a letter of 22 September 1995 – see below) and in evidence he described how he kept track of the work being done by Rattigans, Birse's sub-contractor. Birse produced no records of the remedial work. There must have been extensive records since a contractor such a Birse would record all its costs that it incurred and why they were incurred, especially if it was to put right work that might be the responsibility of a supplier or sub-contractor. The absence of these records is distinctly odd and was rightly the subject of submissions by ET that I should draw inferences adverse to Birse. It is certainly not possible to accept Miss Jackson's submission that 80 doors were replaced. Similarly I cannot accept the submission that doors which were replaced in 1993 may have warped again for reasons which are not the responsibility of Birse as there is no sound evidentiary basis for it. It is however clear that some of the replacement doors did warp again, although they might easily have been indistinguishable from other doors for, as Mr Cherry said, the problem continued. However the Certificate of Making Good Defects was issued in respect of the Residences and part of the Teaching Block, so MJP must have been satisfied with the state of the doors since the covering letter contained no reservation in respect of them.

223. In 1994 Birse had brought in the manufacturers of the doors, Crosby. It did so again in 1995 when Crosby inspected the doors in the summer (when the heating would have been off). It sent the following report to Birse:

"Regarding difficulties, specifically distortion, occurring in solid flush doors supplied to this contract, we have again given consideration to the situation.

As has been advised before, we diagnose the cause of the movement in the doors as being a result of abnormally high ambient temperature being present. In particular, as in wardrobe situations, variations in temperature or moisture levels between one face of the door and the other will set up considerable stresses as the timber components move within their natural inclination.

The basic specification of door prescribed for this project is well proven and established. Regrettably however it is not practical to offer any guarantee of stability when rapid drying out of the leaf can be anticipated particularly in an unbalanced way.

In keeping with all recognised standards the moisture content at production is at 10% - 14%. Reduction to less than 4% as in this case will have an effect and distortion is essentially unavoidable. Accordingly, in consideration of any alternative specification that could be adopted in replacement of doors effected we can offer no practical solution. In addition it is our view that no guarantee could reasonably be offered against similar movement occurring in any replacements that are fitted, assuming the use of a timber based product and no change to the ambient conditions."

Birse thereafter wrote to Mr Cox on 22 September 1995:

"Thank you for your letter ref. FGC/CAB dated 13 September 1995. We apologise for the delay in replying.

In response to your letter of 7 August 1995, regarding outstanding defects we use your numerical sequence.

…..

Replacement doors to residences.

The problems experienced with the doors are not attributable to defective materials or workmanship. The manufacturers, Crosby Sarek Ltd, have re-iterated their earlier advice that the distortion and shrinkage of the doors is a result of high temperature and rapid drying out. We enclose a copy of Crosby Sarek's Report for your information.

Birse have employed joiners to carry out adjustments to these doors so that they perform the function for which they were designed. This operation, for which the college has supplied a security guard as the rooms are occupied, has taken 2 joiners and a decorator some 11 weeks to carry out. In addition to this operation 53 new doors have been placed on order to replace those beyond adjustment; fitting of these doors should commence during week commencing 9.10.95.

This positive action and expense taken by Birse Construction in no way implies our acceptance that the doors are defective; it is in response to a situation requiring essential remedial action and a responsible attitude toward the client's needs.

Please find enclosed our remedial works programme to completion which is based on the procurement and fitting of replacement doors."

That remained the position. For example at a meeting on 26 January 1998 between Mr Barraclough, Mr Allsop and Ms Tooth about the defects in the CTP list, Ms Tooth's notes recorded:

4.10		Corridor cupboards
4.10.1		RA said that there had been shrinkage and warping of the doors because they were softwood.
4.10.2		ET said that there was some warping, but that the main problem was bad fit of the doors (varying gaps between doors and frames) and in her opinion this was due to bad installation.
4.10.3		ET said that some specific doors were on the defects list given to her, but on inspection she had considered the defects to be general.
4.10.4		ET gained the impression that CB considered the complaint to be justified.
4.10.5		CB said he didn't think Birse would have left the doors like this, and CB and RA discussed the possibility of doors having moved since Birse left site.
5.0		Residences
5.1		One room in the east Residences and one in the west were inspected.
5.2		Bathroom and Cupboard doors
5.2.1		ET said that some specific doors were on the defects list given to her, but on inspection she had considered the defect to be general.
5.2.2		RA said that the problem was shrinkage/ warping of the doors because they were softwood.
5.2.3		CB showed ET a letter from Crosby which stated that the doors could not be covered by their guarantee because of the environmental conditions.
5.2.4		RA stated that Crosby had visited site and measured 90 deg F temperature and 0% RA
5.2.5		RA said that a significant number of doors had been replaced ( value approx £26,000) under his supervision.
5.2.6		ET said that there was more warping in the Residences than the Teaching Block, but that the main problem was again bad fit of doors (varying gaps between doors and frames), and in her opinion this was again due to bad installation.
5.2.7		ET gained the impression that CB considered the doors to be of an inadequate standard.
5.2.8		CB and RA discussed the possibility of piecemeal remedial work to each door.
5.2.9		ET said she considered this was unrealistic and that she considered the frames would have to be replaced.
5.2.10		CB said that the work involved and disruption would be enormous.

