"[W]hat the court has to do is to satisfy itself, ex post facto, that the covenanted work has been done, and it has ample machinery, through certificates, or by inquiry, to do precisely this."

This distinction between orders to carry on activities and to achieve results explains why the courts have in appropriate circumstances ordered specific performance of building contracts and repairing covenants: see Wolverhampton Corporation v. Emmons [1901] 1 Q.B. 515 (building contract) and Jeune v. Queens Cross Properties Ltd. [1974] Ch. 97 (repairing covenant). It by no means follows, however, that even obligations to achieve a result will always be enforced by specific performance. There may be other objections, to some of which I now turn.

     One such objection, which applies to orders to achieve a result and a fortiori to orders to carry on an activity, is imprecision in the terms of the order. If the terms of the court's order, reflecting the terms of the obligation, cannot be precisely drawn, the possibility of wasteful litigation over compliance is increased. So is the oppression caused by the defendant having to do things under threat of proceedings for contempt. The less precise the order, the fewer the signposts to the forensic minefield which he has to traverse. The fact that the terms of a contractual obligation are sufficiently definite to escape being void for uncertainty, or to found a claim for damages, or to permit compliance to be made a condition of relief against forfeiture, does not necessarily mean that they will be sufficiently precise to be capable of being specifically performed. So in Wolverhampton Corporation v. Emmons [1901] 1 Q.B. 515, Romer L.J. said that the first condition for specific enforcement of a building contract was that "the particulars of the work are so far definitely ascertained that the court can sufficiently see what is the exact nature of the work of which it is asked to order the performance". Similarly in Redland Bricks Ltd. v. Morris [1970] A.C. 652, 666 Lord Upjohn stated the following general principle for the grant of mandatory injunctions to carry out building works:

 "[T]he court must be careful to see that the defendant knows exactly in fact what he has to do and this means not as a matter of law but as a matter of fact, so that in carrying out an order he can give his contractors the proper instructions."

Precision is of course a question of degree and the courts have shown themselves willing to cope with a certain degree of imprecision in cases of orders requiring the achievement of a result in which the plaintiff's merits appeared strong; like all the reasons which I have been discussing, it is, taken alone, merely a discretionary matter to be taken into account: see Spry on Equitable Remedies (4th ed.) at p. 112. It is, however, a very important one.

     I should at this point draw attention to what seems to me to have been a misreading of certain remarks of Lord Wilberforce in Shiloh Spinners Ltd. v. Harding. [1973] A.C. 691, 724. He pointed out, as I have said, that to grant relief against forfeiture subject to compliance with a repairing covenant involves the court in no more than the possibility of a retrospective assessment of whether the covenanted work has been done. For this reason, he said:

 "Where it is necessary, and, in my opinion, right, to move away from some 19th century authorities, is to reject as a reason against granting relief, the impossibility for the courts to supervise the doing of work."

This is plainly a remark about cases involving the achievement of a result, such as doing repairs, and, within that class, about making compliance a condition of relief against forfeiture. But in Tito v. Waddell (No. 2) [1977] Ch. 106, 322 Sir Robert Megarry V.-C. took it to be a generalisation about specific performance and, in particular, a rejection of difficulty of supervision as an objection, even in cases of orders to carry on an activity. The Vice-Chancellor regarded it as an adoption of his own views (based, as I have said, on incomplete analysis of what was meant by difficulty of supervision) in C. H. Giles & Co. Ltd. v. Morris [1972] 1 W.L.R. 307, 318. In the present case, in the Court of Appeal took this claim at face value. In fact, Lord Wilberforce went on to say that impossibility of supervision "is a reality, no doubt, and explains why specific performance cannot be granted of agreements to this effect." Lord Wilberforce was in my view drawing attention to the fact that the collection of reasons which the courts have in mind when they speak of difficulty of supervision apply with much greater force to orders for specific performance, giving rise to the possibility of committal for contempt, than they do to conditions for relief against forfeiture. While the paradigm case to which such objections apply is the order to carry on an activity, they can also apply to an order requiring the achievement of a result.

     There is a further objection to an order requiring the defendant to carry on a business, which was emphasised by Millett L.J. in the Court of Appeal. This is that it may cause injustice by allowing the plaintiff to enrich himself at the defendant's expense. The loss which the plaintiff may suffer through having to comply with the order (for example, by running a business at a loss for an indefinite period) may be far greater than the plaintiff would suffer from the contract being broken. As Professor R. J. Sharpe explains (Specific Remedies for Contract Breach in Studies in Contract Law (ed. Reiter and Swan) at p. 129):

 "In such circumstances, a specific decree in favour of the plaintiff will put him in a bargaining position vis a vis the defendant whereby the measure of what he will receive will be the value to the defendant of being released from performance. If the plaintiff bargains effectively, the amount he will set will exceed the value to him of performance and will approach the cost to the defendant to complete."

