BAILII is celebrating 24 years of free online access to the law! Would you consider making a contribution?
No donation is too small. If every visitor before 31 December gives just £1, it will have a significant impact on BAILII's ability to continue providing free access to the law.
Thank you very much for your support!
[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback] | ||
United Kingdom House of Lords Decisions |
||
You are here: BAILII >> Databases >> United Kingdom House of Lords Decisions >> Beaufort Developments (NI) Ltd v. Gilbert-Ash NI Ltd and Others [1998] UKHL 19; [1999] AC 266; [1998] 2 All ER 778; [1998] 2 WLR 860 (20th May, 1998) URL: http://www.bailii.org/uk/cases/UKHL/1998/19.html Cite as: [1998] UKHL 19, [1998] 2 All ER 778, [1999] 1 AC 266, [1998] CLC 830, [1999] AC 266, [1998] 2 WLR 860 |
[New search] [Buy ICLR report: [1999] 1 AC 266] [Buy ICLR report: [1998] 2 WLR 860] [Help]
LORD GOFF OF CHIEVELEY
My Lords,
I have had the advantage of reading in draft the speech prepared by my noble and learned friend, Lord Hoffmann. I find myself to be in complete agreement with his reasoning and his conclusion; and I too am satisfied that, with all respect to the distinguished members of the Court of Appeal who decided the case, Northern Regional Health Authority v. Derek Crouch Construction Co. Ltd. [1984] Q.B. 644 was wrongly decided and must be overruled. I too would therefore allow the appeal--a conclusion which, I have no doubt, will be welcomed by the courts in Northern Ireland who would, if they had been free to do so, have wished to follow the same course. Like my noble and learned friend, I gladly acknowledge my debt to the writings of Mr. I.N. Duncan Wallace Q.C. on the subject.
LORD LLOYD OF BERWICK
My Lords,
Standard forms of building contract have often been criticised by the courts for being unnecessarily obscure and verbose. But in fairness one should add that it is sometimes the courts themselves who have added to the difficulty by treating building contracts as if they were subject to special rules of their own.
Two recent examples illustrate the point. In Dawnays Ltd. v. F.G. Minter Ltd. and Trollope & Colls Ltd. [1971] 1 W.L.R. 1205 the Court of Appeal held that when a sum is certified by an architect as due under a building contract (in that case the RIBA form) the employer has no right of set-off. The justification for this decision was said to be that cash flow is the life blood of the building trade: see Modern Engineering (Bristol) Ltd. v. Gilbert-Ash (Northern) Ltd. (1973) 71 L.G.R. 162 per Lord Denning at p. 167. The decision came as something of a surprise in the official referees' corridor. It was overruled a few years later when the Modern Engineering case reached the House: [1974] A.C. 689. "It is not to be supposed" Lord Diplock said, at p. 718:
And so the House held, restoring the decision of His Honour Judge Edgar Fay Q.C., that the ordinary common law right of set-off, whereby a breach of warranty may be set up in diminution of the price, applies as much to building contracts as to contracts for the sale of goods.
In the meantime Dawnays' case had been followed in five other cases in the Court of Appeal. This is not surprising when one considers the pressure of litigation in this field. One erroneous decision of the Court of Appeal is bound to lead to others.
The same applies to the second example, although the intervening period has been somewhat longer. The arbitration clause in Northern Regional Health Authority v. Derek Crouch Construction Co. Ltd. [1984] 1 Q.B. 644 gave the arbitrator the power to "open up review and revise any certificate" of the architect, as does the arbitration clause in the present case. The Court of Appeal held this special power was confined to the arbitrator, on whom it had been conferred by the arbitration clause. It could not be exercised by the courts. Since it would have been unjust to the contractors to deprive them of the opportunity of challenging the architect's certificates in that case, the Court of Appeal held that the arbitrations (there were two of them) should go ahead.
As in Dawnays' case, it appears that the decision in the Crouch's case came as a surprise. Official referees had been opening up and revising certificates as a matter of course for many years without any objection from the parties.
It is clear from Pringle J.'s judgment in the present case, that but for the decision in Crouch, he would not have granted the defendant a stay of the plaintiffs' action under section 4 of the Arbitration Act (Northern Ireland) 1937, and the Court of Appeal would have upheld his decision. In my view they would have been right. So the question is whether the Crouch case was correctly decided.
In the present case we are concerned with clauses 30.9, 30.10 and 41.4. Clause 30.9 provides that the final certificate is to be conclusive evidence of the matter certified in accordance with the elaborate provisions set out in that clause. Clause 41.4, the arbitration clause, provides, as one would expect, that the arbitrator's powers to open up and revise certificates are subject to clause 30.9. So the arbitrator has no power to open up and revise the final certificate, save as provided by clause 30.9, and in particular by clause 30.9.3. But we are not here concerned with the final certificate. It has not yet been issued.
Nothing in clause 30.9 affects any certificate other than the final certificate. Indeed clause 30.10 specifically provides:
Interim certificates granted by the architect in the course of a building contract are an important part of the contractual machinery. But there is nothing in the present contract to make interim certificates conclusive; nor was there in the Crouch case. So there is no need for the contract to confer on the courts the power to open up and revise interim certificates. The power already exists, as part of the court's ordinary power to enforce the contract in accordance with its terms.
Then can it be said that the jurisdiction of the courts to open up and revise interim certificates is impliedly excluded by the terms of the arbitration clause? I do not pause to consider whether such an ouster of the court's powers would be effective in law; on any view it would require the clearest of language. I can find no such language in clause 41.4. Since an arbitrator's powers, unlike the powers of the court, are derived ultimately from the contract under which he is appointed, it is by no means unusual to find his powers spelt out in longhand. Thus under the old law (until changed by section 30 of the Arbitration Act 1996) an arbitrator had no power to rule on his own jurisdiction. Since he could not pull himself up by his own boot straps, he could not decide whether a valid arbitration agreement had ever come into existence. But the High Court can rule on its own jurisdiction. Similarly an arbitrator could not rule on a question whether the contract ought to be rectified. So it is not surprising to find the parties conferring on the arbitrator an express power to rectify the contract. But it would be hopeless to argue that because the parties had by clause 41.4 conferred on the arbitrator an express power to rectify the contact, they had by implication curtailed the power of the court to rectify the contract. By the same token, the courts power to open up and revise interim certificates is not excluded by the express power to open up and revise certificates conferred on the arbitrator.
For these reasons, and those given by my noble and learned friend Lord Hoffmann and Lord Hope, with which I agree, I would hold that the Crouch case was wrongly decided, and, like them, would allow the appeal.
