Junior Books v Veitchi

ORB 352

The Plaintiff is one of a group of companies involved in the manufacture and supply of plastic and malleable pipes and pipework systems and fittings. The ultimate holding company is Georg Fischer AG, incorporated in Switzerland, and the strategic control of the group's activities is exercised from Schaffhausen, Switzerland. The Plaintiff is itself the holding company of a sub-group consisting of the UK members of the group.

From early 1987 the group developed proposals under the code-name "Project GB" for the rationalisation of its UK operations, in particular by the centralisation of all distribution activities in a single warehousing complex on a site which would also house the UK head office. A feature of the warehouse design was to be two high bays each containing racks of stores on either side of a central aisle, in which would operate a travelling "man-rider" crane, working in conjunction with a computerised record of the location of every item in stock.

In due course a site was chosen and obtained adjoining the M6 motorway near its junction with the M69 outside Coventry and the building was erected between 6 March 1989 and (subject to an issue as to the true date of practical completion) 14 April 1990.

Disputes arose between the Plaintiff and various parties involved in the construction and fitting out of the building. The actions now before me chiefly concern complaints about the roof, so it is convenient to describe the different parts of the building in a way which takes account of the varying roof levels as well as other characteristics. It comprised:

All the roofs are laid to a very shallow fall. The design pitch was for the most part 1.5 0 although there was, I believe, one surface designed to just under 2 0 . As a result of foreseeable and acceptable deflections in the supporting steelwork the finished slope is in places only just over 1 0 . The roof cladding consists of panels some 1200mm wide with an upstand along each side, abutting upstands being capped and waterproofed by "batten caps". Each panel has a 0.5mm steel upper surface and a 0.4mm steel lower surface separated by a 37mm insulation core. A subject of sharp dispute is the design acceptability of the fact that on none of the slopes do the panels, as laid, extend continuously from apex to eave; there are always either two or three rows, joined by "end laps", which I must describe in more detail later. What is not in dispute is that the roofs as laid were and are defective, in that from the outset they leaked and, despite some reduction following the taping over of the end lap joints, still do so. It is also the case that many of the end lap joints are not watertight and that that is and always has been a substantial cause of or contribution to leaking through the roof.

The main contract for the design and construction of the building was a contract in writing dated 10 April 1989 between the Plaintiff and Multi Construction Ltd ("MCL"). MCL, like the Plaintiff, was a member of a group of companies and as is not uncommon those controlling each group and negotiating on its behalf do not always seem to have had the distinctions between different members of the group clearly in mind or to have pursued a settled policy as to which should become the contracting party. There were also changes of name. Thus on the George Fischer group side another company, George Fischer Sales Ltd, seems at one time to have been expected to be the contracting employer, and it wrote one of the important documents, a letter of intent dated 31 May 1988. With one important exception the same was true of the Multi group; at the date of the letter of 31 May 1988 MCL was called Multi Construction (Southern) Ltd, but it was to Multi Construction (Northern) Ltd that the letter was addressed and by whom it was accepted, and even when formal contracts came to be entered into it was the latter company which employed the steelwork subcontractors. In the pleadings these distinctions continued to be ignored (so that for example the letter of 31 May 1988 was treated as having passed between the Plaintiff and MCL) until the trial was well advanced, and on day 14 I refused leave for amendments seemingly directed to relying upon such distinctions for the first time. They are not therefore treated in this judgment as being of any significance.

The exception concerns the Defendant Multi Design Consultants Ltd ("MDC"). MDC also was at the time a member of the Multi group, but for a number of reasons its separate corporate identity has always been actually or potentially material. Although it was the member of the group responsible for design it also did design work independently on works not constructed by members of the group, and conversely the latter constructed works not designed by MDC. It was because of contacts made in the course of its independent work that MDC, and in particular its Mr Fitt, first came to the attention of the George Fischer group, gained the confidence of its executives and effected the introduction of the rest of the Multi group. Secondly it is not in dispute that MDC did some work, at least, directly for the Plaintiff before any contract by the latter with MCL took effect, and itself invoiced and was paid for that work. Thirdly MDC left the Multi group by way of a "management buy-out" in April 1991, before the present litigation commenced.

Parts of the work were subcontracted. For present purposes it is sufficient to record that the subcontractor for the design and installation of the travelling cranes was a company called Dexion Ltd, and that the First Third Party Roofdec Ltd was responsible for the roof cladding and the Second Third Party Severfield Reeve PLC for the steelwork. It is MDC's case that it was MCL's subcontractor for overall design and for co-ordination of the design work of specialist subcontractors, but any such relationship was admittedly informal; it is not suggested that there was any express contract, written or oral.

The Defendants Davis Langdon & Everest ("DLE") were involved in two capacities. They were intended from the outset to be, and in due course became, the Plaintiff's quantity surveyors for the project, and were named as such in the building contract. That function was carried out from their London office and has attracted no criticism featuring in these proceedings. There was later also a suggestion that they should act as "Employers' Representative" or "Supervising Officer" (it being common ground that for present purposes the two terms may be treated as equivalent), and they entered into a formal contract in writing with the Plaintiff to act as the former and were named as the latter in the building contract. That function was carried out from their Chester office and it is in that capacity that they are joined in these actions.

On 3 April 1991 the Plaintiff commenced proceedings against MCL for breach of contract, alleging defective design of the man-rider crane system. MCL joined Dexion as a third party. That action proceeded to trial before me in October 1992, in the course of which MCL and Dexion admitted liability, and to judgment on issues of damages on 3 December 1992, effectively against Dexion, which unsuccessfully appealed ([1995] 1 BCLC 260).

On 29 September 1992 the Plaintiff commenced further proceedings against MCL, alleging roof defects with resultant leakage ("the roof claim") and defective rainwater disposal and sprinkler systems with resultant floods ("the flood claim"). MCL joined Roofdec as a third party to both claims and Severfield Reeve as a third party to the roof claim. On 14 September 1994 the Plaintiff and its Swiss holding company entered into an agreement with Severfield Reeve under which the latter effectively bought themselves out of that or any future claims relating to the project. That agreement is relied upon by MDC and DLE here as requiring a credit against any damages otherwise payable by them, and if I hold them liable I must return to that point.

MCL became insolvent, went into liquidation in March 1995 and was dissolved in February 1996. I was not told of any formal steps to terminate the 1992 proceedings against it, but at some stage they ceased to be actively pursued, they were presumably stayed by the effect of the winding up order and certainly they must have abated upon its dissolution. That last point is germane to another issue as to credits, since against its claim the Plaintiff held retention moneys and other unpaid sums ostensibly due to MCL which the latter can now never pursue.

The present actions were begun against MDC on 9 February 1994 and against DLE on 22 April 1994. Both arise out of substantially the same facts as the "roof claim" in the 1992 action against MCL. In both Roofdec and Severfield Reeve were joined as third parties. The two actions were not consolidated but were ordered to be tried together.

The third party proceedings in both actions against Roofdec were disposed of by Tomlin Orders at the opening of the trial and those against Severfield Reeve by consent orders on the following day, under which each Defendant discontinued its third party claim, with no order for costs, on terms which excluded further proceedings against Severfield Reeve. The trial was therefore effectively confined to the Plaintiff's claims against MDC and DLE.

Against MDC the Plaintiff rather confusingly pleads certain parts of two documents as amounting to "representations" on which the Plaintiff "relied", but the passages in question were plainly promissory in form rather than representations as to existing facts and the pleading goes on to allege that MDC thereby gave warranties in consideration of the Plaintiff's entering into its contract with MCL. In my view that is in truth a plea of a collateral contract and the claim under it is one for breach of such a contract, not for misrepresentation. It is a major issue in the action against MDC whether there was such a contract. If there was, there are issues whether it was simply one for due care and skill in design or whether there was also a supervisory responsibility, or a warranty that the completed works were "fully functionally operational", or both.

The claim against MDC is also laid in tort. Although it is pleaded as arising "further or in the alternative" to the claim in contract the Plaintiff does not in fact rely on any concurrent duty in tort, no doubt because it is unnecessary to do so in the absence of any limitation defence. The issue, therefore, is whether a duty of care arose in the pleaded respects (design and supervision) if, contrary to the Plaintiff's primary case, there was no contractual relationship with MDC once the main contract with MCL was in place.

It is common ground between the Plaintiff and DLE that there was an agreement in writing between them, made on 11 January 1989, under which DLE were to act as Employers' Representative for the main contract works, providing the services specified in the agreement. That should have left little scope for dispute about the extent of DLE's duties, but there are some minor issues of construction and since DLE's witnesses repeatedly disavowed responsibility for some apparently plain duties and that was not abandoned in closing submisions it must briefly be dealt with.

A number of defects are alleged. In a few cases it was denied by one or both of the Defendants that what was complained of was a defect but in most cases that was not in dispute; the issues of liability to which most time and attention were paid were whether defects were caused or permitted by lack of due care or skill in design or supervision within the area of responsibility of the Defendant in question.

There is a separate group of issues of liability and causation centring on DLE's issue of what was described as a certificate of practical completion. Questions arise as to its nature and status, whether DLE were in breach of duty in issuing it and what was the effect of doing so - in particular whether it resulted in the loss of the Plaintiff's right to call on a performance bond.

There has been a wide measure of agreement between the quantity surveying expert witnesses, but some issues of detail remain and there is a major dispute between the Plaintiff and the Defendants as to which of two rival remedial schemes for the prevention of further leaking through the roof is the more appropriate. The Defendants also dispute the necessity for an important element (the installation of netting) of the temporary remedial works already undertaken by the Plaintiff. Finally, under the head of quantum, there are issues whether (i) in principle, and (ii) on the evidence, the Plaintiff can recover a sum for residual loss of value remaining after defects are remedied.

There are issues whether credit should be given by the Plaintiff against any award of damages for (i) the money received from Severfield Reeve under the settlement referred to in paragraph 11 above, and (ii) what could allegedly have been claimed under the performance bond referred to in paragraph 19 above. It is common ground that in so far as money withheld from MCL (see paragraph 12 above) is available to reduce the Plaintiff's losses from roof defects it must give credit, but there are issues how far it is entitled before doing so to make deductions under various heads from the gross amount so withheld, and on day 18 of the trial I ruled that for the most part those issues would have to be stood over to a later hearing.

If either Defendant is liable to the Plaintiff, or both are, there are contribution issues between them.

On 22 March 1988 Mr Fitt of MDC wrote to Mr Gunthardt of Georg Fischer AG with a "fee proposal for the Detailed Planning and Realisation stages of the Project". He added that the fee proposal was "in the form of a lump sum, so that you are able to assess a total cost for the project, as requested".

The document, after a cover and contents page, is headed "PROJECT PLANNERS & DESIGNERS - MULTI DESIGN CONSULTANTS LTD". Section 1.0 is entitled "General Duties and Responsibilities" and clause 1.1 reads:

Clause 1.2 sets out a list of what are to be included in the "design, specifications, drawings and Schedules". Clause 1.3 reads:

Section 2.0 is entitled "Stage 2 - Detailed Design & Planning Stage: Specific Duties & Responsibilities". Clause 2.1 reads:

Section 3.0 is entitled Stage 3 - Realisation: Specific Duties and Responsibilities". Clauses 3.1 to 3.6 continue, without repetition of clause 2.1, the list of duties from clauses 2.2 to 2.14, and in the same grammatical form, beginning with 3.1: "Complete all detailed design calculations, specifications and drawings necessary to permit Construction of the Works". It ends with a "Note" in two parts, of which the first reads:

Section 4.0 is entitled "Scope of Works" and its contents answer to that description. Section 5.0 is entitled "Fee Proposal" and begins:

An appendix to the document defines the end of stages 1 and 2. I do not think I need set the definitions out in full; nothing seems to turn on the exact date of either, although the chronology is sufficiently clear. On 28 March 1988 the Plaintiff formally instructed MDC to proceed with stage 1. On 20 May 1988 Mr Fitt of MDC wrote to Dr Seiler of the Plaintiff, ending his letter: "I confirm that we shall now commence the Stage 2 design process as discussed to define Key Planning Data, with a full commencement of Stage 2 Planning Work following the 3rd June meeting at Coventry". The critical part of the definition of the end of stage 2 seems to have been "Full approval of G.F. of total scheme design, specification and Guaranteed Maximum Price", which was apparently obtained on or about 10 February 1989, and that is consistent with MDC's statement in its fee calculation of 1 February 1989 that "Stage 2 work will be completed w/e 17th February 1989".

On 20 April 1988 Mr Cheetham of Multi Construction (Northern) Ltd wrote to Mr Masterson of DLE's quantity surveying office, who had been negotiating terms on behalf of the George Fischer group, referring to a meeting on the previous day and enclosing a "final draft" letter of intent to be written by Georg Fischer AG to Multi Construction (Northern) Ltd.

On 31 May 1988 George Fischer Sales Ltd wrote the letter of intent to Multi Construction (Northern) Ltd in the form of that draft. Material passages are as follows:

The letter of intent was signed on behalf of the companies between which it passed. Appendix A was the Fee Proposal of 22 March 1988 already summarised with some minor changes, and one more material for present purposes, namely that MDC's fee for stage 3, instead of being a lump sum, was expressed as : "1.21% of the Actual cost of the works including the addition which accrues as a result of any shared savings on the G.M.P." In Appendix E (not B, which seems to have been a clerical error in the body of the letter) are listed the "Key Project Personnel", including as "Project Design Manager" Mr Fitt.

As already noted in paragraph 6 above the main contract for the project works was entered into on 10 April 1989. The parties were the Plaintiff as employer, MCL as contractor and the holding company of the Multi group, Multi Construction (UK) Ltd, as guarantor. As envisaged in the letter of intent it was a contract for design and construction. It consisted of articles of agreement, conditions of contract and appendices.

By article 2 the conditions of contract and eight appendices were to be deemed "to form and be read and construed as part of this Agreement". Appendix 8 was described as "Side letters between the parties dated 7th April 1989". By article 3 MCL covenanted to execute the works in accordance with the contract. By article 7 the guarantor guaranteed the performance by MCL of its obligations under the contract. By article 8 the provisions of the contract were to supersede all previous agreements between the parties.

The following are relevant extracts from the conditions of contract:

Appendix 8 contains two letters dated 7 April 1989 from MCL to the Plaintiff. The first is not material to the issues before me. The second begins: "The Contractors design responsibilities are defined in the Letter of Intent dated 31st May 1988. The following amendments are hereby agreed:-". There then follow a series of amendments which are plainly not to the body of the letter of intent but to its Appendix A, that is the amended Fee Proposal of 22 March 1988. The substance of the amendments is concerned exclusively with computer systems, and so not material to the issues before me, but their form is in my view relevant. They are all purely verbal revisions of the document by addition, deletion or re-drafting. None of them deletes the title or amends the form of the document as one concerned with obligations on the part of MDC as "Project Planners & Designers". The first excepts from clause 1.3, as quoted in paragraph 25 above (that is from the warranty to be provided by MDC "under the form of contract"), the performance of certain computing and interface systems. Another adds two new clauses, 2.15 and 2.16, to the list of duties in section 2.0 governed by clause 2.1, as quoted in paragraph 26 above. A third re-drafts clause 3.4, one of the list of duties in section 3.0 described in paragraph 27 above.

