England and Wales Court of Appeal (Civil Division) Decisions

IN THE SUPREME COURT OF JUDICATURE
COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM (1) Milton Keynes County Court
(2) Newcastle upon Tyne County Court
(1) His Honour Judge Charles Harris QC
(2) Mr Recorder Thomas QC

		Royal Courts of Justice Strand, London, WC2A 2LL
		11th May 2004

Lord Justice Dyson:

1. This is the judgment of the court.
2. These two appeals raise a question of some general importance: when should the court impose a costs sanction against a successful litigant on the grounds that he has refused to take part in an alternative dispute resolution ("ADR")? There seems to be some uncertainty as to the approach that should be adopted in answering this question: it has been the subject of consideration by courts on a number of occasions. A measure of its significance is that we have received detailed and helpful submissions from no fewer than four interveners, namely the Law Society, the Civil Mediation Council, the ADR Group and the Centre for Effective Dispute Resolution.
3. In the appeal of Halsey, the only ground of appeal is that, notwithstanding that the claim was dismissed, the judge was wrong to award the defendant, the Milton Keynes General NHS Trust ("the Trust") its costs, since it had refused a number of invitations by the claimant to mediate. There are two grounds of appeal in the case of Steel. First, it is said that the judge reached the wrong conclusion on the causation issue that he tried in the Part 20 proceedings between the defendants ("the causation issue"). Secondly, it is submitted that the judge was wrong to award the successful second defendant his costs against the first defendant, since the second defendant had refused a number of invitations by the first defendant to mediate. We shall start by giving some guidance as to the general approach that should be adopted when dealing with the costs issue raised by these two appeals. We shall then turn to the facts of the two appeals.

General encouragement of the use of ADR

4. As was explained in Lord Woolf's Final Report on Access to Justice (p 11), for some time before the Civil Procedure Rules ("CPR") came into force, resort by parties involved in litigation to ADR had been encouraged by the courts in various ways. The CPR, practice directions and pre-action protocols have built on these early developments. It is unnecessary to make extensive reference to demonstrate this. CPR 1.4(1) obliges the court to further the overriding objective of enabling the court to deal with cases justly by actively managing cases, and Rule 1.4(2)(e) defines "active case management" as including "encouraging the parties to use an alternative dispute resolution procedure if the court considers that appropriate and facilitating the use of such procedure". Rule 26.4(1) provides that "a party may, when filing the completed allocation questionnaire, make a written request for the proceedings to be stayed while the parties try to settle the case by alternative dispute resolution or other means".
5. The term "alternative dispute resolution" is defined in the Glossary to the CPR as a "collective description of methods of resolving disputes otherwise than through the normal trial process." In practice, however, references to ADR are usually understood as being references to some form of mediation by a third party. The general rule is that the unsuccessful party is ordered to pay the costs of the successful party (CPR 44.3(2)(a)). The cases in which the question of displacing this rule have been discussed have usually been concerned with the refusal of mediation by the successful party. The two appeals before this court fall into this category. In what follows we shall concentrate on the cost consequences of a refusal by the successful party to agree to mediation.
6. There are those who believe that the virtues of mediation have not yet been sufficiently demonstrated. There is some reference to this by Professor Hazel Genn in her excellent report "Court-based ADR initiatives for non-family civil disputes: the Commercial Court and the Court of Appeal" (March 2002), at pp 58-67. But we are in no doubt that we should proceed on the basis that there are many disputes which are suitable for mediation. This approach is consistent with, and (as we have seen) underpinned by, the Woolf reforms. It is also consistent with the fact there are now a number of court-based mediation schemes for civil non-matrimonial cases, which operate with varying degrees of success. The virtues of mediation in suitable cases are also recognised in the Chancery Guide (paras 17.1 and 17.3), the Queen's Bench Guide (para 6.6), the Admiralty and Commercial Court Guide (para D8.8) and the Technology and Construction Court Guide (para 6.4). Judges in the Commercial Court routinely make "ADR orders" in the form set out in Appendix 7 to the Admiralty and Commercial Court Guide (see further para 30 below).
7. We are also mindful of the position which has been taken by Government on this issue. Thus, in March 2001, the Lord Chancellor announced an "ADR Pledge" by which all Government departments and Agencies made a number of commitments including that: "Alternative Dispute Resolution will be considered and used in all suitable cases wherever the other party accepts it". In July 2002, the Department for Constitutional Affairs published a report as to the effectiveness of the Government's commitment to the ADR pledge. The report stated that the pledge had been taken very seriously, and identified a number of initiatives that had been introduced as a direct result of it. These included the following initiative on the part of the National Health Service Litigation Authority ("NHSLA"):

"The encouragement of greater use of mediation, and other forms of alternative dispute resolution, is one of the options considered by the NHSLA, who are responsible for handling clinical negligence claims against the NHS. The NHSLA is working with the Legal Services Commission to develop a joint strategy for promoting greater use of mediation as an alternative to litigation in clinical negligence disputes.

Since May 2000 the NHSLA has been requiring solicitors representing NHS bodies in such claims to offer mediation in appropriate cases, and to provide clear reasons to the authority if a case is considered inappropriate."

8. Strong support for the use of ADR in general , and mediation in particular, has been given by the courts in cases such as R (Cowl) v Plymouth City Council [2001] EWCA Civ 1935, [2002] 1 WLR 803, Dunnett v Railtrack plc [2002] EWCA Civ 303, [2002] 1 WLR 2434 and Hurst v Leeming [2001] EWHC 1051 (Ch), [2003] 1 Lloyds Rep 379.
9. We heard argument on the question whether the court has power to order parties to submit their disputes to mediation against their will. It is one thing to encourage the parties to agree to mediation, even to encourage them in the strongest terms. It is another to order them to do so. It seems to us that to oblige truly unwilling parties to refer their disputes to mediation would be to impose an unacceptable obstruction on their right of access to the court. The court in Strasbourg has said in relation to article 6 of the European Convention on Human Rights that the right of access to a court may be waived, for example by means of an arbitration agreement, but such waiver should be subjected to "particularly careful review" to ensure that the claimant is not subject to "constraint": see Deweer v Belgium (1980) 2 EHRR 439, para 49. If that is the approach of the ECtHR to an agreement to arbitrate, it seems to us likely that compulsion of ADR would be regarded as an unacceptable constraint on the right of access to the court and, therefore, a violation of article 6. Even if (contrary to our view) the court does have jurisdiction to order unwilling parties to refer their disputes to mediation, we find it difficult to conceive of circumstances in which it would be appropriate to exercise it. We would adopt what the editors of Volume 1 of the White Book (2003) say at para 1.4.11:

"The hallmark of ADR procedures, and perhaps the key to their effectiveness in individual cases, is that they are processes voluntarily entered into by the parties in dispute with outcomes, if the parties so wish, which are non-binding. Consequently the court cannot direct that such methods be used but may merely encourage and facilitate."

