England and Wales Court of Appeal (Civil Division) Decisions

IN THE HIGH COURT OF JUSTICE
COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM QUEEN'S BENCH DIVISION
ANDREW EDIS QC
[2010] EWHC 2061 (QB) HQ08X01154

		Royal Courts of Justice Strand, London, WC2A 2LL
		29/06/2011

President of the Queen's Bench Division

Introduction

1. Awards of costs in litigation are peculiarly fact sensitive, and this court will not disturb a trial judge's costs decision unless it was wrong in principle or otherwise plainly erroneous. In the present appeal from the costs decision of 5^th August 2010 of Andrew Edis QC, sitting as a deputy High Court Judge, the claimant had failed entirely on the only main issue in the action and had recovered a tiny fraction only of the amount he claimed on a part of the claim which had scarcely featured as an issue at all; yet the deputy judge awarded him 50% of his costs of the entire action. The appellant defendants maintain that this was manifestly unjust and wrong, and that they should have recovered all or a substantial proportion of their costs because the claim for practical purposes failed.
2. The deputy judge's costs judgment may be found at [2010] EWHC 2061 (QB) and may be referred to for details which this judgment need not contain. The deputy judge's liability judgment of 22^nd July 2010, which is not appealed, may be found at [2010] EWHC 1888 (QB).

Facts

3. The respondent, Sebastian Marcus, was a young man aged 31 in the first half of 2005, when he contracted a very unusual condition in his left lower leg which resulted in its amputation below the knee on 20^th June 2005. This became necessary because the arteries in his left lower leg became blocked with embolisms deriving from elsewhere. This caused very painful ischaemia in the foot, which resulted in the death of the tissue. The claim against the first defendant essentially was that Dr Ruth Thom, a General Practitioner employed by the first defendants, whom the claimant consulted on 6^th April 2005, diagnosed ischaemia but negligently failed to take appropriate steps to see that it was treated; and that the second defendant, a locum General Practitioner, negligently failed to diagnose ischaemia on any of the occasions on 14^th, 21^st and 28^th April 2005 when he saw the respondent. When the respondent was admitted to Medway Maritime Hospital on 12^th May 2005, it was too late to save his leg. His case was that, if each of the doctors had not been negligent, timely appropriate treatment would have saved his leg. He quantified his claim at £731,255 plus general damages. The appropriate quantum of the claim was agreed at £525,000 shortly before the liability trial. In short, the claim was all about the defendants' alleged responsibility for the leg amputation.
4. The second defendant admitted breach of duty in his defence, which was the first practical opportunity of doing so, there having been no pre-action protocol letter. This was not the respondent's fault, since much time had been spent in obtaining medical notes and tracing the second defendant, so that the limitation period was about to expire. However that may be, neither defendant had an earlier opportunity to admit breach or to take appropriate steps to protect themselves from an adverse costs award. The second defendant's defence denied that the breach of duty had caused the respondent's loss. It was specifically pleaded that, if the respondent had been referred for thrombectomy or embolectomy on 14^th April 2005, this would inevitably have failed because the arterial blockage had been in place too long to avoid amputation.
5. The first defendants' defence denied breach of duty, contending that there was no sufficient indication compelling a diagnosis of critical limb ischaemia on 6^th April 2005 and asserting that Dr Thom had on examination felt a weak left malleolus pulse. In due course a witness statement of Dr Thom was exchanged which contained her assertion that she had felt this weak pulse on that day. But about 2 days before the liability trial and shortly after a meeting of vascular experts, the first defendants changed their position by admitting breach of duty. Dr Thom was not called to give evidence, but her witness statement was available for the respondent to use as hearsay evidence.
6. Mr Stephen Brearley, consultant general and vascular surgeon, was the second defendant's vascular expert. It was he who expressed the opinion that, even if the respondent had been referred to hospital after first seeing his General Practitioner with foot pain on 24^th March 2005, he would already have had blocked arteries for three weeks and neither thrombolysis nor embolectomy would have been at all likely to succeed. Even if any of the doctors who saw the respondent before his admission to hospital negligently failed to recognise that his symptoms were the result of blocked arteries, the blockages were so well established when they saw him, and so far distal, that the situation was irremediable and below knee amputation of his leg was already inevitable. Mr Brearley noted that Dr Thom, when she saw the respondent on 6^th April 2005, believed she could feel a week posterior tibial pulse (at the ankle), but the claimant simply could not have had the symptoms he did have at that stage (and had had for over a month previously), if he had had patent arteries down to the ankle level. Mr Brearley wrote that it is well known that practitioners who are not vascular specialists quite frequently believe that they can feel peripheral pulses when clinical and arteriographic evidence shows that those pulses could not be present. Mr Brearley referred to a paper "Peripheral pulse palpation: an unreliable physical sign" which he, with others, published in the Annals of the Royal College of Surgeons of England 1992 Vol. 74 page 167.
7. The other two vascular surgery experts were Mr Jack Collin for the respondent, and Professor Charles McCollum for the first defendants. The three of them made a joint statement. This records that Professor McCollum agreed with Mr Brearley that the weak pulse which Dr Thom recorded on 6^th April 2005 was unreliable. Mr Collin believed that there was no reason to doubt the observation of Dr Thom that there was a weak malleolus pulse. On the critical causation issue, the statement records that Professor McCollum and Mr Brearley were of the opinion that neither thrombolysis nor thrombectomy/embolectomy would have had any chance of success unless offered within 2-3 weeks of the onset of symptoms and that even then the likelihood of success would have been low. They agreed that the small arteries in the claimant's left foot were occluded by mid-March 2005 and that there was no prospect that any available treatment could have restored perfusion to this foot such that amputation could have been avoided on 6^th April 2005 or any later date. Mr Collin considered that occlusion of the popliteal artery probably occurred about three weeks before angiography on 13^th May 2005 and the treatment to restore perfusion would have been successful if it had been given in early May 2005. This was the essence of the contest on causation which the judge, accepting the opinions of Mr Brearley and Professor McCollum and rejecting that of Mr Collin, decided against the respondent. This being the central and eventually only issue in the case, the main claim failed entirely.
8. There had, however, been breaches of duty for which, as an afterthought, the claimant's counsel claimed modest damage for the additional time during which the respondent had suffered pain by reason of the ischaemia and before the amputation. There was a pleaded general unspecific claim for pain and suffering. The deputy judge ruled that this covered this small claim, and there is no appeal against that ruling. There was a written question to the experts capable of covering the point, which they had not answered in writing. We are told that evidence covered the matter in passing and that the claimant's counsel first raised the matter in his closing address. It was scarcely part of the claim as conducted, but the deputy judge found that it succeeded and awarded the respondent £2,000 damages. Thus the respondent's main (and essentially only) claim, on which virtually all the costs had been spent, failed and the respondent recovered about 0.25% of the amount of his claim. I regard it as forensic hyperbole to suggest, as Mr Mansfield QC did, that this recovery nevertheless constituted vindication for the claimant. The defendants had admitted breaches of duty – one later than the other. But that and £2,000 would have been scant consolation for a respondent whose only real claim was for the amputation of his lower left leg.

