Coflexip S.A. & Anor v Stolt Offshore Ms Ltd. & Ors [2004] EWCA Civ 213

Case No A3/2003/1852Case No A3/2003/1852/A[2004] EWCA Civ 213
IN THE SUPREME COURT OF JUDICATURE
COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE HIGH COURT OF JUSTICE
CHANCERY DIVISION , PATENTS COURT
Jacob J.
Venue Royal Courts of JusticeDate Friday 27 th February 2004
Strand,
London, WC2A 2LL
Before
LORD JUSTICE PETER GIBSONLORD JUSTICE NEUBERGERSIR MARTIN NOURSE
Between
COFLEXIP S.A. AND ANOTHERRespondentsSTOLT OFFSHORE MS LTD. AND OTHERSAppellants
(Transcript of the Handed Down Judgment ofSmith Bernal Wordwave Limited, 190 Fleet StreetLondon EC4A 2AGTel No: 020 7421 4040, Fax No: 020 7831 8838Official Shorthand Writers to the Court)
Mr. Richard Miller Q.C. and Mr. Douglas Campbell (instructed by Messrs Clifford Chance of London) for RespondentsMr. Roger Henderson Q.C. and Mr. Tom Hinchliffe (instructed by Messrs Bird and Bird of London) for Appellants
Judgment