224. The basic issues are therefore clear. Under the contract Birse had to compete the works with doors that had not warped for otherwise they would not be in accordance with the contract, i.e. they would not be of good quality. It had also had to fit them properly with good workmanship so, for example, the door frames should not be out of square (as noted by Mr Woodward in his report). However it does not follow if the doors warped after practical completion Birse must be liable. The reason has to be established, especially in the light of the contemporary assertions that the temperatures in the buildings were too high. Birse relied on the fact that the desks installed by the fit-out contractors in the Residences all warped. On 3 September 1993 MJP wrote to ET to say:

"We confirm below the discussions held on site today, regarding the issue of partial possession certificates. As you are aware, the situation regarding partial possessions and conditions of works is complicated and contains a number of lacunae

To enable delegates to arrive on Saturday 4 September and begin teaching on Monday 6 September 1993 the following partial possessions will need to be issued:

….

Fit- out items: Furniture has been snagged but remedials not carried out. Unfortunately every desk is scratched or damaged and the excessive heat during testing and commissioning has warped the tops."

Mr Cherry could not recall the event but pointed out that "obviously the excessive heat was due to testing and commissioning, turning it up high to make sure it works". Birse was responsible for the central heating so that incident does not assist it. Nevertheless some rooms that were affected faced south and got hot in summer and there was a fair amount of evidence that the temperatures in the College were high. Ms Tooth (who had not carried any investigations into the timber that had been specified or taken any temperature or humidity readings) noticed the residences, when occupied, were very hot and that it was hotter higher up the building so much so that one could break out into a sweat. Mr Woodward spoke of a "sauna-like atmosphere". Mr Fleming acknowledged that some of the south facing rooms were "exceptionally warm". Mr Cox said that the buildings were very hot especially the residences.

225. As appears from his report Mr Woodward inspected 60 of the entrance and shower or bathroom doors complained of in the Residences and found 18 twisted doors and one frame out of alignment. He thought that that 41 doors (i.e. 68%) were acceptable and were unlikely to have been twisted when fitted. He also inspected a large and, in my judgment, a representative sample of the wardrobe doors: 63 of the 231 double and single wardrobe doors complained of in the Residences. He found 19 twisted doors, three with excessive gaps and seven doors that had been poorly fitted. He thought that 34 of the 63 doors (i.e. 53%) were acceptable and that the doors were unlikely to have been twisted when fitted. He inspected the cupboards in the Teaching Ancillary areas and found only about 10% to be twisted. Overall Mr Woodward considered that the preponderance of defective doors would not have been in that state on completion. Mr Fleming looked at about nine rooms only: four in the west wing and about five in the east wing. He agreed that there ought to be evidence of the factors that would cause twisting (although he thought that it was to be found in the statements of Ms Tooth and Mr Cox), but he had made no investigation.
226. Miss Jackson submitted that, in order to establish that it was necessary to replace doors to the extent claimed, ET ought to have presented expert and other evidence of inspection and assessment as to how many doors are unacceptably twisted or out of alignment, and the reasons for those conditions. I agree. Obviously where there are straightforward gaps then, almost certainly, they existed in 1993 for although they could be caused by warping and twisting, Mr Woodward said that they were not found where he found twisting. Equally Birse is liable for any doors specifically identified as defective in 1993 and which were not replaced. However ET has failed to establish that the other doors were not of the requisite quality. Given the recurrence of the problem in doors which were otherwise standard doors, manufactured and supplied both originally and as replacements over quite a long period, it is necessary to establish that there were no extraneous factors and doors that emanated from a reputable source were nevertheless not of the required quality. I do not consider that there is evidence from which I can properly draw that conclusion. Accordingly whilst there may in principle be liability for defects that were present in 1993 and 1996 (such as frames out of alignment) which have not been rectified, Birse is not liable for the extent of the remedial works costs claimed as they have not been shown to have been caused by any breach of contract on its part.
227. Even if I am wrong in that conclusion, ET has had the use of the doors for 10 years or so; had there been any real effect on the use of the College they would have been replaced long ago. It would be unreasonable to do so now, just as it would be unreasonable to hold that ET could recover the supposed claim as damages for Birse's breach of contract in supplying doors, some of which have twisted and warped but not to the extent of making then unusable.