Rules about costs

9. Orders for costs should be made in accordance with the relevant provisions of the Civil Procedure Rules. These relevantly include that the court has a discretion as to whether costs are payable by one party to another and as to the amount of those costs – rule 44.3(1). If the court decides to make an order about costs, the general rule is that the unsuccessful party will be ordered to pay the costs of the successful party, but the court may make a different order – rule 44.3(2). It is thus necessary for the court to determine which party is the successful party. The court must have regard to all the circumstances, including the conduct of all the parties, whether a party has succeeded on part of his case and any payment into court or admissible offer to settle which is not an offer to which costs consequences under Part 36 apply – rule 44.3(4). The conduct of the parties includes conduct before, as well as during, the proceedings and the extent to which the parties followed relevant pre-action protocols; whether it was reasonable for a party to pursue a particular allegation or issue; the way in which a party has pursued or defended the case; and whether a successful claimant exaggerated the claim – rule 44.3(5). When costs are assessed on a standard basis, the question of proportionality is relevant, as is the question whether costs were reasonably incurred or reasonable and proportionate – rule 44.4, but see Lownds v Home Office [2002] 1 WLR 2450. Rule 44.5 provides for matters relevant to reasonableness and proportionality.
10. The effect of a Part 36 offer, if it is accepted, is that the party making the offer has to pay the other party's costs to the date when the notice of acceptance was served – rule 36.10. If in the present case the second defendant, for instance, had made a Part 36 offer to settle the claimant's claim for £2,000 at the first reasonable opportunity, that is when the Claim Form was served, and if that offer had been accepted, we are told that that would have entitled the claimant to claim some £67,000 of costs plus any CFA uplift and ATE insurance premium. The total of this is likely to have been in excess of £100,000, which would be palpably disproportionate to a recovery of £2,000. Neither defendant had made a Part 36 offer, but it would be quixotic if the failure to do so in the circumstances were to have any real bearing on the eventual award of costs.

Authorities

11. In Widlake v BAA Limited [2009] EWCA Civ 1256, the claimant had claimed for personal injury resulting in a back injury for which she eventually claimed £23,906.40, having started the proceedings with a claim valued at nearly £150,000. She recovered £5,522.38 plus some interest, but she was ordered to pay the defendant's costs. The defendant had paid £4,500 into court. The judge regarded the real winner of the trial as the defendants, mainly because the claimant had deliberately concealed a previous history of back trouble, and had misled her own expert and abused the court's process in a cynical and dishonest fashion. The judge was influenced by Mulloy v Shell UK Ltd [2001] EWCA Civ 1272 and Painting v University of Oxford [2005] EWCA Civ 161, both of which the Court of Appeal considered. Reference was also made to Jackson v Ministry of Defence [2006] EWCA Civ 46 and to Straker v Tudor Rose [2007] EWCA Civ 368, where claimants who had exaggerated their claims nevertheless recovered substantial damages and beat payments into court. Ward LJ then said at paragraphs 35 and 36:

"I agree with Waller LJ in Straker that one must go to the rules and apply the approach which the rules prescribe.

Thus the first question is to determine which was the unsuccessful party. There may appear to be some difference in the approach of this Court if one compares Painting with Jackson or Hall and I hope I can reconcile those differences. I prefer the approach of Tuckey, Keene and Wilson LJJ in Jackson that the claimant was successful in the sense that he had established a claim for damages and beaten the payment into court. Although it was a case set in a commercial context, Waller LJ was surely right in Straker to endorse Longmore LJ's views that the most important thing is to identify the party who is to pay money to the other even in a case of personal injury. The claimant had to come to court to establish her claim, a genuine claim, because she had suffered an injury through the admitted negligence of the defendant. The judgment in her favour is a vindication of her stance."

Ward LJ's conclusion was:

"Part 36 now also affects a claimant. Whilst not obliged to make a counter-offer, in this day and age of encouraging settlement, claimants who do not do so run the risk that their refusal will impact upon the costs they may otherwise be entitled to recover. Here there was no attempt to negotiate and that counts against the claimant.