The necessary and sufficient conditions for the formation of a contract are agreement (usually in the form of offer and acceptance), consideration and contractual intent. Where substantial commercial organisations embody or record the progress and outcome of their negotiations in documents, as did the parties here, the issue whether they have entered into a contract or contracts, and if so on what terms, is primarily to be resolved by construing the documents in their factual setting. Leaving aside the law as to misrepresentation, mistake and rectification, none of which features here, the wishes and intentions of the individuals participating and their beliefs, then or later, as to the effect of the documents, are irrelevant except to the limited extent to which matters of shared knowledge or understanding at the time may be part of the factual setting. Only in exceptional circumstances, and on cogent evidence, is a court likely to find that oral exchanges between such parties have produced a result which contradicts the contemporaneous documents.

These are, in my understanding, elementary and familiar propositions. I therefore expressed some surprise and disquiet from time to time during the course of the trial, which I still feel, at the very considerable time occupied in pursuing in oral evidence matters of individual intention and belief which could not reasonably be expected to have any significant bearing on the outcome. I must in due course consider whether that should affect any orders for costs which I may be asked to make. As to the impact of that evidence on the substantive issues I shall not add to its inflationary impact on time and expense by reviewing it in any detail, but in the next section of this judgment I shall refer briefly to that part which is properly material to the factual setting and later, when dealing with objections to my conclusions, I shall touch on those which rely on such evidence.

The way in which the Plaintiff's case was pleaded, and passages in at least one of the authorities to which I was referred, caused some attention to be paid in evidence and argument to the concept of reliance. As I understand it the elements of contract are those set out in the first sentence of paragraph 37 above, and no separate requirement of reliance is to be added to them. What is nevertheless true is that in the sort of collateral contract in question here, where acceptance and consideration are allegedly constituted by entry into a contract or contracts with a third party, "reliance" in a non-technical sense (or its absence) may be germane to the issues whether or not acceptance, consideration and contractual intent have been established.

The factual setting begins on the Plaintiff's side with the matters set out in paragraphs 1 and 2 above. One of the Fischer group's Swiss subsidiaries was a Zurich company called Oehler Whylen Lagertechnik ("OWL"), which specialised in the design, manufacture and installation of automated warehouse systems, in which capacity it was involved in the planning for Project GB. Both the managing director and European marketing director of OWL had come to know Mr Fitt of MDC when he undertook a project with them while he was a divisional manager with Balfour Beatty. In late January 1988 one of them, a Mr Broggi, telephoned Mr Fitt and described the Fischer group and its Project GB. He told Mr Fitt that the Fischer group had engaged a local architect to help in identifying a site in the Coventry area and in negotiating planning and other preliminary matters, but that when asked to take responsibility for the total project he had refused, as a result of which OWL had been asked who should be approached.

Mr Fitt reported this approach to the senior management of both MDC and the construction wing of the Multi group. He also sent to Mr Broggi the current MDC brochure, copies of which were passed to Fischer executives involved in Project GB. This included a "group profile", describing the principal activities of MDC, two Multi construction companies and a property development member of the group which does not feature in these proceedings, a group organisation chart, a description of MDC's "design and project management services", with examples of current projects, a section headed "Design and Construction" and a list of MDC staff. The "Design and Construction" section read as follows:

There followed a meeting on 10 February 1988 between four Fischer representatives, Mr Fitt and one of his subordinates from MDC and a representative of the construction wing of the Multi group. The Fischer representatives were Dr Seiler, then managing director of the Plaintiff, Mr Gunthardt, the project manager of Project GB, Mr Malkin, logistics manager of the Plaintiff and Mr Horton, project co-ordinator. Mr Gunthardt was based in Switzerland and was the link with the Swiss senior management of the Fischer group. Notes were taken by Mr Horton and copies sent to his senior colleagues and to Mr Fitt, apparently without demur. Persons present on both sides, including Mr Horton himself, gave evidence glossing that account in one direction or the other, but they were all speaking long after the event and with recollections inevitably coloured by hindsight. Mr Horton was in my view an honest, competent and conscientious note-taker and his notes are the best evidence of what happened. The relevant parts read:

The GF requirement was for a turnkey project which upon completion should achieve co-ordination between all aspects of the complex .... . MDC would be expected to manage the contract, liaising with system planners and co-ordinating all sub-contractors.

MDC to provide organisational documents showing relationships between MDC/MCL and the various stages of the project.

MDC to nominate one person to act as contact between MDC and GF.

MDC to provide references, contract details, brief details of their Company (including turnover) and staff who would undertake the project.

MDC to prepare a quotation for their fees, broken down into project stages. The quotation to show variances for both "tender" and "open book" options.

What MDC sought to suggest was that it was made clear that MCL, not MDC, would be the main contractor. I am satisfied that that was not made clear and that the Fischer representatives still believed the opposite. On the other hand "MCL" (albeit not the present company of that name) was represented and paragraph 2 of the note shows that some mention was made of relationships within the Multi group and further clarification promised. I have no doubt that the truth of the matter is that the Multi team were determined that eventually a construction company would be the main contractor, but they were in the first stages of playing a promising catch which had (i) a strong preference for a designer/contractor, and (ii) an attachment to and confidence in Mr Fitt which no-one on the construction side shared. Until it was closer to land they were not disposed to frighten it off by being too explicit. In any event the exact nuance of what was said on 10 February about "MCL"'s role is of no consequence, because as will appear Fischers were shortly afterwards persuaded to accept that a construction company would be the main contractor, and it is not in dispute that in the event the main contract was with MCL.

Five days later MDC submitted a document described in Mr Fitt's covering letter as "our proposal document" but in the Statement of Claim and commonly at the trial (perhaps to avoid confusion with the "fee proposal") as the "appraisal report". It is pleaded in the Statement of Claim as one of the documents containing the "representations" and founding the warranties relied upon, but in closing Mr Palmer for the Plaintiff, in my view rightly, relied only on the fee proposal of 22 March 1988 for that purpose. The appraisal report is nevertheless a significant part of the factual setting.

The introduction refers to the meeting of 10 February and states that MDC and MCL "have prepared this proposal document to outline how the joint companies would approach and manage the project". Section 2, headed "Project GB Appraisal" describes the project in general terms and states that it "is currently proposed" to undertake it in three stages: 1, general systems planning; 2, detailed planning; 3, project realisation. Section 3 is headed "Approach to Contract GB" and in 3.1, entitled "Contract arrangements", states:

Paragraph 3.2, entitled "Form of Contract", reads:

In section 4, headed "Project Responsibility", paragraph 4.1, entitled "General", includes the following:

In section 6, headed "Fee Proposals", design fees at stages 2 and 3 are shown as being "over and above the Guaranteed Maximum Price (GMP)". In section 8, devoted specifically to MDC, it is described as "an independent multi disciplined design consultancy within the Multi Construction (UK) Limited Group of Companies".

The appraisal report is a lengthy document and contains other passages relied by one party or the other, which I have taken into account, but it would unduly lengthen this judgment to set them all out here. In my view it was a transitional document in the process by which the Multi group was weaning Fischer from its desire for a main contract with MDC. On the one hand the proposal for one design and build contract with a construction company is clearly stated, but on the other hand there are repeated passages emphasising MDC's responsibility and independence.

Between 15 February and 22 March negotiations continued, Mr Fitt continuing to take the lead on the Multi group side. On 1 March Mr Gunthardt sent a telefax to Mr Fitt, stating that some questions had arisen after he and Dr Seiler had read the 15 February proposal and a later letter, "which I would like to discuss with you next Friday" (a meeting being projected for 4 March in Schaffhausen). They included:

Overall responsibility of MDC/MCL for operationability of all systems must be clearly defined (please new phrasing). ....

We would like if you brought into the contract a clause (at the end of stage 2):

It is common ground that there was a meeting in Schaffhausen on 4 March 1988, but no note or minute of it has been traced and the unaided and often confused recollections of the witnesses as to what was said, and even who was there, were in my view an unsafe basis for any but the most general conclusions. No doubt the points specifically raised in the telefax of 1 March were discussed, but the only reliable guide to the outcome is what was included in that respect in the fee proposal of 22 March and agreed. I find (and this is supported by Mr Cheetham's letter of 21 March below) that Fischers had by then accepted, or did so at the meeting, the principle that if the Multi group obtained an entire design and build contract the main contractor would be a construction company, not MDC. I find, since it is confirmed by Mr Fitt's letter of 9 March, that MDC's stage 1 duties and fees were discussed. With that letter MDC submitted its proposals for stage 1, for which it is common ground that it was the contracting party and billed and was paid directly.

On 21 March Mr Cheetham wrote to Mr Gunthardt in Schaffhausen with a first draft letter of intent to be written by "Georg Fischer" to Multi Construction (Northern) Ltd. The final draft of 20 April 1988 referred to in paragraph 30 above was clearly derived from it, but only after negotiation and amendment; in particular the first draft had no Appendix A, incorporating the fee proposal of 22 March.

That brings the story to the date of the fee proposal of 22 March. Another relevant mutually known fact, about which I need not review the evidence since I do not understand it to be in dispute, is that Fischers wanted design fees to be expressed as lump sums and included in the GMP, not added to it as in the appraisal report.

In that situation the salient facts, known to both sides (in addition to those, such as the nature of the project itself, which were obvious and are not in dispute), against the background of which the fee proposal of 22 March 1988 was submitted were in my view as follows:

For the most part these findings derive from the documentary record, summarised but not exhaustively extracted above. To some small extent I am assisted by the oral evidence, in particular in relation to (3), (4)(ii) and (6), where there was a substantial measure of concurrence between the admissible and relevant parts of what was said by the Plaintiff's witnesses and by Mr Fitt. As indicated in paragraph 38 above, however, I found a high proportion of the oral evidence as to the witnesses' attitudes, expectations and intentions of no assistance. In particular I was unimpressed by attempts on the Plaintiff's side to go behind or qualify the plain truth of the second limb of (2) and on MDC's to suggest that (1) and the second limb of (2) were incompatible with and excluded the possibility or credibility of (3) and (4)(i).

In my view the fee proposal of 22 March 1988 was plainly a response to the factual situation described in paragraph 54 above. The distinctive word "operationability" in the telefax of 1 March 1988 is reflected in the phrases "fully functionally operational" and "operational performance" in clauses 1.1 and 1.3 respectively of the fee proposal. The fees are quoted as lump sums "calculated in relation to" the duties and responsibilities in section 1.0 as well as to the design functions in sections 2.0 and 3.0. MDC is expressed to undertake a direct responsibility.

That being so it would have been ineffective for the fulfilment of its manifest purpose had it been devoid of contractual significance, whether as a deliberately worthless sham on the part of MDC or as a well-meaning but ineffectual piece of inept drafting. In my judgment it did not suffer from either of those defects. I find that it was an offer made with contractual intent and capable of founding, upon acceptance, a binding contract. I can see no other credible explanation for the inclusion of the words "shall be responsible" in clause 1.1, "will provide a warranty to the Employer" in 1.2 and "will be responsible to the Employer" in 2.1.

As to consideration, the fees themselves would not have that function if, as expected, they became part of the GMP under the main contract, but that is a familiar situation in collateral contracts of this kind (the locus classicus being Shanklin Pier Ltd v Detel Products Ltd [1951] 2 KB 854) and there is no difficulty in finding, as I do, that the consideration primarily invited was the entry into contractual relations with a construction company member of the Multi group, which would also constitute acceptance of the offer. Whether the offer would have remained open and unaltered had the construction contract gone outside the Multi group, and if so what the consideration would then have been, are questions which I need not answer, although the existence of such possibilities will be of some interest when I come to deal with the objections to my conclusions.

The performance promised by MDC plainly included, at least, the performance with due care and skill of the design functions specified in section 2.0. I must return to the question how far, if at all, it extended beyond that, but it is convenient to consider first whether there was such acceptance as to constitute a binding contract or contracts.

On that issue, although attention was largely directed in argument to the entry into the main contract on 10 April 1989 it is relevant to remember that by that time MDC had completed its stage 2 work, itself accounting for some 72% of its fees, and was some two months into stage 3. It is therefore important to consider under what contractual regime it had been working until then, and that turns on the effect of the letter of intent of 31 May 1988. That document, although only a statement of "intent", and therefore non-contractual, in relation to the prospective main contract, itself had the following important immediately binding contractual elements (see paragraph 31 above for the wording):

Appendix A was, of course, the fee proposal of 22 March as slightly amended (see paragraph 32 above), and I have no doubt that the entry of Fischers into the contract constituted by the letter of intent was acceptance of MDC's offer, as varied by Appendix A, and consideration for MDC's promises, so far as they were to be performed under the contractual elements of the letter of intent. It was also done in "reliance" on those promises in the sense discussed in paragraph 39 above, so as to establish contractual intent on Fischers' part. There was therefore a contract in those terms.

I do not overlook the fact that the primary contract formed by the letter of intent was with Multi Construction (Northern) Limited - that indeed is why entry into it constituted consideration - but MDC was plainly involved in the negotiations in such a way as to establish that it authorised both the inclusion of Appendix A in general and the variation of its wording from the document of 22 March in particular. Even without direct evidence it would be inconceivable, given Mr Fitt's key position, that he would not have been involved, but in fact there is ample direct evidence. On 15 April 1988 he wrote to Mr Masterson of DLE, who was negotiating on Fischers' behalf, as to the replacement of the lump sum fee for stage 3 by a percentage. On 19 April Mr Masterson wrote to Mr Gunthardt, with copies to Messrs Fitt and Cheetham, after a meeting that day with them both at which the wording of the letter of intent had been agreed, "with Appendices relating to the fee proposal and other matters", and specifically the design and planning fees for stage 2 and stage 3, about which he promised to write further. Mr Cheetham's letter with the "final draft" of the letter and appendices "agreed at our meeting" followed on the next day (see paragraph 30 above) and on 21 April Mr Masterson wrote to Mr Gunthardt, again with a copy to (inter alia) Mr Fitt, commenting in detail on the changes agreed at the meeting and incorporated in the final draft of (inter alia) Appendix A.

That dealt with the position until the main contract was entered into on 10 April 1989, when by its own terms the letter of intent was superseded and the contract with MDC collateral to it came to an end. On this occasion the fee proposal was not appended in full to the main contract, but it was in effect incorporated with further amendments by the relevant side letter. As amended it still plainly spoke in the name of MDC (see paragraph 36 above). There were still substantial design duties for MDC to perform, even setting aside for the moment the significance of the warranties of performance. Unless MDC can in some way dissociate itself from this further amendment and incorporation of its fee proposal it seems to me to be inevitable that the true analysis of the effect of the entry by the Plaintiff into the contract of 10 April 1989 with MCL must be similar to that relating to the letter of intent, namely that there was thereby brought into being a collateral contract with MDC.