LORD NOLAN
My Lords,
I confess to much sympathy with the very distinguished and experienced judges who have expressed or assented to the view that a clause such as clause 41.4 of the building contract giving the arbitrator power to "open up, review and revise any certificate, opinion, decision . . . requirement or notice . . ." confers upon him a discretion wider than that available to a court. The language used is not that of the Supreme Court Practice. It seems to suggest an informal and constructive approach to the resolution of problems occurring in the course of the building work, an approach appropriate to the work of an arbitrator who is chosen because he is an architect rather than a judge.
I am, however, persuaded by the arguments of Mr. Declan Morgan Q.C., and by the opinions of your Lordships whose speeches I have had the opportunity of reading in draft, that the Court of Appeal in Northern Regional Health Authority v. Derek Crouch Construction Co. Ltd. [1984] Q.B. 644 placed a weight on clause 41.4 greater than it will bear. I am persuaded in particular that clause 41.4, read in the context of the contract as a whole, cannot properly be construed as giving an interim certificate (as distinct from a final certificate) any conclusive effect in litigation between the parties. Further, I am satisfied that the clause cannot be regarded as conferring upon the arbitrator the power to modify the contract. I find it difficult to conceive of a contract properly so called which conferred upon a third party the power to modify its terms.
The decision in the Crouch case has stood unchallenged, although not uncriticised, for 14 years. It has now been virtually superseded by section 9(4) of the Arbitration Act 1996, unless and until (if ever) section 86 of that Act is brought into operation. Yet on the view of the law which has prevailed in your Lordships' House the relevant dicta in the Crouch case must clearly be overruled, in justice to the appellants. Pringle J. and the Court of Appeal in Northern Ireland would plainly have refused a stay to the respondents, on the compelling ground that to grant it would lead to duplication of proceedings, had it not been for their reluctant acceptance of what was said in the Crouch case. The same objection to a stay did not, as it happens, arise in the Crouch case itself because all three of the parties concerned submitted to arbitration by the same arbitrator. Mr. Donnell Deeny Q.C., for the first named respondent, persuasively invited your Lordships to assume that the same consequence would follow if the stay were upheld in the present case, but the assumption was not one which Mr. Morgan was prepared to support.
I, too, would therefore allow the appeal.
LORD HOFFMANN
My Lords,
The question before your Lordships is whether an arbitrator appointed to decide a dispute arising under a building contract in the JCT Standard Form has a power to review decisions and certificates of the architect which is not available to a court. The English Court of Appeal so held in Northern Regional Health Authority v. Derek Crouch Construction Co. Ltd. [1984] Q.B. 644 but your Lordships are invited to say that they were wrong.
The clause which is said to give the arbitrator these exceptional powers is 41.4, of which the relevant parts are as follows:
The words particularly relied upon are those which confer a power to "to open up, review and revise any certificate, opinion, decision . . . requirement or notice" and determine matters in dispute as if they had not been given. The Court of Appeal in the Crouch case said that these were special powers conferred exclusively upon the arbitrator. Browne-Wilkinson L.J. said, at p. 667 that in an action "questioning the validity of an architect's certificate or opinion," the jurisdiction of the court would be limited to deciding whether or not the certificate or opinion was invalid for bad faith or excess of power. It could not revise the certificate on the ground that the court thought it was wrong. A clause such as 41.4, on the other hand, gave the arbitrator "power not only to enforce the contractual obligations but to modify them." Sir John Donaldson M.R. also said that the arbitrator could vary the certificates to create new rights and obligations which would not otherwise arise from the contract. Dunn L.J. said that one could not imply a term that if the dispute was litigated instead of arbitrated, the court should have a similar power.
My Lords, I have no doubt that it is open to the parties to enter into an agreement of the kind described by the Court of Appeal in the Crouch case. I put aside the purely theoretical question of whether it is right to speak of the architect or arbitrator having power to modify the contractual obligations of the parties. I find this a strange concept. The powers of the architect or arbitrator, whatever they may be, are conferred by the contract. It seems to me more accurate to say that the parties have agreed that their contractual obligations are to be whatever the architect or arbitrator interprets them to be. In such a case, the opinion of the court or anyone else as to what the contract requires is simply irrelevant. To enforce such an interpretation of the contract would be something different from what the parties had agreed. Provisions of this kind are common in contracts for the sale of property at a valuation or goods which comply with a specified description. The contract may say that the value of the property or the question of whether the goods comply with the description shall be determined by a named person as an expert. In such a case, the agreement is to sell at what the expert considers to be the value or to buy goods which the expert considers to be in accordance with the description. The court's view on these questions is irrelevant.
It is less usual, though certainly theoretically possible, to add a second tier to arrangements of this kind, and to provide that a party who is dissatisfied with the view of one expert shall be entitled to call for the opinion of another, which shall then be final and binding. From the point of view of the court, the final outcome is no different from that in the case of a single expert. The contractual obligations of the parties depend upon the opinion of the one expert or the other and not upon its own view of the matter.
It is this two-tier arrangement which the Court of Appeal in the Crouch case considered that the JCT contract had created; what Sir John Donaldson M.R. afterwards called, cryptically but vividly, an "internal arbitration": see Benstrete Construction Ltd. v. Angus Hill (1987) 38 B.L.R. 115, 118. It is internal in the sense that it does not adjudicate upon the rights and duties of the parties but is part of the machinery for determining what they are. The court appears to have considered that in the absence of a second tier power of the arbitrator to open up, review and revise the architect's certificates, they would (if given in good faith and within the ambit of the relevant contractual provisions) be binding upon the parties. So the critical question is whether, upon the true construction of the contract, such certificates are binding. Unless they are, there is no need for any special second-tier arrangement. They will be open to review by any tribunal called upon the determine the rights of the parties, whether arbitral or judicial.
The judgments of the Court of Appeal contain no very detailed analysis of the provisions of the contract which are said to confer upon the architect this power to issue binding certificates. Although none of the judges say so expressly, there is an implied suggestion that one can infer such a power from the very fact that the arbitrator is given a power to "open up, review and revise." This is the argument from redundancy; the parties are presumed not to say anything unnecessarily and unless the decisions of the architect were binding, there would be no need to confer upon the arbitrator an express power open up, review and revise them. The later judgment of Sir John Donaldson M.R. in Benstrete Construction Ltd. v. Angus Hill (1987) 38 B.L.R. 115, in which he distinguished Crouch on the ground that the contract in the latter case did not have a similar arbitration clause, tends to support the view that he had adopted this form of reasoning.