Having tried to represent these considerations in a balance sheet, where does the balance lie? I start with the claimant getting her costs because she beat the payment in and was the successful party. That is the starting point. Those costs should not include costs related to Miss Porter's reporting and the costs judge must be directed to exclude those matters. Pursuing her claim in the exaggerated way she did had the result that this became heavily contested litigation whereas it might have settled. The defendant has been put to unnecessary expense. But an order for costs against the claimant is less justified where, as here, the defendant failed to alleviate its predicament by making a proper Part 36 offer and so lost the opportunity provided by the rules of recovering those costs from the claimant. The claimant's dishonesty must be penalised. The claimant's failure to negotiate a claim which was clearly capable of being settled must also be recognised. When I balance those factors, and attempt to do justice to both parties and to be fair to them, I conclude that the right order in this case is that there be no order for costs."

12. Oksuzoglu v Kay [1998] 2 All ER 361 is, with some qualification, barely distinguishable from the present appeal. One qualification is that it was decided just before the Civil Procedure Rules came into force, so that the former Rules of the Supreme Court applied. I do not consider that that is of great significance, although there are some differences. The plaintiff's leg had to be amputated because of a malignant tumour. Liability was tried as a preliminary issue. The judge found that both medical practitioners were negligent in not referring the plaintiff to hospital earlier, but that amputation would have been inevitable in any event. He gave judgment against both defendants for damages limited to pain, discomfort and distress. After the liability trial, the judge gave the plaintiff leave to amend the claim to include allegations of psychiatric damage. The plaintiff was awarded £3,000 for pain and suffering and about £2,000 for his mother's care and incidental laundry expenses. The judge nevertheless ordered the defendants to pay all the plaintiff's costs of the action because the defendants had failed to protect themselves by making a written offer accepting a specified proportion of liability. This court held that the provision enabling a defendant to accept a proportion of liability did not apply, and that the judge had been wrong to lump the trials of liability/causation and quantum together. The court ordered that the defendant should pay the plaintiff the costs of the action except that the defendants should recover 90% of their costs of the trial of issues of liability and causation – see paragraph 66 of the judgment – the head note is wrongly expressed in this respect.
13. Mr Mansfield, who helpfully carried out a detailed analysis of the procedural position in Oksuzoglu, seeks to distinguish it on the footing that it concerned a split trial where the defendants succeeded on the preliminary issue and recovered the costs of that issue, but the plaintiff, he says, recovered the costs of the action and the quantum hearing. That is correct so far as it goes. But it does not go far enough, because in the present case the trial was a single trial for liability, causation and quantum, although in the event the quantum did not have to be judicially decided, apart from the £2,000. In awarding the defendants in Oksuzoglu 90% of their costs of the liability and causation issue, the court ignored a payment into court and treated the defendants as the winners of the liability and causation issue. The second defendant did not write a Calderbank letter and the first defendant was not willing to accept the possibility of a finding of negligence until a later date. But so long as the plaintiff was asserting that the defendants' negligence caused the need to amputate the leg, any such assertion was bound to be resisted and the plaintiff lost the major issue. The decision in Oksuzoglu does not indicate that the respondent in the present appeal should be paid his costs of the action, because it was the action (not a preliminary issue) which the deputy judge decided and which had given rise to all the costs. There was no other part of the present case to make a costs order about.
14. Brooke LJ gave the leading judgment in Oksuzoglu, with which Millett and Hirst LJJ agreed. Brooke LJ said at paragraph 55 that, if one put aside both a payment into court and a Calderbank letter, there was still no doubt but that the defendants essentially won the trial on liability and causation. All the plaintiff got out of it was a decision which reduced the potential value of his claim, as then put forward, by 99%, from about £300,000, inclusive of interest, to about £3,000. Brooke LJ then considered a number of authorities including Re Elgindata (No 2) [1992] 1 WLR 1207, which he said should be applied to situations such as that before the court with appropriate caution, and Beoco Ltd v Alfa Laval [1995] QB 137. He then said at paragraph 58:

"In this line of cases, where the plaintiff only recovers between 1% and 3% of his original claim (sometimes, but not always, after a late amendment) the court is entitled to ask itself: 'Who was essentially the winning party?' It will not be distracted from making a just order as to costs by the absence of a payment into court which the plaintiff obviously would not have accepted (see Alltrans Express Ltd v CVA Holdings Ltd [1984] 1 All ER 685 at 692, 693 [1984] 1 WLR 394 at 403, 404 per Stephenson and Griffiths LJJ), or where the defendants did not have a proper opportunity to make a payment into court which obviously would not have been accepted (see Beoco Ltd v Alfa Laval Co Ltd [1994] 4 All ER 464 at 479-480. [1995] QB 137 at 156). Although all these cases are different, in the present case the substantive lis between the parties on the trial of the preliminary issues related to the big claim on which the plaintiff wholly failed."