This was the reason given by Lord Westbury L.C. in Isenberg v. East India House Estate Co. Ltd. (1863) 3 De G.J. & S. 263, 273 for refusing a mandatory injunction to compel the defendant to pull down part of a new building which interfered with the plaintiff's light and exercising instead the Court of Chancery's recently-acquired jurisdiction under Lord Cairns's Act 1858 to order payment of damages:

 ". . . I hold it . . . to be the duty of the court in such a case as the present not, by granting a mandatory injunction, to deliver over the defendants to the plaintiff bound hand and foot, in order to be made subject to any extortionate demand that he may by possibility make, but to substitute for such mandatory injunction an inquiry before itself, in order to ascertain the measure of damage that has been actually sustained."

It is true that the defendant has, by his own breach of contract, put himself in such an unfortunate position. But the purpose of the law of contract is not to punish wrongdoing but to satisfy the expectations of the party entitled to performance. A remedy which enables him to secure, in money terms, more than the performance due to him is unjust. From a wider perspective, it cannot be in the public interest for the courts to require someone to carry on business at a loss if there is any plausible alternative by which the other party can be given compensation. It is not only a waste of resources but yokes the parties together in a continuing hostile relationship. The order for specific performance prolongs the battle. If the defendant is ordered to run a business, its conduct becomes the subject of a flow of complaints, solicitors' letters and affidavits. This is wasteful for both parties and the legal system. An award of damages, on the other hand, brings the litigation to an end. The defendant pays damages, the forensic link between them is severed, they go their separate ways and the wounds of conflict can heal.

     The cumulative effect of these various reasons, none of which would necessarily be sufficient on its own, seems to me to show that the settled practice is based upon sound sense. Of course the grant or refusal of specific performance remains a matter for the judge's discretion. There are no binding rules, but this does not mean that there cannot be settled principles, founded upon practical considerations of the kind which I have discussed, which do not have to be re-examined in every case, but which the courts will apply in all but exceptional circumstances. As Slade J. said in the passage which I have quoted from Braddon Towers Ltd. v. International Stores Ltd. [1987] 1 E.G.L.R. 209, 213 lawyers have no doubt for many years advised their clients on this basis. In the present case, Leggatt L.J. remarked that there was no evidence that such advice had been given. In my view, if the law or practice on a point is settled, it should be assumed that persons entering into legal transactions will have been advised accordingly. I am sure that the learned Lord Justice would not wish to encourage litigants to adduce evidence of the particular advice which they received. Indeed, I doubt whether such evidence would be admissible.

 5.  The decision of the Court of Appeal

     I must now examine the grounds upon which the majority of the Court of Appeal thought it right to reverse the judge. In the first place, they regarded the practice which he followed as outmoded and treated Lord Wilberforce's remarks about relief against forfeiture in Shiloh Spinners Ltd. v. Harding [1973] A.C. 691, 724 as justifying a rejection of the arguments based on the need for constant supervision. Even Millett L.J., who dissented on other grounds, said that such objections had little force today. I do not agree. As I have already said, I think that Lord Wilberforce's remarks do not support this proposition in relation to specific performance of an obligation to carry on an activity and that the arguments based on difficulty of supervision remain powerful.

     The Court of Appeal said that it was enough if the contract defined the tenant's obligation with sufficient precision to enable him to know what was necessary to comply with the order. Even assuming that this to be right, I do not think that the obligation in clause 4(19) can possibly be regarded as sufficiently precise to be capable of specific performance. It is to "keep the demised premises open for retail trade." It says nothing about the level of trade, the area of the premises within which trade is to be conducted, or even the kind of trade, although no doubt the tenant's choice would be restricted by the need to comply with the negative covenant in clause 4(12)(a) not to use the premises "other than as a retail store for the sale of food groceries provisions and goods normally sold from time to time by a retail grocer food supermarkets and food superstores." This language seems to me to provide ample room for argument over whether the tenant is doing enough to comply with the covenant.

     The Court of Appeal thought that once Argyll had been ordered to comply with the covenant, it was, as Roch L.J. said, "inconceivable that they would not operate the business efficiently." Leggatt L.J. said that the requirement "was quite intelligible to the defendants, while they were carrying on business there. . . . If the premises are to be run as a business, it cannot be in the defendants' interest to run it half-heartedly or inefficiently. . ." This treats the way the tenant previously conducted business as measuring the extent of his obligation to do so. In my view this is a non sequitur: the obligation depends upon the language of the covenant and not upon what the tenant has previously chosen to do. No doubt it is true that it would not be in the interests of the tenant to run the business inefficiently. But running the business efficiently does not necessarily mean running in the way it was run before. Argyll had decided that, from its point of view, the most efficient thing to do was to close the business altogether and concentrate its resources on achieving better returns elsewhere. If ordered to keep the business open, it might well decide that the next best strategy was to reduce its costs as far as was consistent with compliance with its obligations, in the expectation that a lower level of return would be more than compensated by higher returns from additional expenditure on more profitable shops. It is in my view wrong for the courts to speculate about whether Argyll might voluntarily carry on business in a way which would relieve the court from having to construe its order. The question of certainty must be decided on the assumption that the court might have to enforce the order according to its terms.