I think, my Lords, that the argument from redundancy is seldom an entirely secure one. The fact is that even in legal documents (or, some might say, especially in legal documents) people often use superfluous words. Sometimes the draftsmanship is clumsy; more often the cause is a lawyer's desire to be certain that every conceivable point has been covered. One has only to read the covenants in a traditional lease to realise that draftsmen lack inhibition about using too many words. I have no wish to add to the anthology of adverse comments on the drafting of the JCT Standard Form Contract. In the case of a contract which has been periodically re-negotiated, amended and added to over many years, it is unreasonable to expect that there will be no redundancies or loose ends. It is therefore necessary to make a careful examination of the contract as a whole in order to discover whether upon its true construction it does confer binding power upon the decisions of the architect or whether there is some other explanation for the "open up, review and revise" power in clause 41.4. It is also important to have regard to the course of earlier judicial authority and practice on the construction of similar contracts. The evolution of standard forms is often the result of interaction between the draftsmen and the courts and the efforts of the draftsman cannot be properly understood without reference to the meaning which the judges have given to the language used by his predecessors.
The substantive provisions of the agreement state the principal obligations of the parties in clear and objective terms. The contractor is obliged by condition 2.1 to "carry out and complete the works in accordance with the contract documents, using materials and workmanship of the quality and standards therein specified." In the this particular contract, the preliminary articles defined the "Works" as the construction of a nine-storey office block as described in the contract documents. Condition 8.1.3 provides that all work is to be carried out in a proper and workmanlike manner and by condition 23.1.1 the contractor is to proceed "regularly and diligently" with the works and complete them on or before the completion date. The contract specified 14 January 1995 as the completion date and said that the contract price was to be £1,700,000.
This framework of carefully defined contractual obligation is not easily reconcilable with a broad discretion, said to be conferred in the first instance upon the architect and subject to review by an arbitrator, to vary or modify the rights of the parties or to have them conclusively determined by the judgment of one or the other. The parties have agreed that a particular building is to be constructed out of specified materials in a workmanlike manner and that the work should proceed regularly and diligently to completion by a specified date. No doubt within this framework there is room for judgment about what amounts to proper workmanship and diligent progress. But one would not ordinarily describe the exercise of such judgment as a power to modify the contractual rights. These are questions which require the application of objective standards and with which the courts are routinely familiar.
The contract provides for the issue by the architect of certificates or statements in writing as to his opinion on various matters. For present purposes, these documents may be treated as similar. As Devlin J. said in Minster Trust Ltd. v. Traps Tractors Ltd. [1954] 3 All E.R. 137, 142: "The mere use of the word 'certificate' is not decisive." In the absence of express words, the parties are highly unlikely to have intended that some of these statements of opinion should be binding and others not. I shall give a few examples. Clause 30 provides for the issue of interim certificates of the value of the work for which the contractor is from time to time entitled to payment. Clause 30.1.1 provides that the contractor is entitled to payment within 14 days after the issue of the certificate. Clause 25, which deals with extension of time, lists a number of "Relevant Events" such as force majeure or failure to provide instructions or information which the parties accept as capable of delaying completion beyond the completion date without breach of the contractor's primary obligation to proceed diligently with the works. By clause 25.3, if the architect is of opinion that a relevant event is likely to delay completion beyond the completion date, he must give the contractor an appropriate extension of time by a written notice fixing a new completion date. Clause 26 deals with claims for loss and expense caused by deferment of giving possession of the site or various matters such as provision of information for which the employer or architect is responsible. Here again, the architect is required to state his opinion that loss or expense has been caused, or is likely to be caused, by one of the specified matters, whereupon the amount is ascertained by the quantity surveyor and added to the contract price. Finally, clause 30.8 provides for the issue of a final certificate stating the balance due from employer to contractor or vice versa.
Clause 30.9 expressly makes the final certificate conclusive evidence as to various matters. But there is no other express provision which says that any certificate or expression of opinion is to be binding upon the parties in the same way as the determination of an expert. Clause 30.10, immediately after the provisions dealing with the final certificate, says:
This clause has itself been the subject of refined arguments of the inclusio unius, exclusio alterius variety. The clause refers to certificates, therefore it must have been intended that other statements of opinion by the architect should be conclusive. The clause refers only to works, materials and goods being in accordance with the contract, therefore it must have been intended that certificates as to other matters such as extensions of time should be conclusive. In a contract such as this, such arguments are just as dangerous as the argument from redundancy, of which they are in truth merely a variety. If arguments of this kind are to be pursued, what seems to me much more compelling is that the contract contains express and elaborate terms which provide for conclusiveness as to various matters for one certificate and one only, namely the final certificate.
If the certificates are not conclusive, what purpose do they serve ? If one considers the practicalities of the construction of a building or other works, it seems to me that parties could reasonably have intended that they should have what might be called a provisional validity. Construction contracts may involve substantial work and expenditure over a lengthy period. It is important to have machinery by which the rights and duties of the parties at any given moment can be at least provisionally determined with some precision. This machinery is provided by architect's certificates. If they are not challenged as inconsistent with the contractual terms which the parties have agreed, they will determine such matters as when interim payments are due or completion must take place. This is something which the parties need to know. No doubt in most cases there will be no challenge.
On the other hand, to make the certificate conclusive could easily cause injustice. It may have been given when the knowledge of the architect about the state of the work or the effect of external causes was incomplete. Furthermore, the architect is the agent of the employer. He is a professional man but can hardly be called independent. One would not readily assume that the contractor would submit himself to be bound by his decisions, subject only to a challenge on the grounds of bad faith or excess of power. It must be said that there are instances in the 19th century and the early part of this one in which contracts were construed as doing precisely this. There are also contracts which provided that in case of dispute, the architect was to be arbitrator. But the notion of what amounted to a conflict of interest was not then as well understood as it is now. And of course the inclusion of such clauses is a matter for negotiation between the parties or, in a standard form, the two sides of the industry, so that what is acceptable will to some extent depend upon the bargaining strength of one side or the other. At all events, I think that today one should require very clear words before construing a contract as giving an architect such powers.
The language and practical background of the JCT contract does not therefore suggest that any certificates other than the final certificate were intended to have conclusive effect. I return, therefore, to clause 41.4, from which the Court of Appeal in the Crouch case drew the opposite conclusion. It is worth noticing in passing that, in addition to the power to "open up, review and revise," it also confers express powers to rectify the contract and to direct measurements and valuations. It seems plain that the reason for the inclusion of these powers in clause 41.4 is to confer upon the arbitrator the plenitude of power to "determine the rights of the parties" which would be possessed by a court. If the power to "open up, review and revise" was intended to be peculiar to the arbitrator, it would at any rate be different in its purpose from the other powers.