In my judgment MDC cannot dissociate itself. In the first place it is as true of April 1989 as of May 1988 that it is inconceivable that Mr Fitt was not involved in anything affecting the contract terms in general and design responsibilities in particular. But in addition there are again positive indications that he was. On 20 February 1989 he was one of the participants in a Project Meeting between Fischers, DLE, MDC and MCL at which the breakdown of the GMP (a contractual document) was discussed and MDC undertook a number of tasks relating to cash flow plans, costings and a potential design change. On 23 March Mr Masterson wrote a letter to Mr Gunthardt, with a copy to (inter alia) Mr Fitt, enclosing a summary of the GMP, giving details of "late amendments" which include the design change and other matters discussed on 20 February. On 30 March Mr Fitt took part in a Progress Meeting between Fischers, DLE, OWL and MDC (but not MCL), at which the form of the contract conditions was reported upon, MDC were to provide final contract drawings to accompany the "Employers Requirement" (another contract document), MDC reported that the GMP had been agreed, detailed attention was given to design progress to date and the future design programme, MDC was to issue a "key personnel" update (Appendix E to the letter of intent) and MDC was requested to prepare a further cash flow forecast. On 4 April Mr Masterson wrote to Mr Fitt at MDC in response to what must have been an important letter raising issues of principle as to the operation of the deviations clause in the contract conditions, from which it is clear that Mr Fitt was still taking a leading role in the negotiations as a whole.

I therefore find that there was a further collateral contract with MDC on 10 April 1989, in replacement of the one expiring on that date, on the terms of Appendix A to the letter of intent as amended by the relevant side letter in Appendix 8 to the main contract.

I return to the question of the extent of MDC's obligations under these collateral contracts. Clause 2.1 of Appendix A made it responsible to the employer for the duties "set out below", which must at least include clauses 2.2 to 2.14 while the letter of intent was in force and the additional 2.15 and 2.16 after 10 April 1989. There is no express limitation of the standard required under these clauses to one of due care and skill, and some might as a matter of construction arguably impose strict obligations, but others (for example to obtain full planning consent) cannot sensibly be construed in that way, design obligations (which are those in this section of which breach is alleged) are normally understood to be to use due care and skill, the breaches of design duties pleaded are also alleged to be breaches of a tort duty of due care and skill, and the evidence and arguments as to breach of design obligations were all directed to the issue of compliance with that standard. I therefore consider that for the purposes of these proceedings that is the standard required by section 2.0.

Section 2.0 applied only at stage 2. The next question is whether the stage 3 duties and responsibilities set out in section 3.0 of Appendix A were incorporated (with, after 10 April 1989, the variations required by the side letter). There was, in fact, no submission by MDC that they should be distinguished from those in 2.0, but I should explain briefly why I agree. There are two reasons. The first is that clause 1.1 makes MDC responsible for the production of all designs, etc, required to "Procure and Construct " the works (my emphasis), and the function of sections 2.0 and 3.0 is to spell out and amplify that responsibility at each of those stages. The second is that for the reasons explained in paragraphs 26 and 27 above the clauses in section 3.0 are grammatically not separate sentences but a continuation of the list of items starting with 2.2 and all subordinate to 2.1. The standard of care was of course the same under section 3.0 as under 2.0.

I turn next to the issue of supervision. The Plaintiff pleads and submits that there was a duty to "inspect/supervise the construction of the project" to ensure compliance with design and good workmanship. I reject that submission. The only basis for it is note (a) to section 3.0 of the fee proposal, as set out in paragraph 27 above, but that provides that all site supervision will be carried out by the production team, "with support from [MDC] as required". Those words, especially when contrasted with those of clauses 1.1, 1.3 and 2.1, are wholly inadequate to impose any direct obligation by MDC to the Plaintiff.

The final issue as to the scope of MDC's obligations is whether there was a warranty of operational performance, having regard in particular to the words "[u]nder the form of contract proposed [MDC] will provide" in clause 1.3. By way of preliminary observation I note that on any view any such warranty related to "the completed works" and did not therefore arise at the date of the letter of intent, but only when the main contract was entered into.

The argument against the existence of such a warranty is that the words "under the form of contract" in clause 1.3 make the obligation conditional upon the inclusion of an express warranty by MDC in the main contract, and that that condition was not fulfilled. In my judgment that argument fails for two independent reasons.

The first is that both at common law and in equity a contract to enter into a contract is binding and is directly enforceable without the circuity of an action to compel entry into the (second) contemplated contract. The common law authorities are concerned to distinguish between the situation in which agreement is incomplete or "subject to contract", and thus not intended to create legal relations until all terms are agreed and a further document is entered into, from those in which all terms are agreed, the parties intend to be bound and the "contract" referred to is intended only as a more formal record. An example of the latter is Rossiter v Miller (1878) 3 App Cas 1124. Equity, subject to the same distinction, "looks on that as done which ought to be done". I have no doubt that in this case the offer constituted by the fee proposal was unconditional and required only acceptance, not the negotiation of some further contract. I reach that conclusion from its nature and the factual setting in which it was made, together with two more specific considerations. In the first place clause 1.3, in which the problematic words occur, does not stand alone. Just as clause 2.1 covers the same ground as the first part of 1.1 and should be read with it, so 1.3 has the same relationship with the second part of 1.1, which uneqivocally provides that "[MDC] shall be responsible for .... ensuring that the completed works are fully functional". Secondly the offer was effectively repeated, with variations immaterial to the present point, by the incorporation of the side letter varying it in the contract documents, when no further or different "contract" could have been contemplated.

The second ground for rejecting the argument that the warranty did not come into effect is that if, contrary to my conclusion above, the "form of contract" (in the shape of the main contract of 10 April 1989) itself had to "provide" the warranty then it did so by the incorporation of the relevant side letter in the contract documents with MDC's authority. It is true that on that basis a more meticulous attention to draftsmanship would have suggested the desirability of the express joinder of MDC as a party and of eliminating the future tense in clause 1.3, so that it would have begun in some such terms as "The Project Planners & Designers warrant to the Employer the sufficiency .... ", but the intention and effect is clear without that.

I have not encumbered my analysis above by detailed refutations at each stage of all the arguments to the contrary, but I propose in this section to identify and respond to the principal objections discernible in Mr Mauleverer's thorough and forceful submissions on behalf of MDC.

The first is based on article 8 of the main contract, summarised at the end of paragraph 34 above - a standard "supersession" clause. That fails on two grounds. In the first place the "parties" to it are plainly the Plaintiff, MCL and the latter's holding company, not MDC. Secondly the collateral contract with MDC was not a "previous" agreement; it came into existence at the same time as the main contract and by virtue of its formation.

The first of those grounds of rejection applies also to the next objection, based on clause 2.1 of the contract conditions, quoted from in paragraph 35 above - that the Plaintiff was "entirely reliant" on the skill and expertise of MCL for design. If surgeon A advises me that a particular operation is necessary and that he is the person best fitted to perform it, and if surgeon B, to whom I go for a second opinion, confirms both pieces of advice, then the fact that just before going under I express to A entire reliance on his advice does not absolve B if the operation was inappropriate, A was incompetent to perform it, and B should have known both facts.

I have already largely anticipated the next objection - that both parties intended a single, "turnkey", contract - in paragraphs 54 and 55 above. That was only one of the ingredients of the factual background; the others were equally important and not displaced by it.

Next, attention is drawn to the fact that under the main contract the design liability lay with MCL. But of course it did; that is fully taken into account in the discussion above, and is why MDC's contract was collateral to the main contract and not in substitution for it.

Then there is the length of time (just over a year) between the fee proposal and the main contract. But offers do not lapse, without express withdrawal, pending the very event contemplated as the consideration and the acceptance, and while active negotiations towards that event are proceeding between the parties. In any event the first acceptance, by the letter of intent, came within six weeks. And, so far from withdrawing its offer, MDC repeated it with minor variations on the occasion of each acceptance.

The above points were all made in MDC's opening and repeated in closing submissions. There is next a rather different and somewhat surprising objection which was made for the first time in closing, although it was foreshadowed by questions to witnesses. It was that the promissory words in the fee proposal relied upon by the Plaintiff were, if an offer at all (which was of course denied), an offer conditional upon Fischers' engaging a main contractor other than MCL on a contract other than a turnkey contract. In my view that flies in the face alike of the document, the facts and common sense.

As to the document, the fee proposal manifestly contains no expression, or even hint, of any such condition. As to the facts, it is of course true that the possibility of a main contractor outside the Multi group was in the minds of both sides, as recorded at paragraphs 50 and 54 (4)(ii) and (6) above, but the hope and expectation of both was that terms could be agreed between them, and the fee proposal was framed with that in mind, as the very phrase relied upon by MDC in clause 1.3 ("the form of contract proposed") shows. It had, indeed, three main functions. By the nature of the dispute before me I have hitherto concentrated on its role as conveying MDC's offer. As its description as a "fee proposal" implies, however, one of its primary functions was to specify the lump sum design fees to be included in the GMP. And, thirdly, it was intended to be, and by incorporation in the letter of intent and main contract became, the specification of design responsibilities under those contracts, which were contracts with the construction arm of the Multi group. The second and third of those elements applied as they stood only if MCL or another Multi construction company were the main contractor. So far from the fee proposal's being directed towards the possibility of there being another contractor, therefore, crucial parts would have had to be omitted or renegotiated in that event. That possibility was indeed catered for, but not in the fee proposal; it was the subject of the provision in the letter of intent for completion of the design, if Fischers so required, notwithstanding the loss by the Multi group of the main contract. As to common sense, the supposition that MDC would have been prepared to warrant the performance of an unknown contractor outside the group but not that of its own construction arm seems to me entirely fanciful.

A great deal of oral evidence was cited in support of this objection, but in my view it was of little or no assistance for the reasons given in paragraphs 37 and 38 above, and certainly of no weight in comparison with the considerations rehearsed in the last paragraph.

The final objection which I feel the need to deal with, also raised in closing submissions, and founded largely on a passage in the cross-examination of Mr Gunthardt, was that any suggested commitment by MDC was superseded by the holding company's guarantee and the performance bond. The fallacy here is to assess the motivations and values of 1988 and 1989 with the hindsight of 1994 to 1997. MDC is now an independent company and is clearly considered to be financially worth suing. Whether that is because it is prosperous, or insured, or both, is not for me to enquire, nor is it relevant to the present point. MCL and, presumably, its holding company, were in the event not worth suing for the reasons given in paragraph 12 above. It is therefore natural, now, to see MDC's disputed obligations in a financial light. But in 1988 and 1989 MDC was merely a fellow-subsidiary, with the construction and development companies, of a UK holding company which was itself, with many other companies all over the world, a subsidiary of a foreign "Societe Anonyme". No-one in his senses would at that time have looked to it for financial backing of MCL's performance. It is quite clear from that consideration on the one side, and from the facts set out in paragraph 54 on the other, that what Fischers wanted from MDC was not a financial fallback but the staking of its professional reputation on the outcome. By the same token the Multi group in general, and MDC in particular, had no reason at that time to regard the latter's commitment as adding anything in financial terms to the obligations or risks which would in any event arise under the design and build contract.

It is against that background that the passage in Mr Gunthardt's cross-examination relied upon must be assessed. It was as follows:

That would be a wholly unsatisfactory basis for a conclusion contrary to that indicated by the considerations set out in paragraph 82. First there is the general inferiority of distant recollection to contemporaneous documents. Secondly, even at face value, the remark about the warranty by the parent company is clearly infected by the "hindsight fallacy" described above. Thirdly it is far from clear that it can be taken to have been intended in the sense relied upon by MDC; Mr Gunthardt immediately goes on to equate the participation of the Multi parent company with the change of participants on the Fischer side. Fourthly "we were obviously satisfied" is not the language of direct recollection but of reconstruction, which brings one back to the effect of hindsight. Fifthly such a reconstruction of his own state of mind is at one further remove from that of other members of the Fischer management, even if an uncommunicated state of mind could prevail over what passed between the parties. Finally, this single remark must be set against the general tenor of Mr Gunthardt's evidence, which was to the opposite effect; almost immediately after the above exchange Mr Mauleverer put it to Mr Gunthardt that the fee proposal was overtaken by the letter of intent, and subsequently by the contract, to which Mr Gunthardt replied:

I therefore find nothing in these objections to invalidate the conclusions reached in the preceding section.

I have already pointed out that no case founded on a concurrent duty in tort is needed or argued. I am therefore concerned in this section only with the questions whether, if I were wrong in my conclusion that there was a collateral contract, MDC would have owed a duty of care to the Plaintiff at common law to protect it against economic loss from defective design and whether, in any event, it did owe such a duty in respect of supervision (which I have found not to be within the scope of the collateral contract). In my judgment it would not and did not. (I disregard the warranty of performance, which although technically within the pleading could not conceivably have been the subject of a tort duty.) With the exception of Junior Books Ltd v Veitchi Co Ltd [1983] 1 AC 520 the authorities are uniformly against the imposition of such duties on the part of a sub-contractor in favour of the employer of the main contractor, and in the light of the subsequent treatment of that case in the appellate courts, including the House of Lords itself, it would be a foolhardy first-instance judge who treated it as other than one turning on its own peculiar facts.

Nor do I feel tempted, in any event, to do so. I understand and accept the force of the arguments advanced by Mr Palmer, for the Plaintiff, that there was a voluntary undertaking of responsibility, but they turn essentially on the facts which go to establish a collateral contract and I have accepted that as to design they do so; if that is wrong it must be so on the basis that there was no direct contractual relationship with MDC, as is in any event true of supervision. The conclusion must be that the Plaintiff accepted the familiar contractual regime of main contract and sub-contracts, under which it would look to the main contractor alone for redress notwithstanding that the breach complained of might lie at the door of a sub-contractor.

Part of the damages claimed by the Plaintiff consists of the cost of placing netting over parts of the roof at particular risk of being blown off. The netting would not avert the failure of the allegedly deficient attachments but would prevent the loose panel or roof light from being blown clear of the building, to the danger not only of persons on the Plaintiff's premises but, it was feared, of drivers, passengers and vehicles on the adjoining motorway. Failing establishment of any general liability for economic loss Mr Palmer argued for an exception covering this expense, in reliance on a passage in the speech of Lord Bridge in Murphy v Brentwood District Council [1991] 1 AC 398 . Having given reasons at page 475A justifying the distinction between a manufacturer's liability in negligence for injury to persons or damage to property caused by defective goods from his absence of such liability for the purely economic loss suffered by the owner if the defect is discovered before any physical harm is done Lord Bridge continued: "I believe that these principles are equally applicable to buildings", and developed that proposition, but continued (and these are the words relied upon by Mr Palmer):

The only other of the seven law lords who heard Murphy to comment on that point was Lord Oliver, who reserved his opinion, adding: "although I am not at the moment convinced of the basis for making such a distinction" (page 489C). As Lord Bridge himself recognised this was a "qualification" of the otherwise general principle which he was propounding. The other reasoned speeches all proceed on the basis of that general principle without qualification. That it is anomalous to award damages for a realised injury but not for the (usually lesser) cost of averting it was explicitly the ground of the decision in Dutton v Bognor Regis Urban District Council [1978] 1 QB 373, and in overruling Dutton the House of Lords in Murphy equally explicitly rejected the claim of that argument to prevail. It is difficult to see why it should linger on where the danger averted is that of liability to a neighbour or passer-by rather than of injury to the Plaintiff himself, damage to his property or liability to his employees, customers or visitors. In my understanding the passage quoted is properly to be regarded as a minority obiter dictum, contrary to the ratio of the decision of the House. I therefore respectfully disagree with His Honour Judge O'Donoghue, who in Morse v Barratt (Leeds) Ltd (1992) 9 Const LJ 158 adopted and applied Lord Bridge's "qualification" as a statement of the law.