Introductory

[1]Coflexip SA is the registered proprietor of European Patent (UK) No 0478742 (“the Patent”). The Patent contains product and process claims relating to the laying of flexible conduits on the seabed, which are of particular value to the off-shore oil industry.[2]In 1996, Coflexip SA and the exclusive licensee of the Patent, Technip Offshore Limited (together “Coflexip”) brought proceedings against Stolt Offshore MS Limited and other companies in the same group (“Stolt”) for infringement of the Patent. Coflexip’s Particulars of Infringement alleged: i) In paragraph 1, that Stolt had infringed “by using the process described in the Specification” of the Patent and by, inter alia, “using and/or keeping … a product made in accordance with the invention described in the … Patent”; ii) In paragraph 2, that “in particular” Stolt had kept and used a vessel called the Seaway Falcon “together with the Flexible Laying and Module Handling Systems installed therein” in the “Magnus Swift Field in the North Sea for the purposes of laying flexible flowlines for British Petroleum plc” and that the Seaway Falcon together with the aforesaid system was “being offered for use and disposal” in connection with the laying of flexible flowlines more generally.[3]Stolt denied infringement of the Patent, and counterclaimed that it was invalid and should be revoked on grounds of anticipation and obviousness. Their case relied on certain prior art.[4]The claim and counterclaim came before Laddie J and the hearing lasted around five days. He gave judgment on 29 January 1999 (“the first judgment”). He held that the Patent was valid, and that Stolt had infringed it. Stolt appealed the first judgment. On their appeal, Stolt sought to put in evidence of further prior art. On 31st July 2000, the Court of Appeal refused Stolt permission to adduce this further evidence, and dismissed Stolt’s appeal. They allowed a cross appeal by Coflexip, as a result of which the extent of Stolt’s infringement was rather greater than that found by Laddie J, and the form of order was amended.[5]On Stolt’s application to adduce fresh evidence in the form of fresh prior art, Aldous LJ said, in paragraph 55 that:
“[Stolt] carried out searches which they believed were appropriate and which they believed and still believe were reasonable. It was their decision not to search more widely and it is not right to allow them to reopen their case in this court. There are no exceptional circumstances which would make this an appropriate case for amendment requiring a new trial.”
[6]Chadwick LJ came to the same conclusion. At paragraph 70, he said:
“This is not a case in which it can be said that the prior art on which [Stolt] now wish to rely … would compel the court to hold that [Coflexip’s] patent is invalid. The most that can be said is that [the fresh prior art] provides a further basis for [Stolt’s] attack on the validity of [Coflexip’s] patent. In those circumstances it is clear that, if [Stolt] cannot succeed without relying on the prior art which they seek to introduce by amendment, permission to reamend would, necessarily, lead to a retrial.”
[7]At paragraph 72, Chadwick LJ agreed with the reasoning of Aldous LJ which I have already quoted. He then said at paragraph 73 that to order a retrial would involve ignoring “the interests of other court users” in light of the “limited number of nominated patent judges and the heavy demands on the Patents Court”. At paragraph 74, Chadwick LJ referred to a further argument by Stolt, to the effect that it was in the public interest that the fresh prior art should be considered by the court, because the Patent would continue in force, with monopolistic consequences, possibly wrongly. He also observed that, if there was prior art, which had not been referred to at trial and which would invalidate the Patent, then “it may be expected that the Patent will be challenged by another party in other proceedings”. He concluded on this issue:
“The court should not allow a party to pursue its own private interests in a way which is wasteful of the court’s resources under the guise of promoting an alleged public interest in challenging the monopoly conferred by the patent.”
[8]The Court of Appeal altered the injunction granted at first instance. The form of injunction granted by Laddie J restrained Stolt “from using or offering … a process laying flexible conduit employing a Flexible Lay System of the design installed in the Seaway Falcon …”. At paragraph 58 Aldous LJ made this criticism:
“That injunction makes no reference to the patent nor to the rights stemming from the patent. But it applies whether or not the patent lapses or is revoked and, unless an application is made to the court, it will continue even when the patent expires. It therefore needs some modification.”
Accordingly, the Court of Appeal effectively substituted the more familiar form of injunction.[9]The order resulting from the first judgment, as amended by the Court of Appeal was to this effect:
“(2) [Stolt] … be restrained … from infringing [the] Patent … (3) There be an inquiry as to damages suffered by [Coflexip] by reason of the acts of infringement of the Patent … by [Stolt] or at [Coflexip’s] option an account of profits made by [Stolt] by reason of such acts … the Inquiry [to] proceed having regard to the contents of the judgment of [the Court of Appeal]. (4) [Stolt] pay Coflexip] all sums found to be due to [Coflexip] on the said account/or Inquiry together with interest thereon ….”
Stolt was also ordered to pay the costs of the action, the counterclaim and the appeal.[10]The inquiry as to damages (“the Inquiry”) has been proceeding. In addition to the specific contract relating to the Magnus Swift field with British Petroleum plc (“the Magnus Swift contract”) referred to in paragraph 2 of the Particulars of Infringement, it covers fourteen other pipe-laying contracts (“the fourteen other contracts”). The hearing of the Inquiry is fixed for 26 April 2004 with a time estimate of between four and six weeks. Coflexip has elected to seek damages rather than an account of profits, and its case currently discloses a claim for around £80 million.[11]Meanwhile, during 2002, a company unconnected with Stolt, Rockwater Limited (“Rockwater”), issued proceedings against Coflexip for revocation of the Patent. On 15 April 2003 Laddie J gave judgment in that action (“the second judgment”) in which he found the Patent to be invalid on the basis of prior art, which had not been referred to by Stolt either at first instance or in the Court of Appeal. Accordingly, he ordered that the Patent be revoked, albeit that he stayed the revocation pending Coflexip’s appeal to this court.[12]In light of the revocation of the Patent, Stolt applied for a stay of the Inquiry. The application is ultimately based on Stolt’s submission that the revocation of the Patent “provides a complete defence to the claim for damages”, or constitutes “a bar to enforcement of the adjudged infringement whether by pursuit of the Inquiry for damages or otherwise”, to quote from the skeleton argument prepared by their counsel Mr Roger Henderson QC and Mr Thomas Hinchliffe. If, as a result of Coflexip’s appeal against the second judgment, the revocation of the Patent is confirmed, then (subject to any further appeal to the House of Lords) this argument, if correct, will come into its own. On the other hand, if the second judgment is reversed, the Patent will not be revoked, and Stolt accept that this argument could no longer proceed.[13]If Stolt are right on the basic point, namely that the revocation of the Patent would effectively put an end to their liability in the Inquiry, then, particularly in light of the fact that Laddie J, in the second judgment, concluded that the Patent should be revoked, there would obviously be a very powerful case for saying that the Inquiry should be adjourned to await the outcome of the Coflexip’s appeal against the second judgment. However, in a decision given on 31st July 2003, Jacob J declined to grant a stay, on the grounds that, even if the Patent was revoked, it would not assist Stolt in the Inquiry, and that, in particular, it would not lead to Stolt’s liability for damages to Coflexip, as a result of the first judgment, being effectively discharged. It is against that decision that Stolt now appeal.[14]The reason Jacob J reached his conclusion was essentially on the basis of a decision of the Court of Appeal in Poulton -v- Adjustable Cover & Boiler Block Co [1908] 2 Ch 430. Jacob J described the facts of that case as “not materially different from those of the present case”. I did not understand Mr Henderson to suggest otherwise. His contention on behalf of Stolt was that we should not follow the decision in Poulton for a number of reasons, namely: i) the decision in Poulton is, on analysis, wrong. ii) the law relating to res judicata, on which the reasoning in Poulton was based, has substantially relaxed since 1908; iii) the provisions of the Patent Act 1977 (“the 1977 Act”), and in particular those relating to amendment, indicate that the law has been changed; iv) the provisions of the Community Patent Convention (“the Convention”) which was, of course, not in force in 1908, indicate a different conclusion.[15]I now turn to the decision in Poulton, and will then deal with these four arguments.