     The respondent argued that the court should not be concerned about future difficulties which might arise in connection with the enforcement of the order. It should simply make the order and see what happened. In practice Argyll would be likely to find a suitable assignee (as it in fact did) or conduct the business so as to keep well clear of any possible enforcement proceedings or otherwise come to terms with the CIS. This may well be true, but the likelihood of Argyll having to perform beyond the requirements of its covenant or buy its way out of its obligation to incur losses seems to me to be in principle an objection to such an order rather than to recommend it. I think that it is normally undesirable for judges to make orders in terrorem, carrying a threat of imprisonment, which work only if no one inquires too closely into what they mean.

     The likelihood that the order would be effective only for a short time until an assignment is an equivocal argument. It would be burdensome to make Argyll resume business only to stop again after a short while if a short stoppage would not cause any substantial damage to the business of the shopping centre. On the other hand, what would happen if a suitable assignee could not be found? Would Argyll then have to carry on business until 2014? Mr. Smith Q.C. who appeared for the CIS, said that if the order became oppressive (for example, because Argyll were being driven into bankruptcy) or difficult to enforce, they could apply for it to be varied or discharged. But the order would be a final order and there is no case in this jurisdiction in which such an order has been varied or discharged, except when the injuncted activity has been legalised by statute. Even assuming that there was such a jurisdiction if circumstances were radically changed, I find it difficult to see how this could be made to apply. Difficulties of enforcement would not be a change of circumstances. They would have been entirely predictable when the order was made. And so would the fact that Argyll would suffer unquantifiable loss if it was obliged to continue trading. I do not think that such expedients are an answer to the difficulties on which the objections to such orders are based.

     Finally, all three judges in the Court of Appeal took a very poor view of Argyll's conduct. Leggatt L.J. said that they had acted "with gross commercial cynicism"; Roch L.J. began his judgment by saying that they had "behaved very badly" and Millett L.J. said that they had no merits. The principles of equity have always had a strong ethical content and nothing which I say is intended to diminish the influence of moral values in their application. I can envisage cases of gross breach of personal faith, or attempts to use the threat of non-performance as blackmail, in which the needs of justice will override all the considerations which support the settled practice. But although any breach of covenant is regrettable, the exercise of the discretion as to whether or not to grant specific performance starts from the fact that the covenant has been broken. Both landlord and tenant in this case are large sophisticated commercial organisations and I have no doubt that both were perfectly aware that the remedy for breach of the covenant was likely to be limited to an award of damages. The interests of both were purely financial: there was no element of personal breach of faith, as in the Victorian cases of railway companies which refused to honour obligations to build stations for landowners whose property they had taken: compare Greene v. West Cheshire Railway Co. (1871) L.R. 13 Eq. 44. No doubt there was an effect on the businesses of other traders in the Centre, but Argyll had made no promises to them and it is not suggested that CIS warranted to other tenants that Argyll would remain. Their departure, with or without the consent of CIS, was a commercial risk which the tenants were able to deploy in negotiations for the next rent review. On the scale of broken promises, I can think of worse cases, but the language of the Court of Appeal left them with few adjectives to spare.

     It was no doubt discourteous not to have answered Mr. Wightman's letter. But to say, as Roch L.J. did, that they had acted "wantonly and quite unreasonably" by removing their fixtures seems to me an exaggeration. There was no question of stealing a march, or attempting to present CIS with a fait accompli, because Argyll had no reason to believe that CIS would have been able to obtain a mandatory injunction whether the fixtures had been removed or not. They had made it perfectly clear that they were closing the shop and given CIS ample time to apply for such an injunction if so advised.

 6. Conclusion

     I think that no criticism can be made of the way in which His Honour Judge Maddocks Q.C. exercised his discretion. All the reasons which he gave were proper matters for him to take into account. In my view the Court of Appeal should not have interfered and I would allow the appeal and restore the order which he made.

LORD HOPE OF CRAIGHEAD

My Lords,

     I have had the benefit of reading in draft the speech which has been prepared by my noble and learned friend, Lord Hoffmann. I also agree that this appeal should be allowed and that the order by which the judge refused to order specific performance should be restored.

LORD CLYDE

My Lords,

     I have had the opportunity of reading in draft the speech of my noble and learned friend Lord Hoffmann. While I should wish to reserve my opinion on the approach which might be adapted by civilian systems I agree that the appeal should be allowed for the reasons which he has given.