10. If the court were to compel parties to enter into a mediation to which they objected, that would achieve nothing except to add to the costs to be borne by the parties, possibly postpone the time when the court determines the dispute and damage the perceived effectiveness of the ADR process. If a judge takes the view that the case is suitable for ADR, then he or she is not, of course, obliged to take at face value the expressed opposition of the parties. In such a case, the judge should explore the reasons for any resistance to ADR. But if the parties (or at least one of them) remain intransigently opposed to ADR, then it would be wrong for the court to compel them to embrace it.
11. Parties sometimes need to be encouraged by the court to embark on an ADR. The need for such encouragement should diminish in time if the virtue of ADR in suitable cases is demonstrated even more convincingly than it has been thus far. The value and importance of ADR have been established within a remarkably short time. All members of the legal profession who conduct litigation should now routinely consider with their clients whether their disputes are suitable for ADR. But we reiterate that the court's role is to encourage, not to compel. The form of encouragement may be robust: see para 30 below.

The costs issue

12. CPR 44.3(2) provides that "if the court decides to make an order about costs (a) the general rule is that the unsuccessful party will be ordered to pay the cost of the successful party; but (b) the court may make a different order". CPR 44.3(4) provides that "in deciding what order (if any) to make about costs, the court must have regard to all the circumstances, including-(a) the conduct of the parties". Rule 44.3(5) provides that the conduct of the parties includes "(a) conduct before, as well as during, the proceedings and in particular the extent to which the parties followed any relevant pre-action protocol."
13. In deciding whether to deprive a successful party of some or all of his costs on the grounds that he has refused to agree to ADR, it must be borne in mind that such an order is an exception to the general rule that costs should follow the event. In our view, the burden is on the unsuccessful party to show why there should be a departure from the general rule. The fundamental principle is that such departure is not justified unless it is shown (the burden being on the unsuccessful party) that the successful party acted unreasonably in refusing to agree to ADR. We shall endeavour in this judgment to provide some guidance as to the factors that should be considered by the court in deciding whether a refusal to agree to ADR is unreasonable.
14. We make it clear at the outset that it was common ground before us (and we accept) that parties are entitled in an ADR to adopt whatever position they wish, and if as a result the dispute is not settled, that is not a matter for the court. As is submitted by the Law Society, if the integrity and confidentiality of the process is to be respected, the court should not know, and therefore should not investigate, why the process did not result in agreement.
15. We recognise that mediation has a number of advantages over the court process. It is usually less expensive than litigation which goes all the way to judgment, although it should not be overlooked that most cases are settled by negotiation in the ordinary way. Mediation provides litigants with a wider range of solutions than those that are available in litigation: for example, an apology; an explanation; the continuation of an existing professional or business relationship perhaps on new terms; and an agreement by one party to do something without any existing legal obligation to do so. As Brooke LJ pointed out in Dunnett at para [14]:

"Skilled mediators are now able to achieve results satisfactory to both parties in many cases which are quite beyond the power of lawyers and courts to achieve. This court has knowledge of cases where intense feelings have arisen, for instance in relation to clinical negligence claims. But when the parties are brought together on neutral soil with a skilled mediator to help them resolve their differences, it may very well be that the mediator is able to achieve a result by which the parties shake hands at the end and feel that they have gone away having settled the dispute on terms with which they are happy to live. A mediator may be able to provide solutions which are beyond the powers of the court to provide."

16. In deciding whether a party has acted unreasonably in refusing ADR, these considerations should be borne in mind. But we accept the submission made by the Law Society that mediation and other ADR processes do not offer a panacea, and can have disadvantages as well as advantages: they are not appropriate for every case. We do not, therefore, accept the submission made on behalf of the Civil Mediation Council that there should be a presumption in favour of mediation. The question whether a party has acted unreasonably in refusing ADR must be determined having regard to all the circumstances of the particular case. We accept the submission of the Law Society that factors which may be relevant to the question whether a party has unreasonably refused ADR will include (but are not limited to) the following: (a) the nature of the dispute; (b) the merits of the case; (c) the extent to which other settlement methods have been attempted; (d) whether the costs of the ADR would be disproportionately high; (e) whether any delay in setting up and attending the ADR would have been prejudicial; and (f) whether the ADR had a reasonable prospect of success. We shall consider these in turn. We wish to emphasise that in many cases no single factor will be decisive, and that these factors should not be regarded as an exhaustive check-list.
17. (a) The nature of the dispute. Even the most ardent supporters of ADR acknowledge that the subject-matter of some disputes renders them intrinsically unsuitable for ADR. The Commercial Court Working Party on ADR stated in 1999:

"The Working Party believes that there are many cases within the range of Commercial Court work which do not lend themselves to ADR procedures. The most obvious kind is where the parties wish the court to determine issues of law or construction which may be essential to the future trading relations of the parties, as under an on-going long term contract, or where the issues are generally important for those participating in a particular trade or market. There may also be issues which involve allegations of fraud or other commercially disreputable conduct against an individual or group which most probably could not be successfully mediated."

Other examples falling within this category are cases where a party wants the court to resolve a point of law which arises from time to time, and it is considered that a binding precedent would be useful; or cases where injunctive or other relief is essential to protect the position of a party. But in our view, most cases are not by their very nature unsuitable for ADR.