I accordingly reject the Plaintiff's claim against MDC in tort based on a duty of due care and skill in supervision and would have rejected a similar claim in respect of design had it arisen for lack of a finding that a collateral contract existed.

As noted in paragraph 17 above DLE were engaged by the Plaintiff as Employer's Representative by a contract in writing between them made on 11 January 1989. It provided that DLE should provide in relation to the project works the following services:

Those provisions are for the most part plain and self-explanatory. Mr Coulson for DLE stressed the point that responsibility for design appraisal under (i)(a) was confined to the pre-contract period, which is true but of little practical moment, since the responsibility under (ii) for such matters as approval of substitute materials, consent to deviations, consent to additional drawings and details and instruction of variations equally required appraisal of design considerations. Similarly the restriction of the first part of (i)(a) to what was "available" and "provided by the .... contractor" cannot avail DLE if inadequate design material made the proposals "unsatisfactory". The frequency of visits required to comply with (iii) was obviously a matter for professional experience and judgment, but only subject to the governing words "sufficient to monitor the Contractor's workmanship and progress".

In opening Mr Coulson relied on a letter from Mr Venning of DLE to Dr Seiler of 22 November 1988 which disclaimed liability for design or performance and asserted that DLE's employer's representative and quantity surveying roles were mainly to act as the "communication link". That cannot alter the effect of a formal contract in writing entered into six weeks later, nor indeed is it pleaded as doing so. The same is true of other assertions by DLE staff in correspondence or in the witness box, of which there were many, amounting to flat denials that DLE had duties plainly laid on them by their contract, such as the duty to consent to drawings (which means, of course, to consent to them so far as they were in accordance with the construction contract and with good design practice).

An estoppel founded on the same sort of evidence was pleaded but not, as I understood it, pursued. In so far as it was I reject it.

If, as I have found, MDC warranted the sufficiency and operational performance of the project works then distinctions between design, workmanship and supervision fall away, as do issues of care and skill. That cuts both ways. On the one hand negligence in any of those respects is not a necessary condition of liability if an alleged defect exists and affects the sufficiency or performance of the works. On the other hand if an alleged defect has no effect on sufficiency or performance that excludes liability under this head, whatever the position on a different kind of claim. There is also, of course, no liability if the alleged defect does not exist at all, whether because the Plaintiff has not established the facts relied upon or because those facts do not show any non-compliance with the contractual requirements.

The major complaint, that the roof leaks, is not denied. That is without question a defect and a failure of sufficiency and performance and there is therefore no need to consider further the Scott Schedule items alleged to have contributed to it, which as I understand it are numbers 1 to 6.

Of the remaining items numbers 12 and 15 were not pursued, and numbers 11, 13, 18, 21, 22, 23 24 and 25 are, in my understanding, admitted on the test to be applied here, having regard to the experts' joint statement under Order 38 of agreements and disagreements and the document furnished at my request by MDC after closing submissions. I therefore turn to consider those not yet listed. In a trial largely occupied with the issues already discussed and the questions of design and supervision responsibility for the leaking roof it was sometimes difficult to strike the right balance between giving these other Scott Schedule items quite cursory individual attention on the one hand and incurring disproportionate time and expense on the other; I shall try to express my conclusions with reasonable brevity and shall certainly not pursue every detail of the evidence and argument.

Item 7 alleges that the junctions between the lower roofs and the high bay warehouses have been so designed that there can be deep snow drifts, producing large roof deflections and areas in which water will collect. I am not clear how far this item was persisted in, since it is not separately dealt with in Mr Palmer's closing submissions. However that may be I reject it. The Plaintiff's own expert evidence was that the design complied in this respect with the relevant code of practice and British Standard guidelines and had sufficient reserve strength to avoid collapse from snow drift loading. The conditions envisaged are exceptional, and although the possibility of "irreversible" distortion is pleaded that was not made good by the evidence. In those circumstances it would have been uneconomic to over-design for an exceptional and temporary situation which could be met by temporary, ad hoc, counter-measures, and not doing so has not made the building unserviceable.

Item 8 accepts that some provision was made for drifting snow, in that intermediate purlins were added in vulnerable areas. The allegation is that they were not fixed to the roof cladding (and that is true, although they were end-fixed to other elements of the steel structure) and that in consequence their load-carrying capacity was significantly reduced and they were liable to collapse. It appeared, at least initially, that the form of collapse envisaged was one in which a purlin, being much deeper than it is wide, buckled or twisted under a vertical downward load on its upper face by partially rotating in the centre of its span, relatively to the fixed ends, around a long axis at or below its (vertical) mid-point. That would involve a displacement of its upper face relatively to the underside of the panel, which a fixing would prevent. But, said MDC's expert, Mr Pepper, friction would have the same effect without the need for a fixing. That was contested and calculations were carried out on both sides. To explore the technical issues in depth would have involved disproportionate time and expense and the parties accordingly invited me, at a certain point in the evidence, to give a provisional impression of my then view, which I did. The matter was not further pursued. In the circumstances I consider that the proper course is to adhere to that view, from which in any event I see no reason to depart. For present purposes it was that, whatever the competence or otherwise of the design, Mr Pepper was correct in his opinion that friction would prevent lateral movement of the upper flange of the purlin and therefore buckling in the mode described above, and that it was not open to the Plaintiff on the pleadings and reports exchanged to rely on other postulated modes of failure. The Plaintiff therefore fails to establish any breach of warranty under this item.

Items 9 and 10 arise out of the undisputed fact that although the clips holding the roof cladding in place should have been fixed to the purlins by three screws each they were not. Those which had only one screw are the subject of the admitted item 11. Items 9 and 10 cover the remainder, which had two. That was plainly a breach by the contractor. The issue here is whether it affected the sufficiency or performance of the building. It is alleged to have done so in two ways, not very clearly distinguished in the two items. The first is that two screws are inadequate to provide lateral stability to the purlins. There was in my view no credible evidence to that effect, which is not surprising in the light of the state of play on the not dissimilar issue considered under item 8 above. The second concerns the adequacy of the fixings to resist wind uplift forces. In the evidence of the Plaintiff's expert, Mr McEwan, criticism was confined to the perimeter areas of the low-bay warehouses, where the wind-induced forces would be greatest. His conclusion that in those areas two screws were inadequate was attacked on a number of grounds. It was said that he had wrongly assessed the "ground roughness" factor at 2 instead of 3. I reject that criticism; on the sample photographs illustrating each factor I consider his assessment justified. It was said that he should not have assumed that the delivery bay doors would be left open during the very high winds postulated. I disagree; a building which is safe only if staff are always available to take such precautions and always act prudently and promptly is not adequate. It was said that he personally had carried out no tests, which was true, and that the data and calculations on which he relied were inadequate. The other experts, however, had done even less and although Mr McEwan's reliance on the manufacturers' data for a two-screw fixing (as distinct from two screws in a three-screw fixing) and on someone else's calculations was by no means ideal it was the best evidence on the point before me. Given that fact and the starting-point that fixings with two screws where three are specified are likely to be inadequate I find for the Plaintiff on this issue.

Item 14 alleges that the roof gutters are undersized, that they have sagged and that their outlets are too small. There was no evidence of undue sagging but I accept Mr Kilbey's evidence that the gutters and outlets were too small and find this item proved as a breach of warranty in those respects.

Item 16, on the face of it, complains that the vertical cladding to certain parapets, which has an appearance reminiscent of box pleating, has the "pleats" running horizontally rather than vertically, as they do on the main external wall surfaces.

Item 17 alleges that in specified locations insulation does not continue from the roof across the cavity to the parapet. That is admitted and I therefore find it established. MDC seems to fear that there is implicit in another column of the Scott Schedule a separate allegation of an absent vapour barrier, which they say is not a defect. In so far as there is such a separate claim I reject it.

Item 19 arises because, as is not disputed, internal block walls have been built up round the purlins to meet the roof cladding, without insertion of the movement joints shown on the drawings. That was plainly a breach by the contractor. Despite some evidence from DLE's expert, Mr Billington apparently casting doubt on the need for a movement joint I am clear that its inclusion was a requirement of good practice. The only real argument against the conclusion that this was a breach of warranty of performance was that no overt cracking or other physical damage had yet appeared. I do not consider that that is sufficient to displace the inference that the building was inadequate in this respect and that this item is therefore established.

Under item 20 three separate defects are alleged. The first, that the external steel ladders giving access to the upper roof levels are not galvanised as required, is admitted. The second, that they rest on the roof cladding, is not strictly true, since there are small base plates, but it is correct that the cladding bears the load; there is no method, either in design or execution, for carrying the load to a structural feature. That may be conceptually inelegant but I am not satisfied that it amounts to a breach of warranty. The load is slight and infrequent, the cladding is well able to support it and alternative treatments, such as penetrating the cladding to reach a structural support below, are likely to be more objectionable. The third alleged defect is that the ladders are inadequately secured by fixing bolts which pass through the wall cladding without adequate weather proofing and hook over internal cladding rails. This is a small matter, but I find it proved. I have used the ladders. The lack of weather proofing is evident and, although not dangerous if negotiated with care they are certainly not as firm as they should be.

That leaves the last item, 26. The complaint here is that rainwater down pipes were not protected near floor level from damage by warehouse traffic. The position is that pipes in areas which were always intended to be exposed to such traffic were protected. The dispute arises because that was not done along one wall where, the Defendants say, the design use was as a manufacturing area for "parking" (the use of steel tubing to make parking meter posts), so that the down pipes would be behind work benches and out of the way of any traffic. In the event that use was relocated, the area adjoining the pipes became open and they were damaged. I find this item proved. The contractual provision for protection was general and unqualified, the contractors obtained no variation order excluding the area in question, the internal layout and use allocation of a large open space like this was always liable to change over time, and the truth is that the contractors simply hoped to get away with an omission, as they did.

Since the argument for the inclusion of a warranty of performance in MDC's collateral contract requires one additional stage beyond that for the establishment of its basic duty of due care and skill in design (see paragraphs 66, 67 and 69 to 72 above) it is possible that I am right about the existence of the latter but wrong about that of the former. I think I ought, therefore, to make findings on the allegations of breach of MDC's design duties. I must in any event do so on the allegations of breach of DLE's duties as employer's representative. They are best considered together, since so much of the factual background is common to both.

I shall consider first the major group of issues concerned with allegations about design, workmanship and inspection in relation to the inclusion and construction of end laps and their contribution to leaking. Then I shall turn to the remaining Scott Schedule items, starting from the findings in the preceding section, where applicable.

I have summarised the basic facts in paragraph 5 above. It is convenient to begin by making good the last sentence of that paragraph, since although much of the trial proceeded as if it were common ground it was not clear to me that it was ever formally conceded, and indeed it seemed to be challenged by some of the cross-examination and argument.

The best starting point is the reports of three early inspections. The first was by Mr Scupham, the manufacturer's technical services manager, on the morning of 16 February 1990. His report begins: "Leaks at numerous endlap positions were in evidence." The second was on the afternoon of the same day by Mr Murfin, an architect employed by MDC, but not on this project. He reported that the end laps did not fit tightly together and thought it highly unlikely that the sealing of the joints would prevent water ingress. It was decided by MCL and accepted by MDC that that report should not be dislosed to the Plaintiff or to DLE. The third was on 13 July 1990 by Mr Smith, a building design consultant instructed by MCL. By that time external taping had been applied as an attempted remedial measure. He tested the incidence and location of leaks by throwing buckets of water over the roof in a number of locations, including end laps covered by the remedial tape. In his report, after an extensive discussion, he summarised his conclusion as being that the leaks had occurred due to inadequate sealing in the end laps.

Amidst much other relevant evidence it is sufficient for my present purpose to record secondly that of one of the expert witnesses called by the Plaintiff, Mr Kilbey, a roofing specialist, who out of a total of over 500 lap joints (DLE say 516) tested 50 (ten on each of the five roofs, at random points where the external taping was already disrupted) by sliding a hacksaw blade under the overlapping weathering sheet, usually at three or four points. In the case of every lap tested the blade penetrated at one or more points the full 100mm depth of the joint to the insulation core of the downslope panel without being prevented from doing so by the internal seal which should have been present, filling the gap. Sometimes there was no resistance at all, indicating the complete absence of a seal; sometimes there was some resistance, either from muck or dirt or from a seal which was present but not filling the gap. The majority of the failures were "around either side of the bend which goes up to the standing seam", but some were in the flat section of the lap. I accept that evidence. Mr Kilbey took it as showing that substantially all the lap joints were defective, and although he was extensively cross-examined by both Mr Mauleverer and Mr Coulson about many aspects of these tests the validity of that conclusion, given acceptance of his results, was not challenged in terms either of common sense or of mathematical statistics, nor could it be; the relevant mathematics confirms that it is statistically sound.

Finally there is the experts' joint statement under Order 38, rule 38, indicating those parts of their evidence on which they were, and those on which they were not, in agreement. It records (inter alia) the following, in which as I understand it the experts consulted by all the present parties concurred (I have ignored reservations or disagreements on the part of Roofdec's experts and additional comments by individual experts not concurred in by others):

That being the position the next question is whether that state of affairs was caused or contributed to by a lack of due care and skill in the performance of MDC's duties as designer. It is pleaded that the designed fall was too shallow in itself, but I did not understand that to be pursued and if it was I reject it; although 1.5 0 is at the lower limit of slope for pitched roofs to function as such it is possible to design and erect a watertight roof at that pitch. The gravamen of the complaint which MDC has to meet is the combination of such a low pitch with the use of end laps, either at all or in such numbers and in such positions, and without special provision for rigorous quality control.

At the relevant date recommendations were available as to the use and location of lap joints in shallow pitched sheet metal roofing of this kind, to which reasonably competent and careful designers would have had regard. In British Standard 5427 of 1976 (the Code of practice for Performance and loading criteria for profiled sheeting in building) the recommendation for lapping sheets at pitches of less than 5 0 was "consult manufacturer". In a guide to good practice published in 1982 by the National Federation of Roofing Contractors entitled "Profiled Sheet Metal Roofing and Cladding" it is advised that designers should adhere to the limits of pitch and methods of lapping shown in a table which, after setting out recommended lap characteristics against pitches classified as more than 15 0 , 15 0 -7 0 and 7 0 -4 0 , reads: "No pitch less than 4 0 ". The manufacturer of the system used, Ward "Moduclad", in its technical handbook dated March 1988, stated "Roof pitches below 5 0 require special consideration. Where possible laps should be avoided .... ."