The Deputy Judge's decision

15. In the present case, the deputy judge declined to follow Oksuzoglu as if it bound him to award the appellants their costs, although he took into account the fact that the respondent failed on causation and that this was a substantial cause of the very high costs and the reason why the respondent recovered so little. He considered that the court's approach in Oksuzoglu may have to an extent anticipated the coming into force of the Civil Procedure Rules. But it was established that decisions under the Rules of the Supreme Court were of little value in the application of the Civil Procedure Rules. He attached particular significance to Rule 44.3(4)(a) and (5)(d), which require the court to take account of the conduct of the parties and specifically mention the case where the claimant succeeds in his claim but has exaggerated it. These provisions were not present in the Rules of the Supreme Court, although I note that neither was of particular significance in the present case. The position under the Civil Procedure Rules was fundamentally different, and the deputy judge considered that it would be wrong for him to approach the case on the basis that he was bound by authority to find that a claimant who has recovered £2,000 when the agreed value of the claim was £525,000 is the unsuccessful party. I note that, although of course the deputy judge was strictly correct that he was not bound by authority, the provisions of the Civil Procedure Rules as to awards of costs, although they are different in some detail, are not fundamentally different from the practice under the Rules of the Supreme Court in the years shortly before the Civil Procedure Rules were brought into force. The deputy judge also noted that the Court of Appeal in Oksuzoglu did not order the plaintiff to pay the defendants' costs of the action, but only of the preliminary issue. As I have indicated, I consider this to be a distinction without a real difference in the present case. The order in Oksuzoglu essentially related to that which had been decided at the trial.
16. The deputy judge concluded that in the present case the respondent was the successful party. He did so, saying that in a commercial context the most important thing was to identify the party who is to pay money to the other, and that in Widlake that had been applied in a personal injury case. The judgment had been a vindication of the respondent's stance, especially when the first defendant had not admitted negligence until 2 days before trial and both defendants denied that any loss at all was caused. The damages awarded were not nominal and the respondent was the successful party in the proceedings. Accordingly the respondent should be regarded as the successful party for the purposes of a costs decision. The respondent had not exaggerated his claim. He had lost on causation, but had reasonably proceeded on the basis of expert opinion which supported his causation claim. The first defendants did not admit breach until very late in the day, and the defendants had agreed that the court should make a single costs decision in relation to them. Until the first defendants admitted breach, they were contending that Dr Thom had felt a pulse on 6^th April 2005. The deputy judge accorded this modest weight. The appellants had succeeded on the most important issue in the case. The damages awarded were not nominal, but were very small. No rational person would issue proceedings in a case of this kind if the recovery was only £2,000. No rational person would contest such proceedings if they were issued. However, neither defendant made an admissible offer to settle the respondent's claim, nor any Part 36 offer. This was surprising and it meant that the appellants could not sensibly complain that an order for costs against them was unjust. The claim which succeeded had not been specifically pleaded, but there was a claim for pain. The deputy judge saw that a late Part 36 offer with reference to the claim which succeeded would, if it had been accepted, have produced an unfair result. The appellants should have had the possibility of a fall back claim in mind at an early stage. Oksuzoglu exemplified such a situation. The award of 50% of the costs against the appellants was not unjust because it was open to them to make a Part 36 offer.

Discussion

17. In my judgment, the deputy judge was wrong in principle to conclude that the respondent was the successful party. The award of £2,000 was insignificant in the context of the claim and the action as a whole, and, although it was technically within the pleaded claim, it was in truth a last minute addition to salvage something (0.25%) from an action which the respondent lost. The whole action was about the cause of the need for the respondent to undergo a leg amputation, and, for all that the first defendants did not admit breach until a late stage, the second defendant's early admission would have carried the entire claim, if the respondent had succeeded on causation. The causation issue was squarely advanced in the original defences. Mr Brearley's opinion on the causation issue was the second defendant's case from the start and it carried the day. I have already indicated my view that such vindication as the action achieved was scant consolation for a claimant whose £525,000 claim had failed entirely. This is not a case in which identification of the party who has to write the eventual (very small) cheque is persuasive as to the costs order. There is little, if any, material distinction between this case and Oksuzoglu. The essential relevant considerations which applied in 1998 have not materially changed with the advent of the Civil Procedure Rules. The fact that the costs order in Oksuzoglu was for the costs of the liability and causation issue does not directly apply to the present case, because the deputy judge decided the action, not a preliminary issue, and the costs of the action in truth related to the causation issue. The fact that the appellants did not make a Part 36 offer or write a Calderbank letter is no more relevant to the costs issue than it was in Oksuzoglu. If the defendants had made a Part 36 offer of (say) £3,000 at the outset, that would have carried a costs payment of some £100,000 which would have been disproportionate and unjust. More importantly, however, an award of £2,000 on an afterthought claim for a short period of extra pain is insignificant in the context of the action as a whole and the nominal failure to make a Part 36 offer is of no consequence and a technical triviality. As the deputy judge himself said, no rational person would issue or defend proceedings such as these, if the recovery was only £2,000. The subject matter of the £2,000 was simply not what the action was about, and, other than technically, that claim was not advanced until the respondent's closing submissions. The action was about the cause of the leg amputation and the costs were spent in advancing and defending that. I accept that the damages awarded were not technically nominal, but they were so small as to be insignificant.
18. Mr Mansfield, in addition to urging upon us points which we have already addressed, submits that, as in some other cases, the respondent could not have got his damages without going to court, and no offer was made at any time. The appellants had to write the cheque and the first defendant contested breach until a very late stage when the case that Dr Thom had felt a pulse on 6^th April 2005 was withdrawn. Mr Mansfield says that the appellants are trying to achieve a more favourable result than if they had made an offer. He stresses that there is no criticism of the conduct of the claimant. He also stresses that in Oksuzoglu the defendant did not get the costs of the action.
19. Although some of these points are persuasive to a result that the respondent should not be ordered to pay all the appellants' costs, they are not persuasive on the question of who was the successful party. I am in no doubt but that it was the appellants, and the starting point should therefore be a costs order in their favour. The question is whether it should be reduced and, if so, by what proportion.
20. Mr Hutton, on behalf of the appellants did not argue vigorously for a full unreduced costs order in favour of his clients. In my view, there should be a reduction for (a) the fact that the respondent did succeed to a very small extent; (b) the fact that the first defendants did not concede liability until a very late stage; and (c) the more significant point that the case that Dr Thom felt a weak pulse on 6^th April 2005 (which, given a finding of liability, would predicate that the respondent's causation case might well have succeeded) was not withdrawn until just before the trial. I do not consider that there should be a reduction because there was no Part 36 offer, but I do take into account that the appellants might have written a Calderbank letter offering £3,000 plus costs proportionate to that recovery. It remains the fact, however, that the real claim failed and no rational person would have issued the proceedings which were issued to recover only £2,000.
21. In these circumstances, I consider that the appellants should have recovered 75% of their costs from the respondent and that the appeal should succeed, and that an order to that effect should be substituted for that made by the deputy judge.