18. (b) The merits of the case. The fact that a party reasonably believes that he has a strong case is relevant to the question whether he has acted reasonably in refusing ADR. If the position were otherwise, there would be considerable scope for a claimant to use the threat of costs sanctions to extract a settlement from the defendant even where the claim is without merit. Courts should be particularly astute to this danger. Large organisations, especially public bodies, are vulnerable to pressure from claimants who, having weak cases, invite mediation as a tactical ploy. They calculate that such a defendant may at least make a nuisance-value offer to buy off the cost of a mediation and the risk of being penalised in costs for refusing a mediation even if ultimately successful.
19. Some cases are clear-cut. A good example is where a party would have succeeded in an application for summary judgment pursuant to CPR 24.2, but for some reason he did not make such an application. Other cases are more border-line. In truly border-line cases, the fact that a party refused to agree to ADR because he thought that he would win should be given little or no weight by the court when considering whether the refusal to agree to ADR was reasonable. Border-line cases are likely to be suitable for ADR unless there are significant countervailing factors which tip the scales the other way. In Hurst, Lightman J said:

"The fact that a party believes that he has a watertight case again is no justification for refusing mediation. That is the frame of mind of so many litigants."

In our judgment, this statement should be qualified. The fact that a party unreasonably believes that his case is watertight is no justification for refusing mediation. But the fact that a party reasonably believes that he has a watertight case may well be sufficient justification for a refusal to mediate.

20. (c) Other settlement methods have been attempted. The fact that settlement offers have already been made, but rejected, is a relevant factor. It may show that one party is making efforts to settle, and that the other party has unrealistic views of the merits of the case. But it is also right to point out that mediation often succeeds where previous attempts to settle have failed. Although the fact that settlement offers have already been made is potentially relevant to the question whether a refusal to mediate is unreasonable, on analysis it is in truth no more than an aspect of factor (f).
21. (d) The costs of mediation would be disproportionately high. This is a factor of particular importance where, on a realistic assessment, the sums at stake in the litigation are comparatively small. A mediation can sometimes be at least as expensive as a day in court. The parties will often have legal representation before the mediator, and the mediator's fees will usually be borne equally by the parties regardless of the outcome (although the costs of a mediation may be the subject of a costs order by the court after a trial). Since the prospects of a successful mediation cannot be predicted with confidence (see further para 27 below), the possibility of the ultimately successful party being required to incur the costs of an abortive mediation is a relevant factor that may be taken into account in deciding whether the successful party acted unreasonably in refusing to agree to ADR.
22. (e) Delay. If mediation is suggested late in the day, acceptance of it may have the effect of delaying the trial of the action. This is a factor which it may be relevant to take into account in deciding whether a refusal to agree to ADR was unreasonable.
23. (f) Whether the mediation had a reasonable prospect of success. In Hurst, Lightman J said that he considered that the "critical factor" in that case was whether "objectively viewed" a mediation had any real prospect of success. He continued (p 381):

"If mediation can have no real prospect of success, a party may, with impunity, refuse to proceed to mediation on this ground. But refusal is a high risk course to take, for if the Court finds that there was a real prospect, the party refusing to proceed to mediation may, as I have said, be severely penalized. Further, the hurdle in the way of a party refusing to proceed to mediation on this ground is high, for in making this objective assessment of the prospects of mediation, the starting point must surely be the fact that the mediation process itself can and often does bring about a more sensible and more conciliatory attitude on the part of the parties than might otherwise be expected to prevail before the mediation, and may produce a recognition of the strengths and weaknesses by each party of his own case and of that of his opponent, and a willingness to accept the give and take essential to a successful mediation. What appears to be incapable of mediation before the mediation process begins often proves capable of satisfactory resolution later."

24. Consistently with the view expressed in this passage, Lightman J said that on the facts of that case he was persuaded that "quite exceptionally" the successful party was justified in taking the view that mediation was not appropriate because it had no realistic prospects of success.
25. In our view, the question whether the mediation had a reasonable prospect of success will often be relevant to the reasonableness of A's refusal to accept B's invitation to agree to it. But it is not necessarily determinative of the fundamental question, which is whether the successful party acted unreasonably in refusing to agree to mediation. This can be illustrated by a consideration of two cases. In a situation where B has adopted a position of intransigence, A may reasonably take the view that a mediation has no reasonable prospect of success because B is most unlikely to accept a reasonable compromise. That would be a proper basis for concluding that a mediation would have no reasonable prospect of success, and that for this reason A's refusal to mediate was reasonable.
26. On the other hand, if A has been unreasonably obdurate, the court might well decide, on that account, that a mediation would have had no reasonable prospect of success. But obviously this would not be a proper reason for concluding that A's refusal to mediate was reasonable. A successful party cannot rely on his own unreasonableness in such circumstances. We do not, therefore, accept that, as suggested by Lightman J, it is appropriate for the court to confine itself to a consideration of whether, viewed objectively, a mediation would have had a reasonable prospect of success. That is an unduly narrow approach: it focuses on the nature of the dispute, and leaves out of account the parties' willingness to compromise and the reasonableness of their attitudes.
27. Nor should it be overlooked that the potential success of a mediation may not only depend on the willingness of the parties to compromise. Some disputes are inherently more intractable than others. Some mediators are more skilled than others. It may therefore, sometimes be difficult for the court to decide whether the mediation would have had a reasonable prospect of success.
28. The burden should not be on the refusing party to satisfy the court that mediation had no reasonable prospect of success. As we have already stated, the fundamental question is whether it has been shown by the unsuccessful party that the successful party unreasonably refused to agree to mediation. The question whether there was a reasonable prospect that a mediation would have been successful is but one of a number of potentially relevant factors which may need to be considered in determining the answer to that fundamental question. Since the burden of proving an unreasonable refusal is on the unsuccessful party, we see no reason why the burden of proof should lie on the successful party to show that mediation did not have any reasonable prospect of success. In most cases it would not be possible for the successful party to prove that a mediation had no reasonable prospect of success. In our judgment, it would not be right to stigmatise as unreasonable a refusal by the successful party to agree to a mediation unless he showed that a mediation had no reasonable prospect of success. That would be to tip the scales too heavily against the right of a successful party to refuse a mediation and insist on an adjudication of the dispute by the court. It seems to us that a fairer balance is struck if the burden is placed on the unsuccessful party to show that there was a reasonable prospect that mediation would have been successful. This is not an unduly onerous burden to discharge: he does not have to prove that a mediation would in fact have succeeded. It is significantly easier for the unsuccessful party to prove that there was a reasonable prospect that a mediation would have succeeded than for the successful party to prove the contrary.
29. So far we have been considering the question whether a successful party's refusal of ADR was unreasonable without regard to the impact of any encouragement that the court may have given in the particular case. Where a successful party refuses to agree to ADR despite the court's encouragement, that is a factor which the court will take into account when deciding whether his refusal was unreasonable. The court's encouragement may take different forms. The stronger the encouragement, the easier it will be for the unsuccessful party to discharge the burden of showing that the successful party's refusal was unreasonable.
30. An ADR order made in the Admiralty and Commercial Court in the form set out in Appendix 7 to the Guide is the strongest form of encouragement. It requires the parties to exchange lists of neutral individuals who are available to conduct "ADR procedures", to endeavour in good faith to agree a neutral individual or panel and to take "such serious steps as they may be advised to resolve their disputes by ADR procedures before the neutral individual or panel so chosen". The order also provides that if the case is not settled, "the parties shall inform the court what steps towards ADR have been taken and (without prejudice to matters of privilege) why such steps have failed". It is to be noted, however, that this form of order stops short of actually compelling the parties to undertake an ADR.
31. Nevertheless, a party who, despite such an order, simply refuses to embark on the ADR process at all would run the risk that for that reason alone his refusal to agree to ADR would be held to have been unreasonable, and that he should therefore be penalised in costs. It is to be assumed that the court would not make such an order unless it was of the opinion that the dispute was suitable for ADR.
32. A less strong form of encouragement is mentioned in the other Court Guides to which we have referred at para 6 above. A particularly valuable example is the standard form of order now widely used in clinical negligence cases, and which was devised by Master Ungley. The material parts of this order provide:

"The parties shall by consider whether the case is capable of resolution by ADR. If any party considers that the case is unsuitable for resolution by ADR, that party shall be prepared to justify that decision at the conclusion of the trial, should the judge consider that such means of resolution were appropriate, when he is considering the appropriate costs order to make.

The party considering the case unsuitable for ADR shall, not less than 28 days before the commencement of the trial, file with the court a witness statement without prejudice save as to costs, giving reasons upon which they rely for saying that the case was unsuitable."

33. This form of order has the merit that (a) it recognises the importance of encouraging the parties to consider whether the case is suitable for ADR, and (b) it is calculated to bring home to them that, if they refuse even to consider that question, they may be at risk on costs even if they are ultimately held by the court to be the successful party. We can see no reason why such an order should not also routinely be made at least in general personal injury litigation, and perhaps in other litigation too. A party who refuses even to consider whether a case is suitable for ADR is always at risk of an adverse finding at the costs stage of litigation, and particularly so where the court has made an order requiring the parties to consider ADR.

Public bodies

34. Another issue that has arisen is whether the court should be particularly disposed to make an adverse costs order against a successful public body on the grounds that it refused to agree to ADR. We can see no basis for the court discriminating against successful public bodies when deciding whether a refusal to agree to ADR should result in a costs penalty. The only reason for doing so that was suggested to us in the course of argument was that Government departments and agencies (including the NHSLA) should be held to the ADR pledge (see para 7 above). We need, therefore, to consider whether the ADR pledge has any special significance. In Royal Bank of Canada v Secretary of State for Defence [2003] EWHC 1841 (Ch) the main issue was the true interpretation of a lease. Lewison J said that, although it concerned a question of law, this dispute was suitable for ADR. He considered that the ADR pledge given by Government was something to which he ought to attach "great weight". At para 12 of his judgment he said:

"As I have said, however, the most important feature to my mind is the formal pledge given on behalf of the government and its various departments to use ADR in appropriate cases. The government did not abide by that pledge in this case. I am not in a position to form any real view of whether a mediation would or would not have succeeded. It may well have done, but in my judgment a failure to abide by the formal pledge given on the part of government, coupled with the fact that……, justifies a decision that the defendant should not recover any further costs from the claimant."

35. In our judgment, the judge was wrong to attach such weight to the ADR pledge. The pledge was no more than an undertaking that ADR would be considered and used in all suitable cases. If a case is suitable for ADR, then it is likely that a party refusing to agree to it will be acting unreasonably, whether or not it is a public body to which the ADR pledge applies. If the case is not suitable for ADR, then a refusal to agree to ADR does not breach the pledge. It is, therefore, difficult to see in what circumstances it would be right to give great weight to the ADR pledge.

HALSEY v MILTON KEYNES GENERAL NHS TRUST

The facts

36. This claim is brought by Lilian Halsey pursuant to the provisions of the Fatal Accidents Act 1976 (as amended) in relation to the death of her husband, Bert Halsey, on 27 June 1999 at the Milton Keynes General Hospital. At the date of his death, the deceased was 83 years of age. The claim arose out of the allegedly negligent treatment of him while he was a patient at that hospital.
37. He was transferred to the hospital on 25 June 1999. At that time, he was suffering from several serious health problems. It was not in dispute that his life expectation was short. He was being fed by means of a nasogastric feeding tube. The basis of the claim was that his death was caused by the tube being incorrectly fitted, so that, rather than directing liquid food into the deceased's stomach, it was directing it into his left lung. Following his death, a post mortem examination was performed by Dr Mayers, a consultant histopathologist, employed by the defendant Trust. She concluded that the cause of death was airway obstruction due to the introduction of nasogastric nutrition into the airway and lung as a result of the insertion of a nasogastric tube into the major airway.
38. An inquest was held into the deceased's death. For this purpose, HM Coroner requested statements from the medical and nursing staff most closely involved in the deceased's care at the hospital. The claimant instructed Messrs Osborne Morris and Morgan ("OMM") to attend the inquest on behalf of the family. The Trust refused a request that it should meet the cost of OMM's preparation and attendance at the inquest. On 11 January 2000, OMM wrote to the Trust's solicitors saying:

"We should perhaps point out that our clients would have agreed to limit their entire claim in this matter to the costs of the representation of the inquest. Indeed they would have been prepared to limit the preparation and attendance to £5000. No doubt your clients will bear this in mind when we issue proceedings against them following the inquest. The family have been forced to issue proceedings simply because of your clients refusal to meet the costs of the attendance".

39. The inquest was held on 14 January 2000. The Coroner heard evidence from several nurses, Dr Mayers and Dr Lanzone Miller, a consultant physician and gastro-enterologist. Dr Miller disagreed with Dr Mayers' view that nutrition had entered the airway and lung due to the insertion of the nasogastric tube into the major airway. Dr Miller considered that the explanation for the presence of nutritional feed in the lung was regurgitation followed by aspiration of the stomach contents at or around the time of death. At the conclusion of the evidence, the Coroner recorded the medical cause of death as:

"1(a) airway obstruction;

1(b) introduction of nasogastric nutrition into airway and lungs;

2 chronic renal failure; old myocardial infarct; chronic obstructive pulmonary disease and fractured ribs".