Those warnings added weight to what would in any event have been to a competent designer the obvious difficulty of achieving a watertight roof at such shallow pitches if joints were introduced in the form of end laps. Much of the argument and cross-examination by the Defendants was based on the implicit assumption that since Ward's handbook contemplated the possibility of pitches less than 5 0 any such pitch was equally permissible and it made no difference how much less. That is manifest nonsense. Recommendations of this kind are necessarily couched in terms of bands, with upper or lower limits or both, but the gradation of physical properties and consequences is continuous. The designed slope here was at the absolute minimum and all the warnings and contra-indications against its combination with a system employing lap ends had their maximum force.

The questions which arise, therefore, are whether it was possible to have avoided the use of end laps and, if not, whether the design could have provided for fewer of them and in better positions, and whether any other steps could have been taken to remove or minimise the risk of leaking. I say "possible" and "could" rather than "desirable" and "should" because in the light of the considerations set out in the last paragraph I find it plain that if any of these steps were possible they should have been taken.

MCL's specification of February 1989 in the "Contractor's Proposals" incorporated in the contract documents of 10 April 1989 provided for roofing by "'Speed Deck' or similar". "SpeedDeck" is the trade name of a metal roofing system manufactured by Stramit Industries Ltd. Like Ward's Moduclad it consists of upper and lower metal surfaces separated by an insulation core. The crucial difference for present purposes is that panels are made up on site instead of in the factory, so that there is effectively no upper limit to their length and the need for end laps is completely eliminated. The specification also required the contractor to state in his tender the size of all laps and jointing materials, if any, he intended to use.

The roofing sub-contract went out for quotations on the basis of that specification. One quotation was for Stramit SpeedDeck, one for Ward's Moduclad and one for a third system, Ash and Lacey's "Dekex". A fourth quotation, from Roofdec Ltd, was for SpeedDeck, but in its covering letter Roofdec wrote:

Roofdec's quotation was accepted in its alternative form, using Moduclad. Mr Fitt's evidence, which I accept, was that MDC was "clearly under instructions from MCL that we should pursue the Roofdec option because it was cheaper", and he was under the impression that that was part of an exercise which was undoubtedly undertaken to bring the GMP within Fischers' budget, but the chronology does not by any means clearly establish that point. The budget overrun had emerged at a meeting on 21 November 1988 and it seems to have been after that that SpeedDeck was first specified, in place of MDC's original proposal of Moduclad, so at that stage it may have been thought to be the cheaper. That the subsequent reversion to Moduclad was before the GMP was finally agreed on or about 10 February 1989 seems doubtful; Roofdec's original quotation was dated 25 January 1989 and the revised version 21 February, but the GMP cost breakdown prepared on 23 March 1989 still shows the proposed roofing sub-contractor as Caunce (the company which had quoted for SpeedDeck alone, and had, as it happens, also expressly complied with the requirement as to the use of laps by stating there would be none), and the main contract of 10 April itself, as already stated, still specified "'Speed Deck' or similar". I was not taken through the roofing sub-contract documents and do not know the date of the sub-contract. Unless the change to Moduclad affected the GMP the reduction in cost was not necessarily or wholly for the Plaintiff's benefit.

What is clear is that it was not suggested that the Plaintiff was a party to the decision or consulted or advised as to any potential implications it might have for the watertightness of the roof. I was told that the sub-contract entered into by MCL with Roofdec made Roofdec responsible for detailed design, but the implication of MDC's evidence about what happened when detailed drawings were received from Roofdec was that until then there had been no specification or discussion about whether there should be lap joints, or if so how many and where, but that it remained within MCL's power (in practice exercised by MDC) to approve or reject Roofdec's proposals in that respect. In any event nothing in the sub-contract could affect MDC's obligations to the Plaintiff.

As to the panel lengths available in Moduclad, Ward's technical handbook showed drawings of "Roof panel options" as part of a section on "Component dimensions (roof)", with a length of "Up to 24,000[mm]" and a sidenote "Longer lengths are available, please contact WBC Sales Office", and Mr Fitt confirmed that his understanding was that the normal maximum was 24 metres but that longer panels could be specially made. Mr Scupham said that Ward could manufacture and supply panels up to 30 metres long, but that the longest panel he could personally recall was 27 metres. The average was between 11m and 15m. There was no evidence of any specific length above which special arrangements would have to be made for road transport; Mr Scupham's understanding was that "the transportable length varies to suit road requirements such as police awareness and transportation generally on the roads". Having been referred to concerns advanced by the Defendants as to the safety of handling the longest panels he was asked:

In that situation Roofdec's proposals for panel lengths and lap joints appeared for the first time in the form of drawings dated April 1989 but apparently received by MDC on 22 May. They provided, on one bay of the high level roof, for panels 9.280m, 9.025m and 6.935m long successively from top to bottom, a total of 25.240m for the complete slope, from which some small deduction is needed for the overlaps at the two joints, so that a single panel would have needed to be about 25m long (the dimension of 24.560m shown cannot be trusted, because (i) it is not taken to the edge of the lower panel, and (ii) it may be a horizontal distance). On the other bay there is an upper row of panels shown as 4.645m long and a lower row shown as 8.670m, although neither dimension seems to be accurate, because they scale at about 5.125m and 8.875m respectively (without overlap), which accords better with the indicated and scaled total of about 14m, that being therefore the length of panel needed to avoid lap joints.

The drawing for the low bay warehouse adjoining the office block shows on one slope successive panel lengths from top to bottom of 9.275m, 9.700m and 9.540m, a total, including overlaps, of 28.515m, corresponding to a single panel of about 28.2m. On the other slope there are two rows, the upper 10.280m long and the lower 10.935m, a total of 21.215m with one lap or about 21m for a single panel. The roof of the other lower bay warehouse also has two slopes. One has an upper row of panels 8.755m long and a lower one of 5.255m, giving a total with overlap of 14.010m and a length of about 13.8m for a single panel. For the other slope the corresponding figures are 9.155m, 4.980m, 14.135m and about 14m.

I have not found Roofdec's drawings for the two mid-level roofs, but as I understand it both were smaller and with slopes no longer than about 14m; nevertheless both had two rows of panels on each slope.

It was when MDC's project engineer, Mr Brooks (who unhappily died in February 1990) came to him with those drawings on 22 May 1989 that Mr Fitt was first involved with the question of lap joints. The occasion was not referred to in his witness statement, but in oral evidence in chief he was asked what Mr Brooks told him, and replied:

Mr Coulson cross-examined Mr Fitt on this subject:

Finally, on this topic, Mr Fitt was cross-examined by Mr Palmer:

There was no re-examination on this point.

I have reached the conclusion that MDC was in breach of its duty to exercise due care and skill. It was perfectly possible, and indeed straightforward, to have avoided the use of end laps by accepting one of the quotations which were based on the use of the SpeedDeck system specified by MDC itself in the contract documents. The cost saving advanced as the motive for the change fails to justify it on a number of grounds, each sufficient in itself. In the first place it was at best insignificant in comparison with the risk incurred. Secondly it did not arise on the cladding system itself, which was actually more expensive, but on the roof lights, and there was no evidence of any investigation into the merits of that alteration or into the possibility of achieving the same end by other means. Thirdly it is not established that the saving was for the Plaintiff's benefit.

Even if, contrary to that conclusion, the use of Moduclad is to be treated as justifiable the manner of doing so was not. All the slopes could have been covered without laps within Ward's maximum panel length of 30 metres. Even if the "normal" maximum of 24 metres had been the limit the only laps needed would have been on the two longest slopes, and there they could have been quite close to the ridge, so minimising the water flow to which they would have been subject. Even the "average" length of 11 to 15 metres would have dealt with the 14-metre slopes; indeed MDC's consent to the insertion of lap joints on those slopes is incomprehensible. If some minimum number of laps had been reasonably necessary MDC could and should have warned the Plaintiff of the consequential dangers and advised of the necessity of arranging for the closest and most rigorous inspection and supervision of the process of making the joint seals.

It is argued for MDC that the defects in the lap joints were purely matters of bad workmanship; reliance is placed on the experts' agreement that the cause of water ingress is that lap seals have not been properly installed and on evidence that a perfectly made seal will not admit water, even at slopes of as little as 1 0 . But to make every lap joint perfectly is difficult and expensive and unlikely to be achieved by the ordinary standards of workmanship and ordinary levels of supervision which suffice in less extreme conditions; that is precisely why the recommendations summarised in paragraph 114 above were current at the relevant time, were particularly important at such shallow slopes as this and should have been followed here by any competent and careful designer. This argument is similar to that discussed in paragraph 115 above and is open to the same criticism.

Much play was also made by MDC of Mr Scupham's answer to a question by me:

I turn to consider the position of DLE, and need not repeat the facts and considerations which are common to both claims. The claim against DLE has two aspects, one arising out of their duty under heads (i) and (ii) in paragraph 91 above, the other out of that under head (iii).

At the stage of pre-contract appraisal of design under head (i) the only relevant provision was the specification of "'Speed Deck' or similar", which there was no reason for DLE to question. Under head (ii) DLE had, inter alia, to "consent to deviations" and "consent to additional drawings, details and other documents". In the light of the words "or similar" I do not consider that the use of Moduclad was in itself a deviation, but its use with the number and location of end laps proposed by Roofdec was, and it was moreover for DLE to give or withhold their consent to Roofdec's drawings showing those details. For the reasons given above in dealing with the claim against MDC they should have withheld that consent.

There was some dispute as to whether Mr Gardiner of DLE, who dealt with such matters, actually saw the material drawings, but I do not find it necessary to resolve that dispute, since it was undoubtedly DLE's duty to obtain and approve all working drawings, although Mr Gardiner and the partner to whom he was answerable, Mr Carter, had managed to persuade themselves and represented to others that they had no concern with the approval of such drawings for construction purposes. It is equally irrelevant that Mr Gardiner may well not have had the professional training or experience to assess the acceptability of the design, although Mr Carter equally mistakenly took the view that there was no obligation on DLE to engage sub-consultants where its own staff had insufficient expertise to discharge its responsibilities.

As to head (iii) DLE made no visits to the roof whatever during the period when the panels were being laid and the lap joints formed, so they were undeniably in gross breach of duty. Mr Gardiner's only excuse for that omission was that access was not safe. That is obviously no answer; he was entitled to require the contractor to provide safe access.

The only defence seriously advanced in respect of that breach was that the defective formation of the seals at the lap joints would not have been detected even had DLE carried out inspection visits, because the work of making the seals would not necessarily or probably have been going on during visits, and if it were the workman would have taken untypical care while under the eye of the employer's representative. That defence fails at every level. As to missing the relevant phase of the operation, first, it is clear on the evidence that on visits of the frequency and length which Mr Gardiner says he carried out elsewhere on site, and even without any special attention to this point, the likelihood is that he would on many occasions have had the opportunity of seeing lap joints formed and sealed. Secondly, the formation of the joints was so obviously crucial that even if the overall frequency of visits was not increased special attention should have been paid to ensuring that they fully covered this aspect. But, thirdly, since this whole discussion predicates the acceptance by DLE of the very risky and inadvisable inclusion of lap joints in such shallow slopes, it was incumbent upon them to exercise the closest and most rigorous inspection and supervision of the process. The last point also disposes of the suggestion that workmen will "put on a show" - either they cannot do that all the time, or if they do that achieves the object anyway. Moreover it is any event part of the necessary skill of a competent inspecting officer to detect and make allowances for such behaviour.

I therefore find both MDC and DLE responsible for the major contribution to leaking roofs made by the inclusion and defective construction of end laps in the roof panelling. The truth is that DLE are in a dilemma which I put to Mr Coulson in the course of his closing address (and which also affects MDC, despite the absence of any supervisory responsibility on its part, for the reason given in the last sentence of paragraph 130 above) as follows:

On this second tour of the Scott Schedule that disposes of items 1 and 4. As on my first visit, I can deal with the remainder much more briefly. In particular, where I have found in paragraphs 97 to 106 above no breach of warranty because the sufficiency and performance of the building are unimpaired, then although there could still theoretically be a breach of the duties here under consideration the damages would be only nominal, and I do not propose to go into such issues unless the parties persuade me on the appointment to hand down this judgment that I should do so. That disposes of items 7 and 8 and parts of 14,16, 17 and 20. Items 12 and 15 remain out of consideration, having been abandoned, but on the other hand some of the items admitted by MDC to be breaches of warranty are now in dispute.

Item 2 alleges that the end laps were in positions at which the roof structure is liable to flex upwards and downwards by relatively large amounts, which will eventually lead to the opening up of the joints. On my finding that there should have been no lap joints this allegation is superfluous. If I were wrong on that score I would not find it made out as a separate complaint.

Item 3 complains first that for reasons which I need not describe in detail, since they were substantially accepted by the Defendants' witnesses, the lowest section of each run of roof panels, adjoining the gutter, has its slope reduced or reversed, so as to produce a "dished" profile, leading to ponding of rainwater. This was a design defect for which MDC are liable, and by the same token DLE should not have consented to the drawings and details, but it requires no separate remedial work. The second part of this item concerns the undisputed defect of workmanship consisting of the fact that instead of being turned over properly by the use of a tool adapted to the purpose the lower edges of panels oversailing the gutter have been kicked or "heeled" down roughly and unevenly. As well as being unsightly this contributed to the ponding already referred to. It was plainly a breach of DLE's duty of inspection not to have observed and prevented this practice from the outset of the roof work in June 1989.

Item 5 concerns the upstand joints between adjacent roof panels. The batten caps covering these joints are alleged to have been defectively fixed. I can see no valid design complaint here. As to workmanship I find that the batten caps were forced on instead of being fixed in the way recommended by Ward, and that in the course of the attempted remedial work involving external tape they were cut near the lap joints so that a short section could be taken off during taping and then replaced. The first defect could and should have been observed and stopped by DLE as supervising officer, but there is no evidence that it has caused or contributed to leaks. The second raises the whole question of responsibility for the consequences of the unsuccessful attempt to cure the leaks by taping the laps externally. That was not much canvassed, if at all, in argument. Prima facie liability for it lay with MCL alone, but in Mr Palmer's closing submission the point is made that this work was done before practical completion. If the cuts in the batten caps are said to give rise to any separate head of damage I may need to be further addressed on DLE's liability, but otherwise the question seems academic.

Item 6 raises what seem to be two unconnected complaints. One is that ridge caps were fitted by rivets at 450mm centres instead of screws at 400mm centres. Since it is not alleged that the fixing is inadequate this is in my view at best de minimis. The second allegation, or group of allegations, concerns the flashings at the abutments between roofs and walls or parapets. It is admitted by DLE in their response to the Scott Schedule that these flashings have been incorrectly detailed by Roofdec and that that seems to have caused some leaks. Although Mr Kilbey was cross-examined to suggest the contrary I accept that there were defects and find this part of the item proved against DLE, since I see no reason why they should not have detected them on inspection, indeed Mr Gardiner agreed that when at a late stage he did make a roof inspection he saw flashings which were improperly fitted. I do not find any design fault proved against MDC.

The facts concerned in items 9 and 10 and my findings on the claim for breach of warranty are set out in paragraph 100 above. On the basis of those findings there was no design fault and the only question is whether Mr Gardiner or his assistants (if they had been there, which of course they never were) should have observed that only two screws were being used instead of three. In my view they should. If the process of fixing the clips did not happen to occur during a routine visit (although on the evidence it is quite likely that it would have done), then this was a sufficiently significant part of the process to warrant making arrangements to be there at the right time.