At this stage, however, I wish to refer to an important authority on a clause in similar language which may be taken to have formed part of the background to the inclusion of clause 41.4 in the JCT Standard Form Contract 1980 edition which was used in this case. It is Robins v. Goddard [1905] 1 KB 294. This concerned an R.I.B.A. form of contract, of which clause 17 dealt with defects "arising in the opinion of the architect from materials or workmanship not in accordance with the drawings or specification." It provided that the contractor should make good such defects at his own cost "unless the architect should decide that the contractor ought to be paid for the same." The contract also included an arbitration clause which conferred power to "open up, review and revise" any certificate, opinion etc. of the architect. The contractor sued upon unpaid architect's certificates and the employer counterclaimed on the ground that the work done and materials supplied were not in accordance with the terms of the contract. Farwell J. held that the fact that the architect had not expressed an opinion in accordance with clause 17 that the work and materials were not in accordance with the contract was conclusive and that the court therefore had no jurisdiction to entertain the counterclaim.
Mr. Duke K.C. argued for the contractor that the contract meant that "the certificate of the architect is to be final unless and until it is appealed under the arbitration clause." He mentioned one express exception in the contract but said that "in all other cases the certificates are final so long as the only mode of reviewing them by means of the arbitration clause is not adopted." In other words, he was contending for precisely the two-tier system of conclusive determination which the Court of Appeal adopted in the Crouch case.
The Court of Appeal unanimously rejected this argument. In fact, they stood it on its head. Collins M.R. said that the arbitrator's power to open up review and revise showed that the architect's certificates were not intended to be conclusive at all. And if they were not conclusive, they were no more conclusive in litigation than in arbitration. The power to open up and review, said the Master of the Rolls, at p. 301, "negatives the contention that the defendant is debarred by the certificates of the architect from setting up bad workmanship on the building and the introduction of improper materials." It followed that he could challenge them in an arbitration or, if there was no arbitration, before the court. Stirling L.J. said, at p. 303 that, rather as I have suggested to your Lordships is the case with the J.C.T. contract in this case, the language of the rest of the R.I.B.A. contract did not support the view that certificates were intended to be conclusive. But, he added: "When we come to the arbitration clause the matter is free from doubt." The effect of the power to open up and revise was that:
I have said, my Lords, that Robins v. Goddard [1905] K.B. 294 is an important case. This is not because it lays down any proposition of law but because it tells us what the Court of Appeal, nearly a century ago, when the "open up, review and revise" formula seems to have been relatively new, thought that it was intended to do. Not, as the Court of Appeal said in Crouch, to enable certificates otherwise conclusive to be revised by an arbitrator and no one else, but to make it clear that such certificates were not conclusive at all. The court clearly took the view that the draftsman had seen no need to confer an express power on the court in the same terms as the arbitration clause. The court's jurisdiction was unlimited. It was the arbitrator's powers which need to be spelled out. On this view, the power to open up, review and revise falls into place alongside the other powers conferred by clause 41.4 as a power which a court would in any event possess.
During the 80 years between Robins v. Goddard and Crouch [1984] Q.B. 644, I can find no authority in which a construction inconsistent with the earlier case was adopted. In Neale v. Richardson [1938] 1 All E.R. 753 similar reasoning was used in a case in which the question was whether the contractor could sue without a certificate which the architect had refused to issue. The arbitration clause empowered the arbitrator to decide all disputes, which, on the authority of Brodie v. Corporation of Cardiff [1919] A.C. 337, the court construed to include disputes as to whether or not a certificate should have been issued. There had been no arbitration because the arbitrator (who was also the architect) refused to act, but the court decided in general terms that the non-issue of the certificate was not conclusive and therefore if the arbitrator had power to decide that the money was owing, the court must have it also. Scott L.J. said, at p. 758:
It is true that Robins v. Goddard seems to have been a cause of perplexity to some members of your Lordships' House in East Ham Corporation v. Bernard Sunley & Sons Ltd. [1966] A.C. 407. Lord Upjohn in particular said (at p. 441) that he found it a "rather difficult case" and that while not doubting the actual decision, he did not find it easy to follow some of the observations in the judgments. Lord Pearson also said (at p. 447) that the effect of the case was not clear. I venture to suggest that the problem lay not so much in what Robins v. Goddard decided but the extraordinary proposition for which counsel was seeking to rely upon it in the East Ham case. He appears to have submitted that even if the contract expressly made a certificate conclusive (as did the contract in the East Ham case) but conferred upon an arbitrator a special power to revise it, the court would automatically acquire a similar power if the matter was litigated. In other words, a two-tier structure of conclusive certificates could not be created even by express language. Viscount Dilhorne dealt with the matter accurately and concisely when he said, at p. 424:
P. & M. Kaye Ltd. v. Hosier & Dickinson Ltd. [1972] 1 W.L.R. 146, another decision of this House, also concerned a final certificate expressly declared by the contract to be conclusive. The arbitration clause (clause 35) gave the arbitrator power to decide any dispute including "any matter or thing left by this contract to the discretion of the architect" but there had been no request for arbitration. The House held that a final certificate was conclusive not only in relation to any later litigation but also in relation to litigation already commenced. Your Lordships are not concerned with this aspect of the decision, but Lord Wilberforce said near the end of his speech, at p. 158:
I understand Lord Wilberforce to have meant that the contract in question, which expressly declared final certificates to be conclusive but gave an arbitrator a special power to revise them, had successfully created a two-tier structure of binding certificates. There is nothing to support the view that certificates which are not said to be conclusive or binding can be assumed to have such effect merely because the arbitrator is given an express power to open up and revise them.
In Gilbert-Ash (Northern) Ltd. v. Modern Engineering (Bristol) Ltd. [1974] A.C. 689 the issue was whether a contractor, sued by a sub-contractor on interim certificates, could set off an unliquidated claim for damages for late and defective work. The Court of Appeal had held in a number of decisions that there was a strong presumption (amounting, as Lord Diplock said (at p. 717) virtually to a rule of law) that the contract excluded the right of set off. The House overruled these cases, Lord Diplock saying, at p. 717 that so far from there being a presumption that set-off was excluded:
It was submitted to your Lordships that this presumption supports an argument that the contract should be construed so as to preserve the common law remedies of the parties for breach of contract rather than making their rights subject to the binding decision of the architect. But I think that such an argument may be circular: if the decision of the architect is as conclusive as that of an expert, subject only to second-tier revision by an arbitrator, then the rights of the parties are defined by reference to the opinions of the architect or arbitrator and there is no question of any independent breach of contract. More to the point, however, is a later passage in Lord Diplock's speech, in which he referred to the power of an arbitrator to open up, review and revise any certificate and went on to say, at p. 720:
This passage seems to me a clear and explicit statement that in the case of a certificate expressly stated (by the equivalent of clause 30.10 of the JCT contract) not to be conclusive, the court has exactly the same right to interpret the contractual obligations of the parties as an arbitrator would have had.