Lord Justice Jackson:

22. I gratefully adopt the summary of the facts set out in the judgment of the President of the Queen's Bench Division, but regret that I cannot agree with his conclusions.
23. This is a case in which there always was, on the pleadings, a fall back claim for pain and suffering consequent on the late diagnosis of the claimant's ischaemia. It was always likely that, if the claimant succeeded on liability but failed to establish that the defendants' breach of duty caused the loss of his leg, the claimant would still recover a modest amount of general damages on his fall back case. This is precisely what happened in Oksuzoglu v Kay [1998] 2 All ER 361, a case very similar to the present case on its facts. Both parties in the present case were represented by experienced lawyers specialising in clinical negligence and must have been well aware of the reported decision in Oksuzoglu.
24. In those circumstances the obvious way in which the defendants could and should have protected their position in respect of the likelihood of the claimant's success on his fall back case was by making a Part 36 offer of modest damages, perhaps £2,000 or £3,000.
25. Mr Hutton argues that it would not be practicable to make a small Part 36 offer because of the drastic costs consequences if such offer were accepted. I do not agree. If the defendants had made and the claimant had accepted an offer of, say, £3,000 then the claimant would have been entitled to recover costs assessed on the standard basis. Costs assessed on the standard basis are no more than is reasonable and proportionate: see rules 44.4(1) and (2). Mr Hutton argues that rules 44.4(1) and (2) have been interpreted by the courts in such a way that the costs awarded would in fact be disproportionate. Mr Hutton draws attention to the decision of the Court of Appeal in Home Office v Lownds [2002] EWCA Civ 365; [2002] 1 WLR 2450. Mr Hutton may be right in his implicit criticism of the Lownds decision, but that is a matter for law reformers not for this court. We must proceed on the assumption that any assessment of costs made pursuant to rules 44.4(1) and (2) on the standard basis would result in a sum which is no more than is reasonable and proportionate.
26. If the defendants had offered £3,000 at the outset and that offer had been refused, then the defendants would have been entitled at the end of this litigation to recover all of their costs incurred since the date of the Part 36 offer. That did not happen.
27. It was suggested in argument that the defendants could have obtained costs protection at an earlier stage by making a Calderbank offer along the following lines: the defendants offer to pay £3,000 damages plus a modest sum (perhaps £5,000) in respect of costs. In my view such an offer would give the defendants no effective protection. Unless the defendants also offered to pay the claimant's costs to date assessed on the standard basis, such an offer would for all practical purposes be of no benefit to the claimant.
28. The blunt fact is that the claimant had a good claim for £2,000 and the defendants were refusing to pay anything. The only way the claimant could recover the £,2000 due to him was by issuing proceedings and pressing on until the defendants agreed or were compelled to pay (a) £2,000 damages and (b) costs assessed on the standard basis.
29. The morass of case law which has developed concerning the effects of Part 36 offers and the effects of offers outside Part 36 reveals an unwelcome and unnecessary degree of uncertainty. This is an area of law where all parties need to know where they stand and understand the costs consequences of their actions.
30. In my view, in a personal injury case where (a) the claimant has pursued his claim in a reasonable manner, (b) the claimant recovers damages (other than nominal damages) and (c) there is no or no sufficient Part 36 offer, the starting point should be that the claimant recovers his costs. That flows from rule 44.3(2)(a). The next question to consider is whether any adjustment should be made to reflect the issues on which the claimant has lost. In a typical personal injury or clinical negligence case where the claimant is successful, the claimant will win on some points and lose on other points along the way. That is not normally a ground for reducing costs recovery: see the judgment of the Court of Appeal presided over by Sir Anthony Clarke M.R. in Goodwin v Bennetts UK Ltd. [2008] EWCA Civ 1658. If, however, the claimant has lost on major issues which generated significant costs, then the court will exercise its discretion under rule 44.3 to reduce the costs recovery.
31. I have carefully considered whether the approach set out in the preceding paragraph is inconsistent with any of the authorities relied upon by the defendants in this appeal. I do not think that it is.
32. Painting v University of Oxford [2005] EWCA Civ 161 was a personal injuries action in which liability was admitted, subject to a 20% deduction for contributory negligence. The case was fought out on quantum. The defendant succeeded in its defence that the claimant was exaggerating her claim, but the claimant recovered more than the payment into court. The Court of Appeal held that the claimant should pay the defendant's costs after the date of the payment into court. The adverse costs order in that case was justified by reason of the plaintiff's unreasonable litigation behaviour. That case is materially different from the paradigm discussed in the paragraph 30 above. It is also materially different from the present case, where the claimant has acted entirely reasonably; he only failed on his main claim because the deputy judge ("with considerable regret") preferred the expert evidence of the defendants on causation.
33. In Widlake v BAA Ltd [2009] EWCA Civ1256 the claimant recovered £5,522 damages for personal injuries, thus beating the payment in of £4,500. However, the claimant had deliberately concealed material facts and exaggerated her claim: see the judgment of Ward LJ at [18] and [44]. The Court of Appeal held that in the circumstances the proper order was no order for costs. Ward LJ (with whom Smith and Wilson LJJ agreed) said that although the claimant had only recovered a small part of her claim she should be regarded as the "successful" party, because she had beaten the payment into court: see [36]. In this regard he preferred the approach of the Court of Appeal in Jackson v Ministry of Defence [2006] EWCA Civ 46 to the approach in Painting. The only reason why Ms Widlake did not recover costs was because of her unreasonable conduct in the litigation. Again that is a material distinction from the paradigm discussed in paragraph 30 above and from the present case.
34. Oksuzuglu v Kay [1998] 2 All ER 361 is a case which was decided under the former Rules of the Supreme Court and is therefore not binding in relation to cases proceeding under the Civil Procedure Rules 1998, as amended. Having said that, I regard the case as materially different from the paradigm discussed in paragraph 30 above and from the present case. In particular, the defendant in Oksuzoglu paid £3,500 into court on 14^th November 1994 (i.e. before the liability trial) which was more than her claim as then pleaded was worth. The claimant ultimately recovered £3,000 general damages plus an agreed sum of £1,720 in respect of psychotherapy and travel costs, but she only recovered these sums because of amendments made after the liability trial. The costs order ultimately made reflects the particular and unusual circumstances of that case.