Item 11 arises out of the admitted defect of using only one screw instead of three in certain locations. There was a design allegation of failure to recognise that the purlins were too narrow to receive three screws, but that fails because there was a drawing providing for the insertion of landing plates for that very purpose, which Roofdec nevertheless failed to implement. That fact makes the inspection case against DLE a fortiori that under items 9 and 10, since there could have been no possible excuse for failing to notice the absence of landing plates.

Item 13 concerns another admitted group of defects, concerning the fixing of the roof lights, which led to water penetration into their interior. The design complaints against the use of PVC as a material were not supported by the Plaintiff's experts with any enthusiasm, if at all, and as I understand it were not maintained in Mr Palmer's closing submission. I reject them. The question is therefore again whether DLE were at fault in failing to observe the defective workmanship. There was some discrepancy between the Plaintiff's case as pleaded and the criticisms relied upon in Mr Kilbey's report and evidence, and the highest that Mr Palmer put it in his closing submission was that "these defects should have been looked for once the trail of suspicion had been created by the early existence of leaks". I do not find this allegation made out.

Item 14 is dealt with under breach of warranty in paragraph 101 above and for the reasons already given I need not discuss the part rejected there. The other part concerns the inadequate size of the gutters and their outlets. The only allegation of lack of due care or skill pleaded against either Defendant is of defective design. As to that Mr Kilbey agreed that his criticisms were of the gutters and outlets as built; he had not seen any drawing which specified the dimensions. If there were a pleading against DLE of defective inspection then they would be in the dilemma that they must either have approved a defective design or passed work which departed from a proper design, but as things stand I do not consider that I would be justified in inferring that "as built" was "as designed", which would be the only basis on which I could find against either Defendant.

Item 16 is similarly dealt with in paragraph 102 above. There is no design defect. The origin of the admitted leaking was not established, since I have rejected the orientation of the cladding as a cause. There is no basis, therefore, on which to conclude that it should have been detected on inspection. This item is accordingly not made out against either Defendant at this stage.

Paragraph 103 above makes findings on item 17 as an allegation of breach of warranty. The missing insulation was shown on the drawings, so there is no design defect on the part of MDC. Its absence would in my view have been apparent for a sufficient time before being covered up to be visible on reasonably spaced inspections, and DLE is therefore liable.

Item 18 concerns the admitted defect of dishing at the centre of parapet cappings. There is no pleading of any design fault against MDC. As to inspection by DLE, the alleged cause of the dishing was that supporting straps were absent, or not fixed, or inadequately fixed. The gravamen of the case against DLE, however, was not that that would have been apparent on reasonable inspection, but that the dishing itself would have developed "pretty well straight away", in Mr Kilbey's words. I accept Mr Kilbey's evidence on that point; it is supported by the fact that DLE's own expert witness, Mr Billington, states in his report that the fact that some parapet cappings were dished was reported to MCL before completion of the contract. I therefore find DLE liable.

In paragraph 104 I have found item 19 proved as a breach of warranty. It concerns the absence of movement joints shown on the drawings, so there is no design fault on the part of MDC. It was not seriously suggested that the departure from the drawings would not have been apparent on reasonable inspection; Mr Billington's evidence was directed rather to the proposition that movement joints were not really necessary. It is not, however, the function of an inspecting officer to decide that departure from design requirements can be ignored, nor indeed was there any evidence that that, rather than oversight, was DLE's reason for inaction. They are therefore liable.

Item 20 is dealt with under breach of warranty in paragraph 105 above and as before I say nothing about the part rejected there. The inadequate fixing bolts and lack of weatherproofing show inadequacies of both design and inspection, and I find against both MDC and DLE in that respect. The lack of galvanising, on the other hand, is pleaded as a breach of specification and there is accordingly no design fault. As to inspection DLE says that since the ladders were painted a competent employer's representative might reasonably have failed to detect the lack of galvanising. The Plaintiff's expert witnesses effectively accepted that proposition and this claim against DLE must therefore fail.

Items 21 to 25 were, as already noted, admitted as breaches of warranty, but need consideration here. Item 21 complains of rusting of the steel platform over the Low Bay RLS warehouse roof. The specification required compliance with relevant British Standards and Codes of Practice, and I find no design fault on the part of MDC. As to DLE it is common ground that in the "snagging list" they noted in the section dealing with the roof that "all steelwork requires painting at roof level". They did nothing effective, however, to require that this omission be made good, and are therefore liable. On the other hand the consequences would have been less serious had there been adequate maintenance by the Plaintiff and prompt remedial work when rust first began to appear. I am not aware whether I have the material on which to base a finding as to the effect of that last conclusion on the quantum of damage, and that should be clarified when this judgment is handed down.

Item 22 also concerns rust, this time of the structural steelwork within the perimeter parapet voids. Again there can by the nature of the complaint be no design fault. This steelwork seems to have been painted off site, and there was no evidence that any defect was apparent on delivery; the point mainly relied upon in the Plaintiff's closing address was that damage to the paintwork should have been noted on inspection. There was no evidence, however, that mechanical damage was responsible for most of the rusting, or even for a significant part, and there is in my view no satisfactory basis for a finding against DLE in respect of this item.

It is common ground that item 23 is a very minor matter, and I shall limit my treatment of it to a finding that neither Defendant is liable under the heads now being considered.

Item 24 returns to the topic of leaks. The allegation is that the "soakers", or sheeting laid to protect the perforations in the roof cladding necessary to accommodate extractor fans, roof lights and other services, are not adequately sealed and leak at the side and end laps. In so far as issues of design were raised they seem to have been dropped, and in any event I find no design fault proved. Defects of workmanship are admitted by DLE. Whether they should have been discovered on inspection raises essentially the same issues as in the case of the end laps of the roof cladding itself, as Mr Coulson accepts in his closing submissions, and I reach the same conclusion; they should.

Item 25 concerns the admitted lack of adequate support for rainwater down pipes. MDC's case is that this was a matter left to Roofdec. In view of MDC's design responsibilities, as found by me, that can be no answer and I find MDC liable. Against DLE an allegation of design responsibility was deleted by amendment, and had there been design details, however faulty, which were followed it might have been difficult to find DLE at fault in a purely supervisory capacity. That does not, however appear to have been the case, and DLE themselves noted the inadequacy of the supports on more than one occasion, but did nothing to follow up those observations or require the defect to be made good. I therefore find DLE liable also.

Item 26 is dealt with under breach of warranty in paragraph 106 above and found proved. It concerns the failure to protect the down pipes from damage. On my findings in paragraph 106 as to the contractual provisions there was no design fault. The omission of protection was obvious, and was indeed noted by DLE, but nothing was done to remedy it, and DLE are liable.

By the Statement of Claim against DLE, as finally amended, the Plaintiff alleges that one of DLE's express duties under clause 1(ii) of its contract was to certify practical completion, that on 6 June 1990 it certified practical completion as having been achieved on 14 April 1990, that they were negligent in doing so and that as a result the Plaintiff paid to MCL and lost one half of the the retention under the building contract and also lost the sum of , 715,000 otherwise recoverable under a performance bond. It is alleged in the alternative that the certificate was not a valid certificate of practical completion under the building contract and that DLE were in breach of contract in a number of ways on that basis, again resulting in the loss of a claim on the , 715,000.

DLE's case was that the document of 6 June 1990 was not the certificate required by the contract but a "qualified" certificate of practical completion issued on the Plaintiff's instructions in pursuance of an agreement between the Plaintiff and MCL with the purpose of enabling the Plaintiff to have possession of the premises and of exempting MCL from liability for liquidated damages, notwithstanding that defects remained to be remedied.

I must summarise the relevant provisions of the building contract. By clause 28.1 the employer was on the date for possession to permit the contractor to occupy the site, "but not so as to exclude the Employer", and the contractor was to complete the works on or before the date for completion, subject to the provisions for extension of time. The date for completion was 14 April 1990 and was not extended. The words quoted above are important, since much of the argument for DLE proceeded on the assumption that the contractor, as is common under many standard forms of contract, had exclusive possession during the works and that the employer could accordingly not make any use of the premises until certification of practical completion or the operation of provisions for partial possession. In my judgment that assumption was false; the Plaintiff here did not need a certificate for that purpose, and indeed it is common ground that it was already using parts of the premises before 14 April 1990. That use did not, as I understand it, involve excluding MCL; there was provision in clause 21 for a formal process of taking partial possession which would have had that effect and would have carried other consequences, including the release of a proportionate part of the retention and a "deemed practical completion" and the beginning of the maintenance period in respect of the relevant part of the premises, but clause 21 was not implemented.

By clause 20.1:

What are, however, non-standard are the provisions of clause 20.2. The usual provision is that any defects which appear during a specified defects liability period starting at the date of practical completion are to be specified by the responsible officer and made good by the contractor. The implication is that defects already apparent, unless very minor, are inconsistent with the achievement of practical completion. Clause 20.2, however, is prefaced by a statement of intent that the the works shall "at or as soon as practicable after the expiration of the Period of Maintenance" be delivered up in the condition required by the contract, requires the contractor to "finish the work (if any) outstanding at the date of Practical Completion as soon as may be practicable after such date", and provides for the making good of defects in terms which carry no temporal implication as to the date of their appearance. I conclude that whatever the precise meaning of "practically complete" in clause 20.1 (and the expression is not defined) it is not inconsistent with some degree of outstanding or defective work.

Clause 30 provides for the payment of liquidated damages by the contractor in the case of failure to complete by the date for completion or within any extended time fixed under the provisions in that behalf. Clause 36 provides for certificates and payments. By clause 36.5(a) the employer is to retain a percentage of 3% of the value of work certified for interim payments before the certificate of practical completion and by clause 36.6(b) the supervising officer is to release one half of that retention on issuing the certificate of practical completion.

Clause 43 requires the contractor within seven days to deliver to the employer a performance bond, executed by an approved bank or insurance, for 10% of the GMP and substantially in the form attached. It provides that the bond "shall, if not called, expire and be returned to the Contractor on the date falling one month after the issue of the Final Certificate of Practical Completion". The form of bond attached provides that the obligation under it, which is in the sum of , 715,000, is to be null and void (inter alia) "if .... (c) The Supervising Officer named in the Contract shall pursuant to the provisions of Clause 20 thereof issue a Final Certificate of Practical Completion then upon the date stated therein".

By the middle of March 1990 it was apparent that the works would not reach practical completion by 14 April. I need not refer in detail to the host of items of incomplete or defective work; it is sufficient to mention two decisive considerations. The first is that leakages from the roof were already a major problem, with no assured remedy in sight. The second is that on 15 March 1990 DLE wrote to the Plaintiff with MCL's latest programme revision for the office block, showing a number of trades continuing after 14 April, or not starting until then or later, including suspended ceilings, electrical final fix, decorations and floor coverings. Nevertheless DLE, on 14 March, issued what they described as "our Substantial Completion Certificate .... as required by the contract". It read: "Under the terms of the above mentioned Contract, We certify that Substantial Completion of the Works was achieved on 12 March 1990 ". No satisfactory explanation was given for the production of this thoroughly misleading document. The contract made no provision for a completion certificate otherwise than by clause 20, and the similarity between "substantial" and "practical" had no ascertainable purpose if not to imply to an incautious reader that this was a clause 20 certificate. Mr Gardiner said that he understood that what was required was evidence to be produced to the local authority as freeholder that practical completion would be achieved within a month, that being the condition (he was told) upon which the Plaintiff could begin preparatory work at the premises, rent-free, in advance of commencing operations there. He was unable to explain, however, why in that event he worded the document as he did instead of certifying his expectation or belief that the work would be practically complete within one month, which would have been directed to the ostensible purpose, albeit still untrue.

On 14 April 1990 practical completion had not been achieved, for the reasons summarised at the beginning of the last paragraph, but the Plaintiff carried out a planned move of part of its operations into the premises so far as the progress of the works allowed. The plan was for a phased build-up of the use of the new buildings which would enable Fischers' existing London premises to be vacated in May or June 1990, those at Brierley Hill in about July and those in Manchester by August or September. MCL at the time, DLE's witnesses in evidence, and Mr Coulson in argument, repeatedly referred to what happened on and before 14 April as "taking possession" but in the light of the facts set out above I do not see how the Plaintiff can have made any significant use of the office block for some weeks after that at best and there was no evidence that MCL (which still had a permanent force on site) was excluded from possession of any part of the premises or hindered in what it was doing to complete unfinished work and make good defects. That being so, and having regard to the terms of clause 28.1 of the contract, I see no basis for the suggestion that what happened on 14 April gave MCL any sort of claim to the issue of a certificate of practical completion in flat contradiction of the true situation. I reject the plea of contributory negligence based on the alleged "taking possession", both for the above reasons and because it was for DLE to advise on such matters and there was no evidence that they did.

It is common ground that from 14 April 1990 onwards MCL exerted great pressure on DLE and the Plaintiff for the issue of a certificate of practical completion as of that date. It was put by Mr Coulson to Mr Cook, the Plaintiff's managing director, and accepted by him, and I find, that the reason advanced by MCL for its demand was that the Plaintiff was "in possession". It could not plausibly have been argued, and there was no credible evidence that MCL did argue, that the work was in fact practically complete; even at the trial Mr Waide, the construction manager, had not the hardihood to accept without reservation Mr Coulson's friendly suggestion in cross-examination that that was his view at the time, but instead retreated behind the word "virtually" and reliance on the two certificates. There were of course great financial advantages to MCL in obtaining a certificate, in particular the release of half of the retention, exemption from the liquidated damages for delay for which they would otherwise be liable, and the acceleration of the "period of maintenance" and therefore of the prospect of a Final Certificate and full payment at the end of that period.

On 29 May 1990 there was a project team meeting attended by Mr Cook and Mr Malkin for the Plaintiff, Mr Fitt for MDC, Mr Lovering and Mr Neely for MCL and Mr Gardiner for DLE. Under item 1.20 of the minutes there are 33 heads of "actions", the great majority in the nature of things still to be done to remedy incomplete or defective work. Under item 2.10, "Progress of snagging/outstanding work", it is reported that "70% of original snagging list" was done and "20% of balance of office snagging". The basic office work was clearly still not complete, because the minute continues: "MCL are intending to complete all office work by Friday [1 June]. MCL to inform GF by Thursday night if this will not be done". Item 2.20, "Design Report", relates to other outstanding matters, as do several of the heads under item 3.00, "Matters Arising". There is no reference to the problem of leaks. Item 4.00 reads: "Any Other Business: .1 Practical completion certificate to be issued to MCL, MDC and GF". In the action column against that appear the initials "DLE". There is no reference in the minute to such a certificate's being backdated or qualified.

On 1 June 1990 DLE sent to Fischers copies of the "final snagging sheets" dated 23 May. They comprise some 184 items on eight pages and seem for the most part to involve comparatively minor omissions and defects; they do not include the leaking roof or certain other major subjects of complaint apparent from the letters exchanged shortly afterwards.