This was the state of the authorities at the time when Crouch came before the Court of Appeal. Dunn L.J. introduced his discussion of this question by saying, at p. 663: "Perhaps surprisingly there is no direct authority on the point which is binding on us." He made no reference to Robins v. Goddard [1905] K.B. 294 and appears to have regarded the issue as being, not whether the certificates were conclusive in the first place, but whether (assuming them to be conclusive) one could imply into the contract a term that the court was to have the same power to revise them as the arbitrator. Not surprisingly, he rejected the implication of such a term. Browne-Wilkinson L.J. also made no reference to Robins v. Goddard, agreed that there was no authority directly in point and said, at p. 669 that his view was supported by the "weight of judicial dicta." Sir John Donaldson M.R. did refer to Robins v. Goddard, but only in relation to the comments upon that case in East Ham Corporation v. Bernard Sunley & Sons Ltd. [1966] A.C. 406. He mentioned the comment of Viscount Dilhorne as to the proposition which Robins v. Goddard did not support but made no reference to his summary of what the case actually decided. It was the latter which was, in my opinion, binding upon the Court of Appeal in Crouch and should have been determinative on the question before them. None of the judges made reference to what Lord Diplock had said in Gilbert-Ash (Northern) Ltd. v. Modern Engineering (Bristol) Ltd. [1974] A.C. 689, although it appears from the report (at p. 649) that counsel drew attention to the passage and submitted (in my view rightly) that it supported the proposition that
In my opinion, therefore, the dicta on this point in Northern Regional Health Authority v. Derek Crouch Construction Co. Ltd. [1984] Q.B. 644 were both obiter and wrong. Since then, however, 14 years have passed and the building industry has lived with the Crouch construction of the standard building contracts. There have also been legislative changes. Section 43A of the Supreme Court Act 1981, inserted by section 100 of the Courts and Legal Services Act 1990, provides that if all parties agree, the High Court may exercise any specific powers which the contract confers upon an arbitrator. The discretion of the court to refuse a stay on the grounds of an arbitration clause has been abolished by section 9(4) of the Arbitration Act 1996, which provides that "a stay must be granted unless the arbitration agreement is null and void, inoperative, or incapable of being performed." This provision is subject to section 86, which excludes its operation in domestic operations and retains the court's discretion. Section 86 has not however been brought into force and it is not clear whether it will be. For the moment, the mandatory stay required by section 9(4) appears to be of general application. So the possibility of litigating contracts containing an arbitration clause except by the consent of all parties has been much reduced.
Nevertheless, it seems to me that cases since Crouch show that the decision has caused such uncertainty and even injustice that its dicta should be disapproved. I refer in particular to the recent decision of the Court of Appeal in Balfour Beatty Civil Engineering Ltd. v. Docklands Light Railway Ltd. (1996) 78 B.L.R. 42. It was a claim for extension of time and loss and expense under the ICE Conditions of Contract, which had been amended, first, by substituting the employer's representative for the engineer and secondly, by deleting the arbitration clause. The contract provided for the employer to certify extensions of time and loss and expense claims. But there was no provision that they were to be binding or conclusive. Nevertheless, the court held that there was no power to "open up, review or revise" them such as an arbitrator might have had if there was an arbitration clause in the usual form and that, as a matter of construction, "the Contractor's entitlement was to depend upon the Employer's judgment.": per Sir Thomas Bingham M.R. at p. 57. Your Lordships will remember that in Benstrete Construction Ltd. v. Angus Hill (1987) 38 B.L.R. 115, 118 Sir John Donaldson M.R. appeared to be saying that the Crouch construction of the certification clauses as conclusive in litigation was based upon the fact that the contract created an "internal arbitration." But in the Balfour Beatty case the contractors were held, even in the absence of an arbitration clause or any express language as to the certificates being conclusive, to have subjected themselves to the judgment of the employer. It is true, as Sir Thomas Bingham M.R. remarked, at p. 57, "It is not for the court to decide whether the contractor made a good bargain or a bad one; it can only give fair effect to what the parties agreed." On the other hand, in deciding exactly what the parties did agree, it seems to me that in the absence of express language, one should not assume so uncommercial a bargain. I do not think that the Balfour Beatty case would have been decided as it was if not for the shadow of the Crouch decision that certificates, even if not declared to be conclusive, can be questioned only for bad faith or excess of power. I do not think that anyone in the industry can be said to have acted in reliance on the Crouch case and I would therefore overrule it. I must acknowledge the assistance which I have had in reaching this conclusion from the writings of Mr. I.N. Duncan Wallace Q.C.
The significance of doing so in the present case can be briefly stated. Beaufort Developments Ltd ("the employer") entered into a contract dated 3 May 1994 with Gilbert-Ash N.I. Ltd. ("the contractor") for the construction of a nine-storey office block in Belfast. The contract was in the standard JCT form, 1980 edition Private Without Quantities. By a separate contract it employed the firm of Parker & Scott ("the architects") as architects. The Works were not completed on time; an outcome for which the contractor blamed the architects and the employer blamed them both. For example, some work had to be done over again and there are disputes over whether this was on account of the contractor's bad workmanship or use of wrong materials or the architects' failure to provide adequate drawings and information. There is also a dispute over whether this and other matters actually caused the delay in completion. The contractor claimed that it was entitled to payment from the employer, both under certificates issued by the architects for work done under the contract and by way of payment for extra work. The employer claimed that it was entitled to damages against contractor and architects for breach of contract.
Litigation commenced on 31 August 1995 when the contractor issued a writ claiming about £230,000 and interest due under six architects' certificates. On 9 October 1995 the employer served an unilluminating defence, denying liability and alleging that it was entitled to set off a cross-claim in a larger amount. On 15 November 1995 the contractor served a notice to refer and concur in the appointment of an arbitrator pursuant to the arbitration clause in the contract. It referred in general terms to the areas of dispute such as the responsibility for delay and the contractor's claims to payment for extra work. On 5 December 1995 the employer issued a writ which named both the contractor and the architects as defendants. It claimed damages for negligence and breach of contract. On 7 February 1996 the contractor issued a summons for a stay of the employer's action pursuant to section 4 of the Arbitration Act (N.I.) 1937. The agreement between the employer and the architects also contained an arbitration clause but the architects did not make a similar application.
Master Wilson granted the stay and on appeal his decision was affirmed by Pringle J. He did so with reluctance because he said that the architects could not be required to take part in the arbitration between the employer and the contractor and there was a very real risk of conflicting decisions in the arbitration and the litigation against the architects. But he considered that he was bound by the Crouch case [1984] Q.B. 644 to hold that an arbitrator would have the power to "open up, review and revise" certificates or opinions of the architect which the court did not possess. If a stay was refused, the contractor would therefore be at a "grave disadvantage in that it will be faced with architect's certificates which the court will not be able to review." In the Court of Appeal, Carswell L.C.J. prefaced his judgment by saying:
As in my opinion the Crouch case was wrongly decided, I think that the discretion should have been exercised as Pringle J. would have done if he felt free to do so. I would therefore allow the appeal.