35. In the present case the judge adopted the course set out in paragraph 30 above. He correctly noted that the claimant had succeeded on a fall back element of his claim. That element of the case had first received attention in the days leading up to trial: see question 20 of the questions formulated by the lawyers for consideration at the experts' meeting. It was in this period that the claimant's primary case on causation was materially weakened by the first defendant's change of position and abandonment of reliance on the evidence of Dr Thom. Thereafter the claimant pursued his fall back case by questions put to the experts in their oral evidence and by argument in counsel's closing speech. At the end of the day the claimant failed on his primary case but recovered £2,000 on his fall back case. There was no Part 36 offer. In my view the judge was correct to take as his starting point that the claimant was the successful party. Therefore the claimant should recover the costs of the action, subject to adjustment to reflect the very substantial issue on which the claimant had lost.
36. In determining what adjustment to make to the costs order in order, to reflect the claimant's defeat on a very substantial issue in the case the judge was required to weigh up a number of conflicting factors. In doing so the judge was exercising the discretion conferred upon him by rule 44.3 of the Civil Procedure Rules. One material factor which the judge took into account was that the claimant's case on causation was substantially stronger right up until the eve of trial by reason of Dr Thom's evidence that she felt a pulse on the 6^th April 2005. Shortly before trial the first defendant changed its position and no longer relied upon the evidence of Dr Thom. That change of position by the first defendant undermined the claimant's case on causation. By then of course a great deal of costs had already been incurred on both sides and there was still no Part 36 offer on the table.
37. The judge decided to reflect the issues on which the claimant had lost by making a 50% reduction in the recoverable costs. That assessment was undoubtedly a generous one to the claimant. However, the judge is best placed to make an assessment of this nature. The trial judge has not only heard all of the evidence but also is familiar with the detailed course of the pre-trial proceedings. This court should only interfere with the judge's exercise of discretion under rule 44.3 when the trial judge has fallen into an error of law. I am unable to detect any error of law in the present judgment. The judge has considered the various conflicting factors. The judge took into account that "the claimant has established in the proceedings that the doctors who saw him in April 2005 were negligent and that by being so they caused him severe pain. This is not a case where he recovered nominal damages." The weight which the judge attached to the various factors as set out in his judgment was a matter for him. It does not seem to me that the judge took into account any irrelevant factors. The judge did not fail to take into account relevant factors. The judge did not reach a decision which was outside the generous ambit of his discretion, even though it was a decision more favourable to the claimant than I would have reached.
38. One way of testing the above analysis is to consider the following scenario. Suppose that, when first presented with the claim, the defendants had appreciated (as they should have done following Oksuzoglu) the likelihood that even if they won on the main issue the claimant would still recover modest damages in respect of pain and suffering during the period of delayed diagnosis. Suppose that the defendants had nevertheless under-estimated general damages and had made a Part 36 offer of only £1,200. The claimant would ultimately have beaten that offer, even though he failed on the main claim. In my view the normal Part 36 consequences would have followed. The claimant would have recovered the costs of the action, less a discount to reflect the substantial issue on which he lost. The Court of Appeal's decision in Carver v BAA [2008] EWCA Civ 412 would have been distinguished. The impact of Carver has been greatly reduced by the Court of Appeal's subsequent decision in Gibbon v Manchester City Council [2010] EWCA Civ 726. Indeed (although not relevant to the present appeal) anyone who attended the open meeting of the Civil Procedure Rule Committee on 13^th May 2011 will be aware that the effect of Carver is likely soon to be reversed by an amendment to the Civil Procedure Rules.
39. Since the defendants have not made any Part 36 offer at all, their costs position should not be any better now than it would have been if they had made an inadequate Part 36 offer.
40. There are underlying the issues in this appeal some important policy considerations. The defendants are understandably aggrieved by the substantial liability which they face in the event of an order of costs against them in a case where the claimant has a conditional fee agreement with after-the-event insurance. I understand the defendants' concern in this regard, but those are matters for the consideration of law reformers not for the consideration of this court. There are, however, other policy considerations which this court can properly take into account. Part 36 (amongst its other purposes) affords protection for defendants who have a weak case on liability but a strong case on quantum. Such defendants can and should protect their position on costs by making an appropriate Part 36 offer at an early stage. If only defendants and their insurers would take this course, a large amount of unnecessary litigation would be avoided. I accept that in this case an appropriate Part 36 offer (of £2,000 or £3,000) would probably not have been accepted. Nevertheless it seems to me quite wrong that defendants, who do not avail themselves of the Part 36 machinery and who press on in the hope of escaping liability altogether, should nevertheless escape the ordinary costs consequences because the claimant has only recovered a small sum in damages.
41. There is an acute need for clarity and certainty in the field of Part 36. Parties need to understand (a) the consequences of making or not making Part 36 offers and (b) the consequences of accepting or not accepting such offers. Absent unreasonable litigation conduct or some similar factor, the starting point must be that a claimant who does better than any offer made by the other side recovers costs. In appropriate cases an adjustment can then be made for issues on which the claimant has lost.
42. The defendants made no Part 36 offer in this case and in my view they should accept the consequences. I would dismiss this appeal.