The "certificate" itself (using that word without prejudice to the issues as to its status and effect) is, as I have said, pleaded as having been issued on 6 June 1990 but is in fact undated. MCL wrote a letter on 6 June stating that it had been received and receipt of it by Fischers was acknowledged in a letter of 7 June, but the copy in evidence is endorsed: "Original sent to Mr Bullock ccc 8/6/90". Mr Gardiner thought he had signed it on 6 June and I find that he did. It is typed on paper printed simply with DLE's name, followed by "Chartered Quantity Surveyors". The typed heading is "Certificate of Practical Completion". After identifying the parties and the works it reads:

It seems not to be contested, and I find, that enclosed with that certificate was a two-page document on the same type of headed paper entitled "Reserved Matters" and dated 14 April 1990. It read: "The following matters are considered as conditional to Practical Completion", followed by eleven items, of which item 8 was "Completion of snagging as per sheets:-". Sheets 1.00 to 6.00 match the description of the "final snagging sheets" referred to in paragraph 167 above, and 7.00 and 8.00 describe the car park, landscaping and other external works as "still to be snagged". The other ten items require "satisfactory reviews" of a number of matters and "suitable answers" to various attached queries and include a final "sweeping up" item which reads: "Compliance with all design requirements whether statutory or contractual". They do not include any reference to the leaking roof or the other major subjects of complaint already mentioned.

Those matters were the subject of a letter from Mr Cook of Fischers to Mr Gardiner of DLE dated 7 June 1990, which after acknowledging the certificate continued:

On 8 June Mr Gardiner wrote to MCL mentioning the same meeting and taking up and expanding upon the complaints made in the letter of 7 June, referring in particular to "considerable concern regarding leaks from the roof" and the possibility of a claim for "financial recovery of any loss should these leaks continue to occur". As to all the systems the letter said: "we propose that the 12 months defects liability for each item should only start from the date each system is fully operational". MCL's reply on 18 June began: ".... we note you accept that the works were practically complete on the due date of 14th April 1990". The substance of some of the complaints, in particular of that as to the leaking roof, was not challenged, but others were. The response to the proposal for postponement of the defects liability period was that "we have written to all the contractors concerned making this point and we await their response". That bears all the marks of a temporising ploy, and it did not appear in evidence that any such postponement was ever agreed.

It is important to distinguish three separate issues arising out of these events. The first is whether practical completion within the terms of the contract had in fact been achieved by 14 April 1990 or any other relevant date. The second is whether the document issued by DLE on or about 6 June 1990 was the "Final Certificate of Practical Completion" required by clause 20.1 of the contract (whether or not properly issued). The third is whether that document (whether or not such a certificate) was properly issued as between the Plaintiff and DLE, either under the contract or pursuant to some subsequent transaction.

I have already found that practical completion had not been achieved by 14 April. Mr Gardiner agreed in evidence that that was so, and indeed that it had still not been achieved by 29 May or 6 June. I am satisfied that that was correct. Whatever the ambit of clause 20.2 - even if it is wide enough to have accommodated all the items in the snagging sheets - it could not have justified a conclusion that the leaking roof and other matters complained of in DLE's letter of 8 June were consistent with practical completion.

As to the second issue two grounds were relied upon by DLE for the contention that the document was not a certificate under clause 20. The first was the omission of the word "Final". The second was the inclusion of the words "subject to the enclosed Schedule of Defects and Reserved Matters". It may first be observed that any submission that this was not a contractual certificate would lie ill in the mouth of a professional adviser who had chosen to use the words "certify" and "Practical Completion" and to describe it as issued "under the terms of the .... contract", even if the basis for such a submission were in other respects stronger than it is in this case.

As to the specific grounds advanced, the first is wholly lacking in substance. There is provision in clause 36.8 of the contract conditions for a "Final Certificate", tout court, in the conventional sense of a certificate of the final balance due after all defects appearing during the defects liability period have been made good. The inclusion of the word "Final" in clause 20 seems to have been an inconsequential drafting slip; there is no provision for any earlier or different "certificate of practical completion", nor would any such provision have had any apparent meaning or purpose. Its omission from the certificate itself, if by considered choice, would therefore have been entirely sensible and proper. If, as is more likely, it was unconsidered, the draftsman of the certificate simply heading it with the usual title from habit, then a fortiori it cannot have had the intention or effect of invalidating the certificate.

Whether the second ground would have been well-founded, or could at least plausibly have been argued by a certifier, in the teeth of the words quoted in paragraph 177 above had the contract been in a more usual form I need not decide, although I consider it doubtful. In my view it is clear that in the light of clause 20.2, as worded in this contract, no such argument can be sustained. If (as is expressly provided) work may be "outstanding at the date of Practical Completion" and (as is implicit) there may at that date be identified but unremedied defects, then it is not only permissible but may well be prudent to refer to any such matters. It is true that since the certificate is by implication accepting them as being consistent with practical completion as understood in this contract it would strictly have been better to have referred to them not by way of qualification of the certificate but by way of an addendum to it, but that is certainly not a nuance of which the authors of the document can take advantage in order to invalidate their own workmanship. Indeed Mr Gardiner at one point told me that he considered that he had issued the certificate of practical completion provided for by the contract, although he was so confused on the subject that I do not place any particular reliance on that answer in itself. Mr Coulson's final position in his closing oral address seemed to be yet a further variation, namely that the document had the effect of certifying that practical completion would occur at the date at which the scheduled "defects and reserved matters" were remedied, which he said was in September 1990, but that is as implausible as any other of DLE's formulations, if only because it directly contradicts the words "on 14 April 1990" in the document itself.

I therefore conclude that the document issued by DLE took effect as the "Final Certificate of Practical Completion" required by clause 20.1. I emphasise that that conclusion is separate from and independent of the questions (i) whether practical completion had in fact been achieved, (ii) whether it would have been achieved if the only omissions and defects had been those listed in the "Schedule of Defects and Reserved Matters", and (iii) whether for any other reason DLE were justified as against the Plaintiff in issuing the document.

DLE's pleaded and argued case that only a "qualified" certificate was issued on the Plaintiff's instructions, pursuant to an agreement with MCL, is part of their argument that it was not truly a contractual certificate at all, which I have just rejected. On that basis the third issue identified in paragraph 175 above does not strictly arise. I propose to deal with it, however, in order to cover two contingencies. The first is that I am mistaken in my conclusion on the second issue and the document is not a contractual certificate after all. The second is that DLE may wish to rely on the "due authority" point independently of the rest of their argument, that is even if the document was a contractual certificate. I shall in this section, therefore, refer simply to "the document" where it is necessary to cover both of those two contingencies.

If DLE are to succeed on this issue they must establish two propositions explicitly advanced by them, namely (i) that there was an agreement between the Plaintiff and MCL as to the issue and form of the document and the consequences of issuing it and (ii) that DLE issued the document on instructions given by the Plaintiff in pursuance of that agreement, and the Plaintiff must also fail to make good an allegation implicit in the nature of DLE's functions and duties and the plea of negligence, namely (iii) that DLE failed to advise the Plaintiff with due care and skill as to the desirability or otherwise of entering into any such agreement, as to its terms, as to the consequences of issuing a document in the form agreed and as to any precautions which should be taken against any avoidable untoward consequences.

As to (i) I accept that in consequence of MCL's pressure for a certificate there were discussions between the parties. I also accept that the situation was such that some agreement to depart from the strict requirements of the contract might well have been reached between reasonable and well-advised parties who had addressed their minds to the relevant considerations. Each had incentives to reach such an agreement. On the one hand MCL was clearly not entitled to an unconditional contractual certificate, or even to one allowing for such incomplete or defective items as could be accepted under clause 20.2. It was therefore not entitled to release of any part of the retention, was in peril of liability for full liquidated damages as from 14 April 1990 and could not rely upon final payment until after the conclusion of a "period of maintenance" which would not begin to run until practical completion was actually achieved. On the other hand the Plaintiff wanted the fullest possible freedom to carry on business on the site, to which it had already moved some of its operations. I have held that on the true construction of the contract MCL had no exclusive possession and was not entitled to exclude the Plaintiff; nevertheless the Plaintiff was equally not entitled to exclude MCL and there was clearly the possibility that if the Plaintiff was not careful MCL might be in a position to allege either that it was being excluded from all or part of the premises or was being interrupted, delayed or disturbed in the regular progress of its works. Moreover, although possibly weak in law, there was clearly an argument of some cogency in terms of commercial reality and fairness that the rate of liquidated damages, which had probably been fixed in contemplation of the Plaintiff's being unable to make any effective use of the premises, merited some reduction if that was not so. Some sensible compromise agreement was clearly possible in such circumstances, involving for example the recording of outstanding omissions and defects, the regulation of the parties' respective use of the premises and the adjustment of one or more of the following: (i) the amount of retention to be released, (ii) the date of such release, (iii) the rate of liquidated damages, (iv) the date from which they should be payable, (v) the date of commencement of the period of maintenance.

There is simply no evidence of a concluded agreement of that kind at any time before the issue of the document. It is inconceivable that a matter of such importance would not have been reduced to writing, or at the very least have been recorded at the time by one of the parties to the bargain or by DLE, to whom it would have been vital as justifying what would otherwise have been the commission of a plain dereliction of duty in issuing either a false (contractual) certificate or a document of some different kind, intended to have some different effect. There is no such writing or record. It is DLE's case that the agreement was reached at the meeting of 29 May 1990, but the minutes reveal no trace of any such occurrence. On the contrary the letters of 8 and 18 June show that DLE on the Plaintiff's behalf were trying, and failing, to obtain MCL's agreement on crucial issues. MCL indeed, by 18 June, was taking the rigorous line that the document was not only a (contractual) certificate but amounted to acceptance by DLE that practical completion had truly been achieved on 14 April 1990. As to oral evidence that of Mr Gardiner, on which DLE's case depended, was hopelessly confused and contradictory as to the effect which he understood the certificate was to have, and as to that which he intended and believed it did have. At one point, for example, he said that there was still (at the trial) no clause 20 certificate, so that the period of maintenance had still not commenced, and then again, in answer to me, that what the document said on that point was "not really correct", but at another point he said that the document was valid as a certificate for the purpose of "triggering" that period. As to the other participants, subject to some matters of detail on which there is help from the June correspondence and to which I shall return, the most that I can extract with any confidence from their evidence, given well over seven years after the event and with no help from the minutes, is that MCL was pressing for the certificate in reliance on Fischers' "possession", that Fischers gained the impression from DLE that they could not resist that but wanted some sort of assurance that their rights in respect of outstanding omissions and defects would not be prejudiced, and that it was indeed agreed that a certificate in some unspecified form, attempting to reconcile these viewpoints in some unspecified way, would be issued.

As to the specific points listed in the last sentence of paragraph 183 above the question of the occupation and use of the premises and numbered items (i) to (iv) inclusive are simply ignored in the document itself and the correspondence and for the sake of brevity I am prepared for present purposes to assume in favour of DLE, without deciding, that the parties were agreed that that should be so. As to omissions and defects some are of course referred to in the document and its enclosure, and it does seem likely from that fact and from the oral evidence, hazy as it is, that there was some discussion of that point on 29 May, as I accordingly find there was. Fischers' letter of 7 June and the responses of DLE and MCL, however, show clearly that no agreement was reached as to the contents of a complete list or of how it should be dealt with. Mr Coulson put to Mr Cook, and he accepted, that DLE's letter of 8 June was "all entirely as you anticipated ... what you would expect following the discussions about the Certificate", but MCL plainly did not accept that. Item (v) is perhaps the clearest indication of the absence of a concluded and binding agreement. It was pressed by the Plaintiff's representatives at the meeting on 6 June and is so closely connected with the issue of the list of omissions and defects that it seems likely to have been at least aired at the meeting of 29 May, but plainly there was in DLE's understanding no agreement, as is apparent from their use of the word "propose" in the letter to MCL of 8 June.

Proposition (ii) in paragraph 182 above is that DLE issued the document of 6 June 1990 on instructions given by the Plaintiff in pursuance of an agreement between the Plaintiff and MCL. It therefore arises only on the hypothesis that I am wrong in deciding that there was no such agreement. On that basis it must clearly be rejected, if only because the "Schedule of Defects and Reserved Matters" omitted the leaking roof and other admitted omissions and defects and because the "certificate" expressly provided for the period of maintenance to end on 14 April 1991, contrary to what was put to Mr Cook on behalf of DLE and accepted by him as having been agreed. It is not possible even to present the remainder of DLE's case on this point coherently, let alone uphold it, because of the contradictions and confusions in Mr Gardiner's evidence as to precisely what the agreement was. That would also suffice to dispose of DLE's pleaded case of contributory negligence by the Plaintiff in instructing DLE to issue a certificate of practical completion, if it were still pursued, but as I understand Mr Coulson's closing submission it is not.

Proposition (iii) is not therefore necessary to the Plaintiff's success on the issue of DLE's authority to issue the document or the propriety of its doing so, but I shall nevertheless deal with it for completeness and also because of its relevance to the issues as to the performance bond considered next. The undisputed obligations to certify practical completion and to report generally to the Plaintiff plainly included duties to use due care and skill in relation to those functions. In my judgment DLE were clearly in breach of those duties and negligent. Some of the grounds for that conclusion appear from the facts which I have already considered. The Plaintiff should have been advised, but was not, as to the merits and demerits of MCL's demands and the strengths and weaknesses of its own bargaining position along the lines of the discussion in paragraph 183 above. If any agreement emerged DLE should have made a careful record of its terms and obtained confirmation by both parties of its accuracy. If, as I have held, the document of 6 June 1990 was a clause 20 certificate DLE were manifestly in breach of their duties as supervising officer in issuing it in view of the undisputed fact that practical completion had not been achieved. If it was not then DLE were equally in breach of duty in not making that plain to the parties and on the face of the document.

If the document was a contractual certificate DLE were also in breach of duty and negligent in a respect unconnected with the relationships between the Plaintiff and MCL with which paragraphs 166 to 186 above are chiefly concerned. The issue of a certificate of practical completion terminated the Plaintif's rights under the performance bond - immediately upon issue, according to the bond itself, or one month after the date certified, according to clause 43 of the contract conditions, but the discrepancy is immaterial for present purposes. Mr Gardiner, by his own admission, gave no thought to that consequence and no advice to the Plaintiff about it. Of course, as he said, a supervising officer should not refrain from issuing a certificate for that reason if practical completion has truly been achieved (although that would even in those circumstances be no excuse for failing to advise) but, again by his own admission, it had not.

There remains the alternative case pleaded by the Plaintiff in relation to the performance bond if, contrary to my finding, the document of 6 June 1990 was not a contractual certificate of practical completion, and therefore did not of itself terminate the Plaintiff's rights under the bond. That turns on what happened to the bond itself, for it is common ground that those rights ceased, in any event, on its physical surrender. On 6 June 1990 MCL, which was clearly very much alive to the significance of the matters discussed in paragraph 185 above, wrote to the Plaintiff requesting the return of the bond "as we have now received the Certificate of Practical Completion (copy attached), which in turn is the release date for the bond". On 10 July 1990 Mr Cook replied stating that a large number of the items in the schedule of defects enclosed with the certificate, and a number of those outlined in the letter of 8 June 1990, had not been resolved and that there was still an unsatisfactory situation concerning the roof, which stil leaked. He concluded: "As soon as all of these are resolved we will be happy to release the Performance Bond".