LORD HOPE OF CRAIGHEAD
My Lords,
The application by Gilbert-Ash NI Ltd. ("the contractor") for a stay of the action by Beaufort Developments (NI) Ltd. ("the employer") was made under section 4 of the Arbitration Act (Northern Ireland) 1937. The grant of a stay under that section is discretionary. It is plain from the reasons which Pringle J. gave for affirming the Master's decision to grant the stay that he would have refused the application had it not been for the decision of the Court of Appeal in Northern Regional Health Authority v. Derek Crouch Construction Co. Ltd. [1984] 1 Q.B. 644. As he pointed out, the circumstances in the present case are very similar to those in Taunton-Collins v. Cromie [1964] 1 W.L.R. 633.
In that case, as here, the contract between the employer and the contractor contained an arbitration clause. The architect, in response to the employer's claim against him, put part of the blame for the unsatisfactory building on the contractor. The employer then joined the contractor as a defendant to his action against the architect. The contractor's application for a stay in reliance on the arbitration clause was refused by the official referee, and an appeal against his decision was dismissed. This was because to grant a stay would have resulted in two sets of proceedings. There would have been an arbitration as against the contractor and an action as against the architect. There would have been a substantial risk of different decisions on the same question and on the same facts. Pearson L.J. said, at p. 638 that there were very strong reasons based on the principle of avoiding a multiplicity of proceedings for permitting the action to continue as an action by the employer against both defendants.
Pringle J. noted that there were many issues of fact to be determined in the present case. They included the reasons for the delay which had occurred in the completion of the works by the contractor, the standard of workmanship and materials and the question whether acceleration of the works had given rise to additional costs. He said that he could foresee very considerable difficulties in dealing with these issues, and that in his opinion the risk of conflicting decisions was a very real one. Nevertheless he felt obliged to uphold the stay in view of the consequences to the contractor of the decision in the Crouch case if the dispute between it and the employer were not to be dealt with by an arbitrator. The Court of Appeal decided, with some hesitation, to follow the decision in the Crouch case. But Carswell L.C.J. made it clear in the course of his judgment that the court had considerable reservations about the soundness of that decision. He said that, if the matter were res integra, he would have been attracted to an interpretation of the contract which would have avoided the need to go to arbitration to avoid injustice to the contractor.
The situation in this case has therefore brought out into the open difficulties created by the decision of the Court of Appeal in the Crouch case which have been lying not far below the surface since it was made. The fundamental question is whether the court has been deprived, by the power which the parties have given to their arbitrator to open up, review and revise certificates, opinions and decisions of the architect, of its ordinary power to determine the rights and obligations of the parties and to provide them with the usual remedies. In the Crouch case [1984] Q.B. 644, 667D Browne-Wilkinson L.J. said:
I shall return to this passage later when I come to examine that case in more detail. For the time being it is sufficient to notice that the basis for this view is the difference which was said to exist between the powers of the court and those conferred by the agreement of the parties on the arbitrator. The power of the court, it was said, was to enforce the contract, while the arbitrator had been given the power, which the court does not possess, to modify it. This proposition has come to be applied generally to all cases where the arbitrator has been given power to open up, review and revise certificates.
The contract in the present case was entered into under the JCT Standard Form of Building Contract 1980 Edition (Private without Quantities). It incorporated amendments numbers 1 to 12 together with the Adaptation Schedule for Northern Ireland and the Contractor's Designed Portion Supplement. By article 5 of the Articles of Agreement the parties agreed to refer their disputes to arbitration in accordance with clause 41 of the Conditions. Clause 41.4 of the Conditions, so far as relevant to this case, provides:
Clause 41.5 provides that, subject to clause 41.6 which enables either party to appeal to the High Court on any question of law, the award of such arbitrator shall be final and binding on the parties. Clause 30 deals with certificates and payments to the contractor. Clause 30.9 makes provision as to the effect of the final certificate, while clause 30.10 makes provision as to the effect of certificates other than the final certificate. Clause 30.9 provides that, except in certain circumstances, the final certificate shall be conclusive evidence "in any proceedings arising out of or in connection with the contract (whether by arbitration under article 5 or otherwise)" as to various matters about which decisions have had to be made by the architect under the contract. Clause 30.10 provides:
Had it not been for the weight of contrary authority I would not have found much difficulty in reaching the following conclusions about the effect of these provisions relating to the powers of the arbitrator and the finality to be given to certificates. In the first place, the function of clause 41.4 is to define the powers which are to be given to the arbitrator. An arbitrator has no jurisdiction except that which the parties choose to confer upon him by their agreement to refer their disputes to an arbitrator. The whole question as to the extent of his powers rests upon contract. So it is necessary that the agreement should set out all the powers which he is to have in order that he may determine all the matters which are in dispute. But it is not to be thought that by conferring powers on the arbitrator the parties are limiting the ordinary powers of the court to determine their rights and obligations under the contract. In the present case, for example, clause 41.4 gives power to the arbitrator to rectify the contract. The court already has that power, but it might well have been in doubt as to whether the power of the court could be exercised by the arbitrator. There is nothing in clause 41.4 to suggest that, by conferring this power on the arbitrator, the parties intended to remove this power from the court.
Then there are the provisions about the certificates. In the present case the contractor seeks payment of the sums certified as due for payment under six interim certificates. It appears that it will also seek to maintain a claim against the employer for additional costs which are not the subject of any certificate by the architect. The employer for its part claims, by way of set off against any sums due to the contractor, amounts in respect of delay in completion of the construction and fitting out works and damages for breach of its obligation to provide materials and workmanship to the standard which the contract required. These are matters about which the contract provides for decisions to be taken or opinions to be given by the architect. But there is no express contractual provision to which one can point which has the effect of giving finality to the various decisions and opinions which he has made. We are not concerned in this case with any question as to the conclusive effect of the final certificate because, although a certificate of practical completion was issued in June 1996, the final certificate has not yet been issued. It is made quite clear by clause 30.10 that the interim certificates which the architect has issued are not of themselves to be conclusive evidence.
On this approach, if there is no stay, the court will be able to exercise all its ordinary powers to decide the issues of fact and law which may be brought before it and to give effect to the rights and obligations of the parties in the usual way. It will have all the powers which it needs to determine the extent to which, if at all, either party was in breach of the contract and to determine what sums, if any, are due to be paid by one party to the other whether by way of set off or in addition to those sums which have been certified by the architect. It will not be necessary for it to exercise the powers which the parties have conferred upon the architect in order to provide the machinery for working out their contract. Nor will it be necessary for it to exercise the power which clause 41.4 confers on the arbitrator to revise certificates. This is because the court does not need to make use of the machinery under the contract to provide the parties with the appropriate remedies. The ordinary powers of the court in regard to the examination of the facts and the awarding of sums found due to or by either party are all that is required. There would be no risk of any injustice to the contractor.