Poulton -v- The Adjustable Cover & Boiler Block Co

[16]In Poulton, the plaintiff had succeeded at trial in establishing that the defendants had infringed its patent, and an inquiry as to damages was ordered. Subsequently, the defendants obtained evidence of prior use, and presented a petition of revocation of the plaintiff’s patent which was successful (the revocation being affirmed on appeal). At the ensuing inquiry, the Master held that, as a result of the revocation, the plaintiff had suffered no damage, but that, if the revocation was not to be taken into account, the damages would be assessed at £110.[17]Parker J, whose decision is included in the report at [1908] 2 Ch 430 (but is also reported at (1908) 25 RPC 529), disagreed. He said this at [1908] 2 Ch, 432:
“Now I think that on principle it is absolutely clear that a judgment for an injunction and an inquiry as to damages in an action for infringement does estop the defendant from setting up, on the inquiry, the invalidity of the letters patent. So far as the defendant is concerned, he is estopped from denying the validity of the letters patent at the date of the trial, which is the date with reference to which the inquiry as to damages is directed, and, of course, estopped from denying their validity at the date of the infringement in respect of which the inquiry as to damages is directed.”
[18]In an earlier passage, reported only at 25 RPC 532, he had said this:
“I think it is obvious on the face of it, that a judgment for the revocation of a Patent would in effect put an end to the injunction, because the question whether a Patent has been infringed or not is a question to be tried at the date of the alleged infringement, and, if there is no Patent at that date because it was revoked, that would not be an infringement; so that in effect an Order for revocation of a Patent would in every case put an end to the injunction. But it does not follow from that that it would put an end to the inquiry as to damages which had been directed.”
[19]At [1908] 2 Ch 435, Parker J considered the fact that the revocation of the patent was a decision in rem. However, he rejected the notion that he could apply the “principle of an estoppel upon an estoppel in the present circumstances, so as to avoid having to give the plaintiff in the action a sum by way of damages, when, as a matter of fact, the patent having been ab initio void, he has suffered no damage at all.”[20]The Court of Appeal (whose decision is also reported at (1908) 25 RPC 661) agreed. Vaughan Williams and Fletcher-Moulton LJJ were equally firm in their views, and, although he was “not so clear upon this question” (at least according to the report at 25 RPC 664), Buckley LJ concurred.[21]Vaughan Williams LJ said this [1908] 2 Ch 437:
“In my opinion the judgment so obtained by the plaintiff made the question whether there had been an infringement by the defendants of a patent then valid, which had been granted to the plaintiff, res judicata as between the parties to the action, and operated as a complete estoppel between them”
[22]He continued at 438:
“The question in this case appears to me to be whether the estoppel created as between the plaintiff and the defendants by the judgment in the action remains a binding and effective estoppel, notwithstanding the subsequent revocation of the letters patent, so as to prevent the defendants from reopening, upon the inquiry as to damages, the question of the validity of the patent, which had been decided as between them and the plaintiff in favour of the latter. In my opinion it remains a good estoppel, and there is no estoppel upon an estoppel as suggested.”
[23]Fletcher-Moulton LJ said this at 438:
“In this case a judgment was obtained by the plaintiff against the defendants, by which it became res judicata as between these parties that there had been infringement by the defendants of a patent granted to the plaintiff, which was then valid. As regards the issues of validity and infringement, the cause had then terminated, but there had to be an inquiry as to the amount of the damages. … The order of revocation is in the nature of a judgment in rem which terminates the res, i.e., the letters patent, at the date when the order is made. It has in my opinion no further effect so far as estoppel is concerned.”
[24]At 440-441, Buckley LJ said this:
“[T]he order made in the action was, in my opinion, a complete judgment inter partes. It contained a subsidiary order, which was made upon the basis that the patent was valid, and directed an inquiry to ascertain what, upon that footing, were the damages sustained by the plaintiff. The effect of this judgment was that the validity of the patent was for the purposes of the inquiry res judicata as between the parties to the action. That being so, the revocation of the patent on a petition subsequently presented cannot affect the estoppel so created, or estop the plaintiff from asserting upon the inquiry as to damages that the patent must be regarded as valid.”
[25]As I have mentioned, it has been accepted by all involved in the present case that, unless the decision in Poulton is no longer good law, Stolt’s argument must fail, and the decision of Jacob J refusing a stay of the Inquiry was correct (subject to the question whether he should have granted a stay pending the appeal against his decision). In those circumstances, I turn to the four arguments raised by Stolt.