On 1 November 1990 MCL again wrote requesting the return of the bond, this time on the ground that "we have completed the works in accordance with the terms of the bond". The letter was dealt with by Mr MacFarlane, the Plaintiff's finance director, who looked out the bond and saw what it said about the effect of a certificate of practical completion. It is common ground that he then rang Mr Gardiner. Neither of them could be expected to have, or had, a detailed recollection of that conversation, but both were agreed, and I accept, that Mr MacFarlane asked whether there was a certificate of practical completion and Mr Gardiner said that there was. Since the only basis on which this point arises is the hypothesis that there was not, that reply was on that basis plainly wrong and negligent. Following that conversation Mr MacFarlane wrote to Mr Gardiner on 8 November enclosing the bond and asking him to forward it to MCL "as it has been effectively cancelled by the issue of the Certificate of Practical Completion". Even if Mr Gardiner had been in any doubt as to why Mr MacFarlane had asked the question about the certificate over the telephone, which I do not believe he was, no such doubt could remain after he read that letter. Moreover he now knew that no harm had yet been done; he could still retrieve the situation by informing the Plaintiff that in his professional opinion the document of 6 June 1990 was not a contractual certificate of practical completion so as to affect the validity or duration of the bond, and that the bond should therefore be retained and enforced. Instead he returned the bond to MCL on 12 November 1990. On the basis on which this point arises that was equally plainly a breach of duty and negligent.

There is a plea by DLE of contributory negligence by failure to call on the bond. It is inherent in my findings in paragraphs 188 to 190 above that that allegation must fail.

There is also a plea of contributory negligence in failing to withhold payment to MCL of more than , 225,896. I do not understand this allegation. This was a contract under which the Plaintiff was obliged to pay the sums certified by the Supervising Officer, and the Supervising Officer was DLE. It was not pleaded that the Plaintiff paid more than was certified.

It is common ground that the only satisfactory cure for the leaking roofs is either to replace them all completely or to add, outside them, a new waterproof element. By the commencement of the trial it was further agreed that replacement was to be excluded as uneconomic and unnecessarily disruptive of the operation of the Plaintiff's business, given the availability of a technically acceptable alternative. The only remaining dispute is between two such alternative schemes. One, favoured by the Plaintiff's experts, would take the form of an "overroof" of PVC reinforced sheets, physically distinct from the existing roof panels and separated from them by insulating panels. This scheme was for brevity referred to by the name of the manufacturers of the sheeting, Sarnafil. The second, supported by the Defendants' experts, would consist of a waterproof membrane applied in liquid form directly to the upper surface of the existing roof panels, with suitable special treatment by reinforcement and otherwise of edges, roof openings, penetrations and other discontinuities in the roof surface. This scheme was referred to by the trade name of the membrane, Soladex. It would cost less than half the Sarnafil scheme, a difference of over , 800,000.

Each scheme was criticised by the proponents of the other. Neither had been designed in full detail, so acceptance of either is to some extent dependent, first, on a judgment as to the ability of a designer, with the assistance of the specialist knowledge of the relevant manufacturer and a contractor experienced in using the system, to devise suitable detailed treatment of all the potential trouble-spots and, secondly, on an assessment of the guarantees and bonds offered by the manufacturer and contractor. Since Soladex would be so much the cheaper and cannot be said to be the more detrimental to the appearance of the buildings - I should have thought, if anything, the reverse - it must clearly be preferred unless the criticisms of its expected effectiveness are, taking the above considerations into account, made good on the balance of probabilities. In my judgment they are not. I have considered them all in detail but do not consider it necessary to add to the length of an already substantial judgment by rehearsing them at length.

I therefore conclude that damages should be assessed on the basis that the appropriate remedial scheme is the Soladex scheme.

On 29 January 1993, during high winds, a section of rooflight blew off the high-bay steel tube warehouse roof. That led to an investigation by Mr McEwan and the discovery by him of the deficient fixings of the roof system to the purlins which are the subject of items 9 to 11 of the Scott Schedule, dealt with in paragraphs 100, 144 and 145 above. There was resultant concern as to the inadequacy of those fixings to resist wind uplift forces. The most pressing aspect of that concern was the danger of personal injuries or loss of life from roof or rooflight sections blown clear of the buildings, including the risk that they might cause serious traffic accidents if blown across the adjacent motorway. The Plaintiff therefore decided to carry out remedial work of a protective nature. Mr McEwan was engaged to advise upon a suitable scheme as a matter of urgency. After considerable research and investigation he advised the installation of containment netting as the fastest and most economical remedy, and that was done. The purpose was not to prevent roof panels or other features from becoming detached from their fixings but to ensure that any which did were contained and not blown away. A large section of rooflight did in fact blow out from the high-bay warehouse roof on 16 February 1995, but was safely contained by the netting.

I have made my findings on liability. As to quantum various points arise. In the first place Mr McEwan agreed that he did not consider netting to be necessary over the whole of the low roofs (where there were two screws per clip) but only over the perimeter areas of those roofs to a depth of 7.5 metres. If the netting actually installed extends further over the low roofs (which is not clear to me from the closing submissions) the cost of the excess is not recoverable. Secondly there was a suggestion by DLE that the work was in truth done against Mr McEwan's advice, in that he would have preferred to go straight to a permanent solution rather than to bother with temporary preventive measures. No doubt as an engineer he would, but the question is whether the Plaintiff acted reasonably in deferring a decision on a permanent scheme until the outcome of litigation and in mitigating meanwhile its loss or prospective loss in the form of potentially enormous liabilities for damages for death or personal injury. In my view it did. I am not clear whether DLE are pursuing as against this item as well as against the Plaintiff's proposed permanent treatment of the fixing deficiency the greater merits of Mr Billington's solution, which involves holding down the clips by the use of wire ties round the purlins, but if so I reject it for two reasons. In the first place the test of expense incurred in mitigation is not what is objectively and by hindsight the best scheme but whether the Plaintiff acted reasonably, and in my judgment it did, not only (as I have already found) in carrying out temporary works at all but also in following Mr McEwan's advice as to their form. Mr McEwan had in fact considered and rejected a "tying" solution in the course of his investigations. Secondly I do not in any event consider Mr Billington's proposals a preferable solution.

The next quantum issue of principle concerns the Plaintiff's claim for loss of value, on the basis that even after completion of remedial work the property will not be worth as much as it would have been had there not been the defects for which the Defendants are liable. DLE resist this claim on the ground that cost of remedial works and diminution in value are in law alternative measures of damage and cannot both be recovered. MDC resists it on the factual ground that the evidence of Mr Castle, the Defendants' valuation expert, that there will be no such diminution if remedial work is properly carried out is to be preferred to that of Mr Lloyd, the Plaintiff's valuer, to the contrary. I have no doubt that the objection in law must be rejected. In point of principle a plaintiff who carries out the best and most economical repair which can be devised to defective property but is left at the end with an asset for which purchasers in the market are not prepared to pay as much as for one which never had the defects has plainly lost both the money expended on the repair work and the residual difference in value. In point of authority none strictly binding on me was cited, but in Payton v Brooks [1974] RTR 169 the Court of Appeal after full argument and citations expressed itself firmly in the sense of the principle stated above; that conclusion was technically obiter, but only because upon a remission to the trial judge for further findings of fact the judge accepted evidence that in the circumstances there was no residual diminution in value. My immediate predecessor in this court, His Honour Judge John Davies QC, awarded such damages in Thomas & ors v T.A.Phillips Builders) Ltd & anor (1985) 3-CLD-08-06 and in Bigg v Howard Son and Gooch (1990) 32 Con LR 39, albeit for a different purpose, I took into account and gave effect to evidence of the reduction in value attributable to the "blight", as it was there called, or "stigma", to use the terminology of the valuers in the present action, attaching to previously defective residential properties, even after repair.

The evidential dispute is not so easy to resolve, since neither valuer had personal experience of either the maintenance or the reduction of price upon discovery of a "stigma" while acting in the sale or purchase of commercial property in such circumstances. Some help can, however, be derived from analogous or related situations. Mr Lloyd did have experience of a stigma discount attaching to previously contaminated land, even after cleansing. The case of repaired residential property is sufficiently common in this court, I think, for me to take judicial notice of the fact that there is such a phenomenon in that field. Mr Lloyd did give evidence, which I accept, that some funding institutions were not prepared to entertain a "purchase and leaseback" of commercial property with such a stigma. I find that there is an effect of this kind, but I consider that it is rather smaller than would be represented by Mr Lloyd's figure of , 200,000, which amounts to nearly 3% of the value free of defects. Mr Castle had, in my view, a legitimate point that the specialist nature of the buildings would narrow the field both of potential buyers and of competing sellers and thus reduce (he, of course, sought to say eliminate) the differential. I assess the diminution in value at , 100,000.

I am not clear from the closing submissions whether there is a further valuation issue arising out of the alleged need for increased future maintenance if the Soladex scheme is adopted. My provisional view is that there should not be; any effect of that kind should be part of the exercise of arriving at the true cost of the Soladex scheme itself rather than a valuation item.

There remain some detailed quantification issues between the quantity surveyor experts, who have reached wide, but not complete, agreement on figures. As between the competing sums on page 10 of trial bundle D3 my findings, which I propose to give summarily, are as follows:

The Defendants allege that the Plaintiff must give credit against its total losses for various sums connected with its claims against or accounts with other parties. As to the money withheld from MCL I ruled on day 18 of the trial on some aspects of the issues raised and in consequence there will need to be a further hearing by way of inquiry on some points. It would, I think, be convenient if the parties could arrange to obtain and make available a separate transcript of that ruling for reference when giving directions for and conducting that inquiry.

As to money which might have been claimed under the performance bond, since it was not actually received I have some difficulty in understanding how it raises an issue of credit, as distinct from causation, contributory negligence, quantification or contribution. It cannot in any event, on my findings as to the certificate of practical completion, be of any assistance to DLE.

There was, however, one issue properly of this kind which was not disposed of by my ruling but which it was agreed need not be deferred to the inquiry, since no further evidence was said to be required. I therefore address it now. It arises out of the agreement with Severfield Reeve referred to in paragraph 11 above, about which I must therefore now give such further particulars as the parties before me were able to furnish. I am bound to say that despite the parties' confidence that all the facts were available most of what follows was extracted by me with some difficulty either during the last of the supplementary oral closing submissions or by research in the documents since, and as will appear I have still in one respect had to proceed on assumption rather than by way of knowledge derived from any specific source.

In the 1992 action against MCL two of the Plaintiff's pleaded allegations of breach of contract, constituting what was called "the roof claim", were as follows:

Severfield Reeve was not a defendant and there was never a lis between it and the Plaintiff. Nevertheless on 14 September 1994 they two and the Plaintiff's Swiss parent company entered into the agreement already referred to. After reciting the contract between the Plaintiff and MCL, the sub-contract between MCL and Severfield Reeve and the proceedings the agreement provides in clause 1 for the payment of , 175,000 by Severfield Reeve to the Plaintiff "in full and final settlement of all or any claims made or capable of being made against Severfield in respect of .... services by Severfield and supply of materials relating to [the Coventry premises]". Clause 5 reads as follows:

That [the Plaintiff] and Georg Fischer AG fully indemnify and hold harmless and keep indemnified and held harmless Severfield against any and all claims (whether made to date or not), losses, demands, liabilities, penalties, interest charges, proceedings, damages, costs and/or expenses sustained of whatsoever nature, incurred or payable by Severfield to [MCL], Multi Limited, Roofdec Limited, their assigns and/or successors, and any other third party .... so far as such relates to or concerns or is in any way connected with services by Severfield relating to the Premises and supply of materials thereto.

In my view there are insuperable difficulties facing any argument that the sum paid by Severfield Reeve under that agreement, or any quantifiable part of it, represented compensation to Plaintiff for the same loss as that for which they are entitled to compensation by way of damages from the Defendants, or that it is open to the Defendants to allege that having settled their third party claims against Severfield Reeve in the present actions they are entitled to any credit against that sum. The payment was not made in settlement of any claim actually made by the Plaintiff against Severfield Reeve, for there was none; the consideration for the payment was substantially the indemnity against claims by others given by clause 5, which exposed the Plaintiff to an open-ended liability which might well have exceeded the consideration received, and by no means necessarily because of the merits of any such claims - a very real possibility was that the Plaintiff might successfully have defended the third party claim in Severfield Reeve's name at considerable expense and then found itself unable to enforce judgment for costs against an insolvent MCL. Moreover the potential claims against which Severfield Reeve was indemnified were not confined to those concerned with the roof or with any other of the matters the subject of the present proceedings. There was no evidence how much money the Plaintiff has actually expended in fulfilment of its indemnity, but there must have been some.

What is perhaps even more germane is that each Defendant joined Severfield Reeve as a third party to the action in which it was sued by the Plaintiff. In the case of DLE that was done by Third Party Notice dated 29 September 1995 and on 7 December 1995, in answer to Interrogatories, DLE were supplied with a copy of the agreement of 14 November 1994. It follows that from that date DLE knew that the indemnity under the agreement covered its claim against Severfield Reeve. I take it, although I was not referred to any document and Mr Mauleverer made no explicit admission, that MDC did not remain long in ignorance after that, if indeed it did not know already; certainly it knew, at latest when Severfield Reeve's Defence was served on 5 July 1996, that Mr Palmer had settled it on instructions from the Plaintiff's solicitors. Had either Defendant pursued its third party claim against Severfield Reeve successfully, therefore, to judgment or settlement, it would in reality have recovered the fruits of that success from the Plaintiff, and knew that that would be so. It plainly could not have kept those fruits and also claimed credit for the , 175,000 or any part of it. I do not see how it can be better off by settling, as each has, on terms by which it discontinues the third party claim and thus effectively concedes that Severfield Reeve was under no liability for any of the losses in respect of which the Plaintiff sues, or at best that any potential liability or risk of liability is reflected in the amount of any unrecovered costs of the conduct of Severfield Reeve's defence, since that amount has in effect already been charged against the , 175,000, the Plaintiff being liable for it under its indemnity, as it would have been for any damages recovered from Severfield Reeve.

Some general arguments were included in Mr Mauleverer's and Mr Coulson's closing submissions on the subject of contribution, but that was necessarily done without knowledge of the conclusions which I would reach as to the liability or otherwise of each Defendant and as to the basis for and extent of any such liability. I have come to the conclusion that it will be much more satisfactory if before I reach any decision on this issue the subject is addressed afresh at the handing down of this judgment or on some suitable later appointment, in the light of my conclusions so far.

February 1998 The above reasoned judgment was handed down. It was ordered that formal judgment be entered for the Plaintiff against the Defendant in each action for damages to be assessed, so far as not already quantified, in accordance with the above reasoned judgment. Directions were given to enable outstanding issues and applications to be heard on 4 and 5 March 1998.