40. By a letter dated 18 January 2000, OMM informed the Trust that the deceased's family would be prepared to accept a payment of £7500 bereavement damages together with a contribution towards their costs. The Trust's solicitors replied on 4 February saying that they were instructed "to take all necessary steps to resist a claim, to trial if necessary, should one be forthcoming". On 7 February, OMM asked the Trust whether they were prepared to refer the matter "for alternative dispute resolution that is, mediation, again to ensure that unnecessary costs are incurred (sic). Our client wants to mediate this claim".
41. On 6 April 2000, OMM wrote to the Secretary of State for Health giving him early notice of the proceedings, and saying:

"You will see from the correspondence that I have at every juncture sought to meet, negotiate and mediate this claim with the least amount of cost to the NHS. Unfortunately all such attempts have been rejected. I want you to have this correspondence in mind when the final bill payable by the NHS for legal costs is in the region of £100,000."

42. By letter dated 14 March 2000, the Trust's solicitors replied saying that they did not accept that there was any claim arising out of the care provided to the deceased, and that all necessary steps would be taken to resist a claim, to trial if necessary.
43. On 25 April 2001, OMM wrote to the Trust a letter of claim in accordance with the clinical negligence pre-action protocol. The letter included an offer to accept the sum of £12,500 together with reasonable costs pursuant to CPR Part 36. This evoked a response on 1 May in which the Trust said that it would be defending the claim. On 25 July 2001 the Trust sent its protocol response denying liability, and asserting:

"As there is no negligence, the Trust will not be settling this claim. In view of this, I do not believe it is appropriate to meet with you and discuss this claim or refer this case to mediation as the Trust's stance will not change in this respect. Therefore, any such meeting or mediation will be unnecessary waste of both costs and resources".

44. On 5 November 2001, OMM wrote to the Trust saying:

"We must make it clear to you at this stage that it was never our intention to issue proceedings since this was quite obviously a case that could have been resolved by mediation".

45. The claim form was issued on 13 May, and served on 9 September 2002. On 10 September 2002, OMM wrote:

"Despite the fact that we have issued proceedings in this matter we are anxious to avoid unnecessary costs being incurred. We would therefore invite you to consider referring this case to mediation so that we can perhaps resolve it to the satisfaction of our client without unnecessary costs being incurred".

46. This request was repeated by OMM's letter of 13 September 2002. The Trust replied on 16 September:

"I have as yet been provided with no evidence that it would be an effective use of NHS resources to go to mediation on this small value claim, where liability is in dispute. What your costs are, are obviously a matter for you and your client, the Trust costs on this claim, even to trial will only be low".

47. On 3 April 2003, OMM wrote yet again urging the Trust to consider either mediation or at least a meeting to discuss a possible settlement:

"We are certain agreement could be reached whereby further unnecessary costs would be avoided".

48. The Trust replied on 4 April, repeating its position that "on such a low quantum claim, we do not consider this to be a cost effective use of NHS resources". On 30 April 2003, OMM wrote again pointing out that they had made every attempt to try to avoid the costs of litigation, but all such attempts had been rejected out of hand. They referred to the decision of this court in Dunnett and that of Lightman J in Hurst, and said that they would be relying on these decisions when the court came to consider the question of costs, since the Trust had acted unreasonably. On 1 May 2003, the Trust replied saying:

"In respect of your request that we clarify why "mediation is not a cost effective use of NHS resources on such a low cost claim", (a) it has little chance of success and (b) the costs of mediation would be as great, if not greater for such a low value claim than attending trial. … if you do not intend to consider a "drop hands" agreement, then I look forward to receiving your indexed trial bundle and summary in order that we may consider and amend/agree them".

49. The trial took place on 1 and 2 June 2003. The judge gave judgment on 17 June, and dismissed the claim. He then heard argument on the question of costs. It was submitted by Mr Meakin on behalf of the claimant that there should be no order as to costs, and in support of his submission he relied on the refusal of the Trust to agree to a mediation. The judge was referred to some, if not all, of the correspondence to which we have referred. He said that he had the feeling that the letters written by OMM had been "somewhat tactical". He also thought that it was unusual to commence litigation by writing to the Secretary of State for Health as occurred in this case. At para 32 of his judgment, he said:

"I think that the question for me to decide is was the defendant's attitude to ADR a reasonable one, or not a reasonable one or to use the words of Lightman J, were they justified in taking the view that mediation was not appropriate because it had no realistic prospect of success? Having considered and been taken through much of this correspondence, although, as I say, notably not the correspondence preceding the letter to the Health Secretary in 2000, it seems reasonably clear to me that although there are a number of tactical observations to the effect that the claimant would like ADR, it remains perfectly clear that this is not a case which they were disposed to compromise on any terms which could possibly be reasonable to the defendants. The Defendants took the view, legitimately as I have found, that this was a case in which there simply was no negligence. It is significant that I have been told that on 4^th March of this year, fairly shortly before trial, an offer was made by the defendant to the claimant that the matter should be settled on the basis that both parties simply walked away from the litigation and neither paid their costs, but that did not attract the advisors to the claimant or the claimant herself. She started off asking for £12,000, towards the end was offering to accept £7,500 and funeral expenses. It seems perfectly clear that it would only have been by the payment of some significant, albeit modest by the standards of this type of litigation, sum of money that the defendants would have been able to buy off the claim, whether by ADR or in any other way. I do not think that the CPR is designed to make parties which have a good defence settle claims which they do not wish to settle, when they ultimately end up winning and are vindicated in the view that they have taken, I do not think it proper, at least not in a run of the mill case, to say that they should then suffer by being denied their costs. This was a perfectly straightforward case of a kind which the defendants were justified in defending and justified in saying that they did not want to pay any money to the claimant in respect of. The claimant was only prepared to settle on the basis that she was going to be paid some money, and therefore there was not much point in talking to any greater extent than they did. It is to be observed that the correspondence from the defendant was both full and reasonable."