In Taunton-Collins v. Cromie [1964] 1 W.L.R. 633, 636 Pearson L.J. said that in that case there was a conflict between two well established principles. One was that parties should normally be held to their contractual agreements. Where the parties have agreed that any dispute or difference between them should be referred to arbitration the court should be willing to say by its decision what the parties have already said by their contract. The other principle was that a multiplicity of proceedings was highly undesirable. In that case it was the principle of avoiding a multiplicity of proceedings which prevailed. The effect of the decision of the Court of Appeal in the Crouch case has been to reverse the result of balancing these two principles. But that case also, it may be said, involved the application of two well established principles. The first is that which was expressed by Browne-Wilkinson L.J. [1984] Q.B. 644, 667B in these terms:
The second, which was not referred to at all in the judgments in that case, is that which was described by Lord Diplock in Modern Engineering (Bristol) Ltd. v. Gilbert-Ash (Northern) Ltd. [1974] A.C. 689, 718E:
The facts in the Crouch case can be stated quite shortly. There had been delays in the completion of the work under the main contract and by a nominated sub-contractor. An arbitrator had been appointed on a reference under the main contract between the contractor and the authority. The same arbitrator had been appointed in a reference under the subcontract in which the nominated sub-contractor was proceeding in the contractor's name in its arbitration with the authority. The arbitrator was empowered under clause 35(3) of the conditions under the main contract to open up, review and revise the architect's certificates, opinions, decisions, requirements or notices. There had been no final certificate. The question was whether the arbitration proceedings should be stayed. One of the issues raised in the case--although it was not necessary to decide this issue in order to dispose of the appeal--was whether the official referee in the High Court had the power to open up, review and revise the certificates, opinions, decisions, requirements or notices of the architect which had been given to the arbitrator by clause 35(3) of the main contract.
Dunn L.J. said, at p. 663F that the court had been told that it was common practice for the official referee to open up and review certificates and other decisions of the architect and he observed that there were decisions of high authority either way as to whether this was competent. In his summary of the competing arguments he said that the authority had relied on obiter dicta of Lord Wilberforce in Hosier & Dickinson Ltd. v. P & M Kaye Ltd. [1972] 1 W.L.R. 146, 158 and of Viscount Dilhorne and Lord Cohen in East Ham Corporation v. Bernard Sunley & Sons Ltd. [1966] A.C. 406, 424 and 432. The other side had contended that in order to give business efficacy to the contract there must be an implied term that if the parties were to litigate rather than arbitrate the court was to have the same powers as the arbitrator. At p. 664D-E he went on:
Browne-Wilkinson L.J. began his discussion of this point at p. 667B with the statement of principle that the court had power only to enforce the contract, not to modify it in any way. He went on to say that, if the parties have agreed on a specified machinery for establishing their obligations, the court cannot substitute a different machinery. He then distinguished the powers of the court from those of the arbitrator. The parties had agreed that certain rights were to be determined by the certificate or opinion of the architect. In no circumstances would the court have power to revise such certificate or opinion solely on the ground that it would have reached a different conclusion, as to do so would be to interfere with that agreement. But the powers conferred on the arbitrator were different. He had been given power to modify the contractual rights by varying the architect's certificates and opinions if he disagreed with them and to substitute his own discretion for that of the architect. He summed the matter up at pp. 667H-668A with these words:
Sir John Donaldson M.R., at pp. 671-673 also drew a distinction between the power of the court and those which had been conferred on the arbitrator. He said that the powers conferred on the arbitrator seemed to him to involve the exercise of a completely novel jurisdiction which was quite different from the function of the court. He described the court's function as being to determine facts and to enforce the contractual rights of the parties. The arbitrator had that function also, but he also had the right and duty which the court did not possess to review the architect's decisions and, if appropriate, to substitute his own. He also found support for his opinion that the court would not be able to exercise the power to open up and review which clause 35 of the JCT contract had given to the arbitrator in the dicta to which Dunn LJ had referred in the East Ham case [1966] A.C. 406 and in what Lord Wilberforce had said, in the Hosier & Dickinson case [1972] 1 W.L.R. 146, 158.
The statement of principle with which Browne-Wilkinson L.J. began his discussion of this point seems to me, with respect, to be both relevant and accurate. I do not think that it can be doubted that in a case which has been based on contract the court's function is to enforce the agreement of the parties, not to modify their agreement in any way. But I have the impression that in the discussion which follows, and in the remarks which were made by the other judges, two other important points were overlooked and that the description of the arbitrator's powers as including a power to modify the contract, for what it is worth, is less than accurate.
The first point is that there is a difference between an agreement that machinery is to be used to implement or to give effect to the contract and an agreement that the parties' rights are to be determined solely by means of that machinery. An agreement which falls into the first category will be needed in almost every building or engineering contract. Some method has to be laid down for dealing with such matters as variations to the contract works and the making of interim payments to the contractor as the work proceeds. But an agreement of that kind does not imply any limitation on the ordinary powers of the court. Nor does it confer any powers on the architect or engineer, or in his turn on the arbitrator, which restrict the power of the court, in the event of litigation, to conduct its own inquiry into the facts. Its purpose is simply to enable the contract to be worked out upon the agreed terms to achieve the result to which it was directed. The purpose of an agreement which falls into the second category, on the other hand, is to exclude the point at issue from being determined by the court. If the parties have agreed that a dispute between them is to be determined conclusively by the architect or engineer, or in the event of dispute by an arbitrator, the sole function of the court is to give effect to the agreement which they have made. Its jurisdiction is to enforce the contract. Its duty is to ensure that the decision of the architect or the engineer or, in his turn, of the arbitrator is given the conclusive effect which has been agreed. But none of the judges in the Crouch case addressed the question whether the certificates or opinions of the architect which the arbitrator had power to open up, review and revise were agreed by the parties to the contract to have effect as conclusive evidence.