Lord Justice Tomlinson :

43. I agree with the President that for the reasons which he gives this appeal should be allowed and an order substituted awarding to the Appellants 75% of their costs of the action.
44. Decisions about costs are in my experience often amongst the most difficult which can confront a trial judge. This experience is borne out by the fact that this court is itself divided so that the four judges who have so far had to consider the issue are evenly divided as to the proper order.
45. Whatever costs order is made, one cannot but be dismayed at the outcome of this litigation. We were told that on a conservative basis the claimant Respondent's costs of the action are likely to be in the region £480,000, inclusive of a success fee of 100% of base profit costs and an after the event insurance premium in a sum likely to be in excess of £30,000. We were not told of the level of the Appellants' costs, but they must evidently be substantial, not least given the involvement of expert witnesses instructed by both Appellants separately. I do not mean to suggest that any of the fees incurred have been unnecessary or unreasonable in amount – I am in no position to offer a view on that topic which does not arise for decision by this court. My point is the rather different one that the costs expended in this litigation comfortably exceed the amount, £525,000, which it was agreed would, in the judge's words "have made a real difference" to the Claimant's life. As it is, this expenditure has achieved no more than the decision that where compensation for the effects of clinical negligence is dependent upon proof of fault, this Claimant, who lost his lower leg at a tragically early age, is entitled to £2,000. Whether the costs of this exercise are borne, or principally borne, by the Claimant's insurers or by the Appellants' insurers, it cannot be said to be either an ideal or a cost-effective outcome.
46. The starting point of the enquiry is as all accept to identify the successful party. Although there have been many attempts to describe this surprisingly elusive process, no description is in my view better than that of Sir Thomas Bingham MR in Roache v Newsgroup Newspapers Ltd [1998] EMLR 161:-

"The judge must look closely at the facts of the particular case before him and ask: who, as a matter of substance and reality, has won? Has the plaintiff won anything of value which he could not have won without fighting the action through to a finish? Has the defendant substantially denied the plaintiff the prize which the plaintiff fought the action to win?"

47. Looked at as a matter of substance and reality, there is in my judgment no escape from the conclusion that the Claimant lost his action. £2,000 is of course a substantial sum, but it is not compensation for the loss of a lower leg and it is of no value to the Claimant in meeting the financial needs imposed upon him in consequence of the loss of his lower leg. The award of £2,000 was in truth irrelevant to the purpose of the action. Since, as the judge found, no rational person would either issue or defend proceedings such as these if the sum claimed had been £2,000, it is not in this case realistic to ask whether the Claimant could have recovered £2,000 without fighting the action through to a finish. An action to recover £2,000 would never have assumed this form and, whilst the matter has obviously not been put to the test, it seems unlikely that it would have been defended. What is however beyond argument is that the defendant Appellants substantially denied the claimant Respondent the prize which he fought the action to win.
48. I am in no doubt that the defendant Appellants should be regarded as the winners of this action. That view is reinforced by the circumstance that experienced and responsible Counsel found it necessary and appropriate to seek from the judge during closing submissions a ruling as to whether the claim for damages arising out of the failure to prescribe appropriate analgesia during the period prior to the amputation was open on the pleaded statements of case. The judge ruled that it was, and his decision has not been appealed, but to my mind it speaks volumes (a) that a ruling was sought and neither the application nor the opposition to it thought to be frivolous and (b) that the argument in this court has been conducted upon the basis that an appeal from that ruling could sensibly have been contemplated. I venture to suggest that had the pleader been consciously advancing a claim of this sort more explicit language would have been used. It is clear that no attention was paid to the possibility of recovery on this basis until the agenda was prepared for the meeting of the experts in vascular surgery which took place shortly before trial. Ironically, the relevant question, no. 20, which read "If the Claimant would not have avoided a left below knee amputation, do the experts agree that earlier diagnosis would have resulted in earlier surgery and/or pain control prior to the actual surgery performed on 19 June 2005?" seems not fully to have been understood. The answer given was:-

"SB and CM agree that earlier diagnosis of foot ischaemia would not have influenced the date of amputation as patients with severe ischaemia almost always delay amputation until such time as their symptoms become unbearable. JC refers to his response to question 19."