The costs issue

50. In the light of our exposition of what we conceive to be the correct approach, we have no difficulty in concluding that the judge was correct to decide that the Trust should not be deprived of any of its costs on the grounds that it had refused to accept the claimant's invitations to agree to a mediation. In our view, the claimant has come nowhere near showing that the Trust acted unreasonably in refusing to agree to a mediation. We start by noting that this is not a case where the court made any order encouraging mediation. We accept that the subject-matter of this dispute was not by its nature unsuitable for ADR. But the Trust believed that it had a strong defence, and had reasonable grounds for that belief. Moreover, the judge was justified in saying that the letters written by the claimant's solicitors were "somewhat tactical". We think that, if anything, this was an understatement. The extraordinary letter written to the Secretary of State very early on was an attempt to extort a sum (plus costs) in settlement of a very small claim which, at best, was speculative. The writing of no fewer than 5 letters asking the Trust to agree to mediation was of a piece with this early letter. The requirement that the Trust should pay the claimant's costs is of particular significance, since she had entered into a conditional fee agreement with her solicitors, with a 100% success fee.
51. Another highly relevant feature of this case was that the Trust reasonably took the view that the costs of a mediation would be disproportionately high when compared with (a) the value of the claim if liability were to be established and (b) the Trust's costs of a trial. The Trust was entitled to regard this as a factor strongly militating against mediation.
52. Nor did the claimant discharge the burden of proving that mediation had a reasonable prospect of success. The Trust had taken the view that the claim would not succeed, and had decided not to make any payment to the claimant. Mr Allan Gore QC submits that the claimant might have been persuaded in the course of the mediation to drop her claim: all she really wanted was an explanation of how her husband had died in hospital. This possibility cannot be dismissed as fanciful, although there is no evidence to support it. But in our judgment, the claimant comes nowhere near proving that there was a reasonable prospect that the mediation would have been successful. In the circumstances of this case, the stance adopted by the Trust cannot fairly be said to have been unreasonable.
53. The final point urged by Mr Gore is the ADR pledge. But as we have already said, this pledge adds nothing. If the case was suitable for ADR, the claimant does not need the pledge; and if the case was not suitable for ADR, the pledge did not require the Trust to agree to mediation. The same applies to the initiative of the NHSLA referred to at para 7 above: the statement merely says that NHS bodies offer mediation in "appropriate" cases.
54. For all these reasons, we dismiss the appeal.

STEEL v JOY AND HALLIDAY

The causation issue

55. On 15 December 1996, the claimant was injured in an accident involving the first defendant. On 13 March 1999, he was injured in an accident involving the second defendant. The claimant brought separate actions against the two defendants. These were subsequently consolidated. The first defendant instituted CPR Part 20 proceedings against the second defendant, seeking a contribution in respect of any damages found payable to the claimant. Both defendants admitted liability to the claimant. The claimant and first defendant jointly instructed Mr M J Gibson FRCS, a consultant spinal surgeon. Mr Gibson advised that the claimant suffers from a congenital spinal stenosis, and that, as a result of the first accident, symptoms of the claimant's pre-existing stenosis were accelerated by 7-10 years. Of the second accident, Mr Gibson said:

"He aggravated the pre-existing problems producing an exacerbation of these which would have lasted for in the order of 3 to 6 months. Thereafter persistence of symptoms primarily relates to his pre-existing problem that started after the injury on the 15^th December 1996".

56. Mr Gibson was asked whether the second accident would have had a similar effect to the first accident had the claimant been free of symptoms immediately prior to the second accident. He replied by letter dated 25 June 2002 that: "on the balance of probabilities, this second accident would have exacerbated Mr Steel's condition by 7 to 10 years". Mr Recorder Thomas QC summarised the effect of the agreed medical evidence at para 10 of his judgment in these terms:

"So what it comes to, in very simple terms, is that as a result of the first accident, the claimant's already existing problems were accelerated by a period of years, in the order of 7 to 10 years. In the second accident, two and a quarter years later, Mr Gibson makes it clear that there was an exacerbation of 3 to 6 months by reason of the second accident, and we understand (and when I say "we understand" this is raised by me in argument with Counsel today), we understand that really what he is meaning by the 3 to 6 months aggravation from the second accident is that there was a flare-up of conditions for that limited period of time as a result of the second accident. He says in the report and the correspondence I have referred to that effectively, if the first accident had not happened, then the second accident would have had the same effect as the first accident if the first accident had not already occurred. However, save for the 3 to 6 months aggravation, the second accident did not in fact affect the claimant's existing medical condition which had already been arrived at by then as a result of his pre-existing condition and the first accident".

57. The trial took place on 3 July 2003. It was agreed that the Recorder should determine the causation issue that had been raised on the pleadings, namely whether the second defendant had caused the claimant to suffer any more damage than 3 to 6 months' aggravation of his symptoms of stenosis. On behalf of the first defendant, it was submitted that the second defendant had also caused or contributed to the 7-10 years acceleration of the claimant's symptoms. The second defendant contested this. The Recorder found in favour of the second defendant. He held that the second accident "did not affect the long-term prognosis that there already was from the first accident" (para 19), and derived support for his conclusion from the decision of this court in Performance Cars Ltd v Abraham [1962] 1 QB 33. The first defendant appeals against this decision.
58. Mr Foster's primary submission is that, since there was an "exact overlap of damage", the second accident damage can be said to have overtaken the first accident damage, so that the first defendant is only liable for the damage which was suffered during the period between the dates of the two accidents. His alternative submission is that the two defendants should be regarded as concurrent tortfeasors who, as regards the acceleration of the stenosis symptoms, both caused the same damage, and between whom there would be rights of contribution under section 1 of the Civil Liability (Contribution) Act 1978.
59. The first of these submissions is completely unsustainable. The phrase "exact overlap of damage" is not apt on the facts of this case. The damage attributable to the first accident (acceleration of symptoms by 7-10 years) had already occurred by the date of the second accident. That historical fact cannot be expunged simply because that same damage would have been caused by the second accident if the first accident had not occurred. In these circumstances, it is a misuse of language to describe the acceleration of symptoms by 7-10 years as "exactly overlapping damage", and plainly wrong to say that the second accident damage overtook and extinguished the first accident damage.
60. Nor can we accept the alternative submission. In our judgment, this case cannot be distinguished from Performance Cars. Mr Foster implicitly accepts this, because he contends that Performance Cars was wrongly decided and should not be followed. In that case, the defendant negligently caused his car to collide with the plaintiff's car and damaged its front wing. It was agreed that to make good the damage, the whole of the lower part of the car would have to be resprayed at a cost of £75. The plaintiff had previously been involved in another collision in which his car had suffered damage to the rear wing which had not been made good. This damage also required a similar respray. The plaintiff had sued the person responsible for the first damage and recovered judgment for £75, the cost of the respray. That judgment had not been satisfied. This court held that the plaintiff was not entitled to recover the cost of the respray from the defendant, since he had damaged a car which was at the time of the accident in need of respraying, with the result that the need for respraying did not flow from the defendant's wrongdoing. Accordingly, the claim against the defendant failed. Lord Evershed MR said that "the necessity for respraying was not the result of the defendant's wrongdoing because that necessity already existed" (p 39). At p 40, he said:

"In my judgment in the present case the defendant should be taken to have injured a motor-car that was already in certain respects (that is in respect of the need for respraying) injured; with the result that to the extent of that need or injury the damage claimed did not flow from the defendant's wrongdoing. It may no doubt be unfortunate for the plaintiffs that the collisions took place in the order in which they did."

61. Donovan LJ said at p 42:

"The question as I see it is this: what extra burden in the matter of respraying was put upon the plaintiff company by the second collision? To my mind the answer must be: None, for the earlier collision had already imposed the burden of respraying upon them."

62. Mr Foster submits that the law has moved on since Performance Cars was decided and that the court is now required to apply what he calls "equitable pragmatism" in a case such as this. He says that justice requires the court to hold that the two defendants were concurrent tortfeasors who were both responsible for the same damage. The analysis adopted by the judge in the present case would mean that, if the first defendant had not been before the court or had been insolvent, the claimant would not have been compensated for his loss: such an unjust conclusion can, and should, be avoided. Justice requires the court to hold that both defendants caused the damage, and that, as between themselves, the court should assess the contribution that is just and equitable having regard to their respective responsibilities for the damage in question: see section 2(1) of the 1978 Act.
63. In support of these submissions, Mr Foster relies in particular on Rahman v Arearose Ltd [2001] QB 351. In that case, the claimant who was employed by D1 was assaulted at work by two fellow employees and suffered an injury to his right eye. As a result of medical negligence committed by D2 in the course of an operation, he later lost the sight in his eye entirely. He also suffered post-traumatic stress disorder and depression. At p 364D of his judgment, Laws LJ rejected the submission that the case was one of "concurrent torts", since on the evidence the respective torts committed by the defendants were the causes of distinct aspects of the claimant's psychiatric condition. At paras 26-33, he discussed the issue of causation, and said: "Once it is recognised that the first principle is that every tortfeasor should compensate the injured claimant in respect of that loss and damage for which he should justly be held responsible, the metaphysics of causation can be kept in their proper place" (para 32). It is this passage on which Mr Foster places particular reliance. But it is important to point out that Laws LJ also said at para 34:

"Once one leaves behind, as for the reasons I have given one should, the dogmas of novus actus and eggshell skulls, there is nothing in the way of a sensible finding that while the second defendants obviously (and exclusively) caused the right-eye blindness, thereafter each tort had its part to play in the claimant's suffering."

64. And at para 36, he said:

"Here, the question what the position would have been if the second tort had not been committed is highly material: the second defendants are not to be held responsible for damage the whole of which had already been inflicted on the claimant by the first defendants."

65. It seems to us that there is nothing in the decision in Rahman which supports Mr Foster's alternative submission.
66. Reference was made during the course of argument to Baker v Willoughby [1970] AC 467. In that case, the plaintiff's leg, which was injured in a car accident caused by the negligence of the defendant, was later shot by some robbers. The House of Lords held that the defendant was liable for the full consequences of the injury he caused regardless of the second incident, and had to pay damages based on the plaintiff's losses beyond the time when his leg was amputated as a result of the second incident. This decision was criticised by some of their lordships in the House of Lords in Jobling v Associated Dairies [1982] AC 794. It is unnecessary to consider the speeches in Jobling. For present purposes, it is sufficient to state that Baker should be regarded as an exception to the general "but-for" test, which was justified by the principle of fully compensating the plaintiff for damage tortiously inflicted. It was recognised that, if the defendant's argument were accepted, namely that he had no liability in respect of the period after the plaintiff's leg had been amputated, the plaintiff would fall between two defendants, and not be entitled to full compensation.
67. This point was most clearly articulated by Lord Pearson who said of the defendant's argument at p 495E:

"That is the argument, and it is formidable. But it must not be allowed to succeed, because it produces manifest injustice. The supervening event has not made the plaintiff less lame nor less disabled nor less deprived of amenities. It has not shortened the period over which he will be suffering. It has made him more lame, more disabled, more deprived of amenities. He should not have less damages through being worse off than might have been expected.

The nature of the injustice becomes apparent if the supervening event is treated as a tort (as indeed it was) and if one envisages the plaintiff suing the robbers who shot him. They would be entitled, as the saying is, to "take the plaintiff as they find him." (Performance Cars Ltd v Abraham [1962] 1 QB 33.) They have not injured and disabled a previously fit and able-bodied man. They have only made an already lame and disabled man more lame and more disabled."

68. Lord Reid had also referred to Performance Cars at p 493F, and said at p 493G:

"These cases exemplify the general rule that a wrongdoer must take the plaintiff (or his property) as he finds him: that may be to his advantage or disadvantage. In the present case the robber is not responsible or liable for the damage caused by the respondent: he would only have to pay for additional loss to the appellant by reason of his now having an artificial limb instead of a stiff leg."

69. The importance of Baker for present purposes is that their lordships recognised that Performance Cars was good law. Both Lord Reid and Lord Pearson explicitly stated that the robbers would not have been liable for the plaintiff's loss of a good leg. It was precisely because Performance Cars was good law that under-compensation of the plaintiff could only be avoided by making the defendant liable for the loss attributable to the leg injury even after the amputation. Accordingly, the "but-for" test could not be applied, just as it has not been applied in cases involving multiple tortfeasors such as Fairchild v Glenhaven Funeral Services Ltd [2003] UKHL 22, [2003] 1 AC 32.
70. In our judgment, Performance Cars is still good law. It has been frequently referred to in the textbooks and, so far as we know, without disapproval. As a matter of logic and common sense, it is clearly correct. We do not consider that it produces an unjust result. The claimant is entitled to recover damages from the first defendant for the losses inflicted by him; and from the second defendant for any additional losses inflicted by him. It is true that, if the first defendant is not before the court or is insolvent, the claimant will not be fully compensated for all the losses that he has suffered as a result of the two accidents. But that is not a reason for making each defendant liable for the total loss. In Baker, the issue was whether the tortfeasor who had caused the first injury was liable for its consequences after they had arguably become merged in the consequences of the second injury. In the present case, the question is whether the second tortfeasor is responsible for the consequences of the first injury. To that question, the answer can only be: no. It is true that, but for the first accident, the second accident would have caused the same damage as the first accident. But that is irrelevant. Since the claimant had already suffered that damage, the second defendant did not cause it. This is not a case of concurrent tortfeasors.