6. Certificate of Making Good Defects

228. Were it necessary I would not open up the Certificate of Making Good Defects that was issued, nor would I issue another one for the whole of any part of the works so as to entitle Birse to the release of further retention. First, the special nature of the Works was essentially such that the contractual requirement of achieving the satisfaction of MJP, always important, was particularly so. The parties, and in particular ET, contracted on that basis. Birse undertook to achieve that result, ultimately signified by the Final Certificate, but rather more relevantly by certificates required at the earlier stages of practical completion and making good of defects. The width of the power of revision in clause 41.4 does not entitle an arbitrator or court to substitute its own aesthetic or other judgment for the opinion and judgment of the Architect where that is the contractual standard, unless that opinion or judgment was quite unreasonable or improper or was withheld unreasonably or improperly. If an Architect has exacting standards they must be respected, even if others might not apply them. The contract is made for the attainment of the satisfaction of X, not Y, acting as arbitrator or court. Otherwise a court or arbitrator would force a party, especially an employer, to accept and pay for something it had not contracted for. Secondly, it is clear that by early 1996 there were still a significant number of items outstanding. Some were items which Birse recognised should be the subject of an allowance. Obviously Birse's offer was low but they and the others could not be regarded as de minimis or to be overlooked. There was no such suggestion from Birse at least at the time. Some were clearly matters which concerned MJP as affecting the appearance of the college, i.e. the work was not to the satisfaction of MJP. Here, as Mr Cherry said of the meeting on 22 April 1996, the decision was that of the firm's and, in turn, of the client. A client may decide not to insist on attainment of the standards set by the Architect to which it is entitled. Although Birse did do more work after that meeting, it is in my judgment plain from the inspection and survey carried out after 1 July 1996 that it had not completed the making good of defects in accordance with the contract, including, where required, to the satisfaction of MJP. In this respect I take account of the Certificate of Making Good Defects as indicative of MJP's standards. Mr Prichard and Mr Cherry were in my judgment, very reasonable and measured in their approach (and in their evidence). Both thought that Birse could and should have done better. I have no doubt that Mr Prichard would have issued a further certificate if he had thought that one should be issued. I can see no reason to fault his decision not to do so. One has only to read Ms Tooth's descriptions of what she found when she came to inspect the college to realise that MJP were not at all unreasonable. In the context of this judgment and although it is part of external works (but not at all to be discounted on that ground as it was an integral part of MJP's scheme and was part of the appearance which concerned MJP) the Blanc de Bierges paving provides a good example of work (perhaps the best example) which had not been carried out in accordance with contract and which was not satisfactory (by any standard, I would add). Thirdly, on the basis of the evidence, notably Ms Tooth's reliable descriptions, I would not issue a Certificate of Making Good Defects backdated, as it were, to 1 July 1996. Fourthly, since ET has not yet brought the works up to the required standards, it follows both that no Certificate of Making Good Defects can now be issued (on the assumption, about which there is doubt, that it is immaterial by whom the defects are made good) and that ET cannot rely on its continuing absence as a defence to Birse's claim. It has its rights of abatement, set-off or counterclaim.

7. Damages

229. To the extent that ET is entitled to an award of nominal damages, I award it £2 for breach of contract. To make such an award in respect of each type of defect or, worse, each scheduled defect would be quite wrong. There is one breach, that of the building contract – a failure to complete the Works in accordance with its terms. Except in relation to a snagging defect which was not a breach of the building contract there is no separate breach of the agreement of March 1994 to warrant a further award as the breaches merge. I award ET £2 for breach of that agreement in relation to any snagging defect that neither Birse nor ET has remedied. In the case of each contract, the defects are instances or particulars of its breach. It would be wrong in principle and unreasonable to make an award in respect of each defect.

8. Conclusion

230. I am not able to say precisely how much might be due to ET in respect of the remedied defects for which I have held Birse liable. Some figures are available; others are not. It will include the costs of putting right items ranging from damage to the Armouralia wall via the Blanc de Bierges paving to stop ends etc. As result of the settlement of the proceedings following receipt of the draft judgment, a further hearing will not be required for that or any other purpose.