The second point is this. The powers which the court ordinarily has to determine and give effect to the rights and obligations of the parties to a contract differ from the additional powers which, in the typical building or engineering contract, are given to the architect or the engineer and, in the event of any dispute about their exercise, to the arbitrator. The purpose of these additional powers is not to deprive the court of its ordinary powers to determine their rights and obligations under the contract. Their purpose is to enable the architect or engineer, and in the event of a dispute about their exercise the arbitrator, to do things in the course of the execution of the contract which the court could not do. This point was well expressed by Piers Ashworth Q.C., sitting as a Deputy High Court judge, in National Coal Board v. William Neill & Son (St. Helens) Ltd. [1985] Q.B. 300, 309C-F, where he said:
So there is this difference between the the provision of an agreed machinery for giving effect to the contract and the taking of decisions or the expressions of opinion which may be necessary from time to time to its exercise. Where the parties have conferred additional powers on the architect, the engineer or the arbitrator, the function of the court is to give effect to the agreement of the parties as to the use of that machinery. The court cannot give the instructions or issue the certificates. But the fact that decisions are taken or opinions are expressed in the course of the working out of that machinery does not, of itself, affect the ordinary powers of the court if litigation becomes necessary. That would only be so if the parties had agreed that that those decisions or opinions were to receive effect as conclusive evidence.
Then there is the question whether it was accurate for the court in the Crouch case to describe the power which was given by the conditions in the JCT contract to the arbitrator to open up, review and revise certificates and opinions of the architect as a power to modify the contract. It was primarily on this ground that the distinction was drawn between the powers of the arbitrator and those of the court. In my opinion the correct analysis is that the power which is given to the arbitrator in the event of a dispute about the exercise of his powers by the architect is a power to give effect to the contract, not to modify it.
In Modern Engineering (Bristol) Ltd. v. Gilbert-Ash (Northern) Ltd. [1974] A.C. 689, 717B Lord Diplock said that a building contract is an entire contract for the sale of goods and work and labour for a lump sum price payable by instalments as the goods are delivered and the work is done. Decisions have to be taken from time to time about such essential matters as the making of variation orders, the expenditure of provisional and prime cost sums and the extension of time for carrying out the works under the contract. Decisions also have to be taken from time to time as to the adjustments which may have to be made to the contract sum on account of these matters and on the amounts to be paid to the contractor by way of instalments towards a final settlement of the sums to which he is entitled under the contract. But in taking their decisions on all these matters the duty of the architect or the arbitrator is to give effect to the contract, not to alter or modify it. Variations can only be made to the contract within the limits which the parties themselves have agreed. From time to time in order to exercise these functions the architect or the arbitrator must apply the provisions of the contract to the facts. But in this regard their position in the resolution of disputes between the parties is no different from that enjoyed in the exercise of its ordinary powers by the court.
For these reasons I consider that the Court of Appeal in Crouch, having started from the correct principle, fell into error in its application to the facts. Unlike both East Ham and Hosier & Dickinson it was not a case in which there had been a final certificate. There was no issue between the parties as to whether any of the certificates which had been issued had been agreed to be conclusive evidence of the facts stated in them. In East Ham it was held that the effect of the relevant provisions of the building contract was that the final certificate was conclusive evidence and that it could not be re- opened even by the arbitrator. In the Hosier & Dickinson case Lord Morris of Borth-y-Gest explained, at p. 153B-C that the fact that the parties had agreed to the conclusiveness of a certificate as a matter of evidence did not involve any ouster of the jurisdiction of the court. Lord Wilberforce said, at p. 157D that to describe it as doing so would be to misdescribe the effect attributed to it by the contract. There was no discussion of any of these points in the Crouch case. But the effect of the decision was to confer a similar status on the certificates and opinions of the architect, subject only to their review by an arbitrator, without having identified any provision in the contract which removed these matters from the ordinary jurisdiction of the court.
In the Crouch case [1984] Q.B. 644 both Dunn L.J. and Sir John Donaldson M.R., at pp. 663H and 673C relied on the dictum of Lord Wilberforce in Hosier & Dickinson Ltd. v. P. & M. Kaye Ltd [1972] 1 W.L.R. 146, 158A, where he said:
I agree that, at first sight, this dictum may be taken to suggest that the court can never open up and review certificates where such wide powers in that regard are given to the arbitrator. But I think that to read the dictum in this way would be to take it out of its context. The context is to be found in what Lord Wilberforce said, at p. 157E-H about the provisions of the contract which dealt with the effect of the final certificate. He said, at p. 157E that the court proceedings had raised the question whether the work done and the materials used were such as should have been done and used under the contract, and that an essential question was what standard was to be set for this. It was in that context that he examined the provisions about the final certificate. He concluded that there could be no objection to a clause which provided that it was the architect's standard which was to be relevant and that his final certificate was to be conclusive evidence. The only circumstances in which, by clause 30(7) of the conditions in that contract, the final certificate was not to be conclusive evidence were where there had been a written request by either party, within certain time limits, to concur in the appointment of an arbitrator.
It seems to me that the discussion in the Hosier & Dickinson case put the matter on the correct basis. On the one hand there is the principle which was expressed by Lord Diplock in Modern Engineering (Bristol) Ltd. v. Gilbert-Ash (Northern) Ltd. [1974] A.C. 689, by which clear unequivocal words must be used to deprive a party to a contract of recourse to the court for the ordinary exercise of its powers and the granting of the ordinary remedies. On the other there is the principle that the court must give effect to the contract which the parties have made for themselves. If the contract provides that the sole means of establishing the facts is the expression of opinion in an architect's certificate, that provision must be given effect to by the court. But in all other respects, where a party comes to the court in the search of an ordinary remedy under the contract or for a remedy in respect of an alleged breach of it, the court is entitled to examine the facts and to form its own opinion upon them in the light of the evidence. The fact that the architect has formed an opinion on the matter will be part of the evidence. But, as it will not be conclusive evidence, the court can disregard his opinion if it does not agree with it.
For these reasons I agree with my noble and learned friend Lord Hoffmann that the Crouch case was wrongly decided and should be overruled. I also consider that the answers which the Court of Appeal gave to the questions which were before it in Balfour Beatty Civil Engineering Ltd. v. Docklands Light Railway Ltd. (1996) 78 BLR 42 were the wrong answers. That was a case in which there was no arbitration clause, but there was no provision in the contract agreeing that the opinion of the employer's representative was to be conclusive evidence in the event of a dispute. The fact that the contract did not provide an agreed means of challenging the judgment of the employer's representative did not affect the power of the court to examine the issue and to form its own judgment in the light of the evidence.
I can return now to the facts of this case. There has been no final certificate. No certificates or opinion have been issued or given which the parties have agreed shall be taken to provide conclusive evidence as to the matters which are in dispute. The court is thus in no different position in regard to such expressions of opinion as have been given as would be an arbitrator. It does not have the additional power which an arbitrator has under this contract to issue fresh certificates in place of those already issued by the architect. But it does not need that power in order to resolve the disputes which have arisen in this case. In these circumstances there would be no injustice to the contractor in refusing a stay. To grant a stay would be to risk conflicting decisions in the separate proceedings which would be needed to determine the respective responsibilities to the Employer of the Contractor and of the architect. I would therefore allow the appeal and refuse a stay of the proceedings in the High Court.