The costs issue

71. Proceedings against the first defendant were started by the issue of a claim form on 14 September 1999. Proceedings against the second defendant were commenced by the issue of a claim form on 12 March 2002. On 29 July 2002, an order was made consolidating the two claims. On 9 May 2003, the first defendant's solicitors wrote to the second defendant's solicitors:

"To dispose of all issues as between defendants/claimant, our clients offer to mediate the issues in this case. We have no trial date yet but the trial window envisages a trial within the first three weeks of July. Mediation should take place as soon as possible and certainly by early June.

This letter is being copied to the claimant's solicitors on an equal invitation.

We have also taken the step of asking the ADR Group to liase with all parties with a view to proposing potentially suitable mediators and agreeing a date for mediation if the parties agree mediation. Please let us know immediately if you have any objection to seeking a mediator through the ADR Group.

In the event that any party refuses this offer of mediation, this letter will be drawn to the attention of the trial judge and you will note that this is an open offer to mediate the case. We shall ask the judge to make a ruling as to costs in the event that any party refuses a mediation. We believe that this case is eminently capable of being resolved by mediation".

72. This offer was repeated to the claimant. By letter dated 20 May, the first defendant's solicitors replied:

"After giving very careful thought to the proposal, our insurance principals have come to the conclusion that this would not be an appropriate case for mediation. The issue between the defendants is one of law and therefore is requiring of a decision of the court. In such circumstances we cannot see there is any benefit to either side in mediation.

So as to make our position perfectly clear, our insurance principals would have no objection to mediation taking place between yourself and the claimant with a view to resolving the claimant's claim".

73. By their letter dated 27 May, the first defendant's solicitors responded, stating that they did not accept that the dispute was incapable of resolution, although it related to a point of law. They added:

"If your clients persist in refusing to mediate this case, then we will have no option but to place this and our earlier correspondence before the court. When the court comes to consider the question of costs and conduct in particular in refusing mediation."

74. The second defendant's solicitors replied on 28 May, saying: "We are not prepared to compromise on the point of law and therefore mediation would be pointless".
75. Having decided the causation issue in favour of the second defendant, the Recorder turned to the question of costs. He decided that costs should follow the event. His reasoning is set out in the following passage of his judgment:

"Well I am going to come to the same final view in relation to this case as it now arises before me. The possibility of alternative dispute resolution was raised fairly late in the day here as far as the chronology of this case was concerned. That is not to say that simply because it is raised late on means that it is of no consequence, far in fact from it, but it is a factor in the equation. But here, where the issue that has arisen is the one that I identified earlier today, which Mr Elgot goes as far as to describe as quite exceptional, using the language that arises in that case of Hurst v Leeming, it seems to me that alternative dispute resolution would have been likely to achieve very little. I do accept Mr Foster's general point, although it is not of application I think in the facts of this particular case, that alternative dispute resolution brings to bear a different set of "tools" than formal litigation in open court such as this, and who knows what alternative dispute resolution can achieve in some cases. However here in this case, when Mr Elgot poses for me the question what could alternative dispute resolution really have achieved at this particular late stage of the litigation as far as this particular topic was concerned, I must say I wonder really what could have been achieved."

76. In considering whether the first defendant has shown that the second defendant acted unreasonably in refusing the offer of mediation, a number of points need to be borne in mind. The full value of the claim was agreed by the parties at £195,000. The second defendant made a CPR Part 36 offer of £3,500 before proceedings started on the footing that this was a generous assessment of the value of the 3-6 months' exacerbation of the symptoms caused by the second accident. It can be seen, therefore, that almost £200,000 turned on the causation issue that was eventually tried by the Recorder.
77. As in Halsey, so in this case, the court did not make any order encouraging the use of ADR. This is not, therefore, a case where a party refused even to consider ADR despite a court order that it should do so.
78. This case raised the question whether Performance Cars is still good law, or whether it could be distinguished. Put like that, the claim against the second defendant, therefore, raised a question of law. In our judgment, the second defendant did not act unreasonably in saying that he (or more realistically his insurers) wanted to have that question resolved by the court. In these circumstances, the nature of the dispute was one which was towards the "intrinsically unsuitable" end of the spectrum. It is a far cry from a typical road traffic claim which raises no disputes of law, and where disputed facts are intrinsically suitable for resolution by ADR.
79. The second defendant reasonably believed that the claim against him had no merit. We hope that we have not done injustice to Mr Foster's submissions, but like the Recorder we think that the answer to the question raised by the causation issue was plain. It follows that the first defendant has not shown that there was a reasonable prospect that a mediation would have succeeded. The second defendant had decided to take a stand on the point of law. In doing so, he was not acting unreasonably.
80. There are yet further factors which the second defendant was entitled to pray in aid in support of the reasonableness of his refusal to agree to mediation. First, the costs of the mediation would have been excessive in comparison with the costs of litigating the issue at trial. The issue was disposed of by the Recorder in about two hours. Mr Christopher Purchas QC has suggested that the total costs of a mediation involving all three parties would have been of the order of £20,000. That figure (which was not investigated before us) does look surprisingly high. But on any view this is a case where a mediation would have been unlikely to be successful, and would probably have been relatively expensive when compared with the cost of a trial which, in the event, lasted about 2 hours. Secondly, as the Recorder pointed out, the offer of mediation came comparatively late in the litigation after substantial costs had already been incurred.
81. Taking all these factors together, we are in no doubt that the first defendant has not proved that the second defendant acted unreasonably in refusing to agree to mediation in this case.
82. It follows that the first defendant's appeal is dismissed on both issues.
Order: Appeal dismissed with costs subject to detailed assessment. (Order does not form part of the approved judgment)