Self-evidently, the experts did not understand that they had been asked a question in relation to an alleged failure to prescribe appropriate analgesia during the period prior to the amputation. One or two questions were asked by the Claimant's counsel of two of the vascular surgery expert witnesses about whether earlier referral would have resulted in more effective analgesia, by reason of using opiate analgesia rather than the less effective analgesia prescribed by the second Defendant. It was on this basis that the Claimant succeeded. The judge allowed the claim under the general rubric of "avoidable pain and suffering" notwithstanding that the pleaded particulars of that pain and suffering are couched in terms of earlier admission to hospital leading to a successful thrombectomy/embolectomy and the avoidance of a below knee amputation. The successful claim is in my view properly characterised as an afterthought which happily for the Claimant was technically open on the pleadings. The judge himself described the award of £2,000 as "having an academic feel to it". In these circumstances the judge in my view reached a conclusion as to who was the successful party which was wrong in principle.

49. I too consider that the judge was wrong to attach weight to the circumstance that the Defendants had made no Part 36 offer in anticipation of this outcome. At no stage could the Defendants have made a Part 36 offer without incurring a liability in costs wholly disproportionate to the outcome. In this regard it is worth noting that CPR 36.10(2) provides:-

"Where –

(a) a Defendant's Part 36 offer relates to part only of the claim; and

(b) at the time of serving notice of acceptance within the relevant period the claimant abandons the balance of the claim,

the claimant will be entitled to the costs of the proceedings up to the date of serving notice of acceptance unless the court orders otherwise."

It is in my view implicit in this rule that in such a situation the starting point is that the costs recoverable will include those referable to the entire claim hitherto pursued but, by acceptance of an offer relating to part only of the claim, subsequently in part abandoned. So here the starting point would be that the Claimant would on acceptance of a Part 36 offer in the sum of, say, £3,000, be entitled to recover his costs to date. Whether those costs were proportionate would be a matter for the costs judge. But in Lownds v Home Office [2002] 1 WLR 2450 this court presided over by the architect of the CPR, Lord Woolf CJ, gave guidance as to the resolution of this question in a case where the Claimant recovered significantly less than he had claimed:-

"39. Turning to the specific points of principle raised by May LJ (paragraph 11 above) where a claimant recovers significantly less than he has claimed, the following approach should be followed. Whether the costs incurred were proportionate should be decided having regard to what it was reasonable for the party in question to believe might be recovered. Thus (i) the proportionality of the costs incurred by the claimant should be determined having regard to the sum that it was reasonable for him to believe that he might recover at the time he made his claim; (ii) the proportionality of the costs incurred by the defendant should be determined having regard to the sum that it was reasonable for him to believe that the claimant might recover, should his claim succeed. This is likely to be the amount that the claimant has claimed, for a defendant will normally be entitled to take a claim at its face value.

40. The rationale for this approach is that a claimant should be allowed to incur the cost necessary to pursue a reasonable claim but not allowed to recover costs increased or incurred by putting forward an exaggerated claim and a defendant should not be prejudiced if he assumes the claim which was made was one which was reasonable and incurs costs in contesting the claim on this assumption."

50. The claim here in respect of an avoidable amputation was a reasonable claim supported by respectable expert opinion which however ultimately failed. It is to my mind axiomatic that it is upon those who assume the risk of pursuing such a claim that the consequences in the shape of costs should be visited. However an assessment of costs consequent upon an accepted Part 36 offer made at any stage in this litigation would achieve a reversal of this obviously sensible and just outcome because the costs judge would be obliged to assess proportionality having regard to the substantial sum that it was reasonable for the Claimant to believe he might recover. It cannot in my judgment be appropriate to take into account the failure by a defendant to avail himself of a procedure which will lead to so unjust an outcome and the rules are not so inflexible so as to require us so to do.
51. As the President has already pointed out at paragraph 10 above, even a Part 36 offer made here at the first reasonable opportunity would have attracted wholly disproportionate costs consequences. Moreover it is obvious that a Part 36 offer of the order of £2,000 or £3,000 made at an early stage in these proceedings stood no prospect of acceptance. Such an offer would have been what Devlin J in Anglo-Cyprian Trade Agencies Limited v Paphos Wine Industries Limited [1951] 1 All ER 873 castigated as "a mere matter of ritual". As this court agreed in Alltrans Express Limited v CVA Holdings Limited [1983] 1 WLR 394 no weight should be attached to a party's failure to make such a ritual act – see per Stephenson LJ at page 402 and per Griffiths LJ at page 404. Here too in my judgment the judge approached the exercise of his discretion in a manner which was fundamentally flawed in principle.
52. It is for these reasons that I agree with the President that the judge's decision produces an unjust and disproportionate result which cannot stand. I agree that the factors identified by the President at paragraph 20 of his judgment should lead to a recovery by the Appellants of only 75% of their costs. For the avoidance of doubt I agree with the President that the Appellants might appropriately have made a Calderbank type offer which would have rendered this debate unnecessary. Such an offer would have been of very considerable benefit to the Claimant if accepted because it would have insulated him from an order such as is now being made to pay the Defendants' costs, or a substantial part thereof, incurred in defending over a long period the ultimately unsuccessful claim. Had the Defendants at the first reasonable opportunity offered £3,000 together with costs proportionate to that recovery, their claim to recover the entirety of their costs thereafter incurred would, in my view, have been strong if not unanswerable. However the failure of the Defendants to make such an offer, which would not have been accepted, does not compel the court to make an order for costs which does not reflect the realities of the litigation.