The argument that Poulton was wrongly decided

[26]Stolt’s case is as follows. Given that the Patent has now been revoked, and that (as Coflexip accepts) the revocation is retrospective in its effect, that revocation should be taken into account when assessing the damages on the Inquiry consequential on the first judgment. For a number of reasons, it would be wrong if Coflexip was entitled to receive, and Stolt were obliged to pay, tens of millions of pounds for an infringement of a patent which, in light of its revocation, should never have been granted and, indeed, should be treated as if it had not ever been granted, including at the time of the alleged infringements. It would be wrong in principle, because damages would be awarded for infringement of a patent which ex hypothesi did not exist. It would be wrong in terms of the public interest, which is engaged because of the monopolistic effect of a patent, that what one now knows to be an invalid patent should be enforced. It would be wrong as between the parties: there would be a wholly disproportionate penalty to Stolt, and a wholly disproportionate windfall for Coflexip, as a consequence of Stolt or its advisers having, through hard luck, poor judgment or ineptitude, overlooked the prior art which was successfully adduced by Rockwater.[27]This argument is obviously attractive. However, the issues determined in the first judgment (substantially upheld by the Court of Appeal) in these proceedings, namely that the Patent was valid and that Stolt infringed it, are res judicata as between the parties. As a result of that judgment Coflexip has obtained against Stolt, in the normal way, an injunction to restrain any further infringement and an inquiry as to damages in relation to past infringements. As between the parties, therefore, there is obvious force that it should follow that the validity of the Patent is no more open to argument than the fact that Stolt infringed it.[28]Nonetheless, I consider there is force in the contention that it does not necessarily follow from this that, at the hearing of the Inquiry, Stolt should be prevented by res judicata from relying upon the subsequent revocation of the Patent to justify the argument that the damages awarded on the Inquiry should be nil. In this connection, there are a number of important features of revocation, which I believe are uncontroversial, and indeed many of which were accepted, indeed advanced, by Mr Richard Miller QC, who appeared with Mr Douglas Campbell, for Coflexip.[29]First, revocation of a patent is not the same thing as invalidity of a patent, although, at least in this jurisdiction, it will automatically flow from a holding of invalidity. The point is well illustrated by the fact that, in the United States, a finding of invalidity does not lead to revocation (see for example Mendenhall -v- Barber-Greene Company (1994) 26 F3d 1573). Secondly, a judicial decision to revoke, unlike a judicial decision as to validity, is a decision in rem (ie, it is a “conclusion against all the world as to … status” - see Ballantyne -v- McKinnon [1896] 2 QB 455 at 462), and not a decision in personam (which is “only evidence between the parties” - ibid). Thirdly, revocation is retrospective in its effect. Thus, the Patent is to be treated as non-existent when Stolt’s acts of infringement occurred.[30]Fourthly, once it occurs, revocation is a fact, albeit that the basis for its existence, and its consequences are, at least in part, a matter of law. In particular, the revocation of the Patent in this case is not merely a fact, but it is a fact which could not have been raised on the issue of invalidity or infringement, or indeed at any time before the first judgment, because it only occurred after the first judgment, although I accept, of course, that it arose from facts which could have been discovered by Stolt. Fifthly, revocation could be for a reason which may have nothing to do with the issues raised in an earlier invalidity action, and for which the person in the earlier invalidity action may have had no reason to know. Thus, under s72(1)(b) of the 1977 Act, a patent can be revoked if “granted to a person who was not entitled” to its grant, and one can fairly easily imagine circumstances in which this ground could exist, where a person seeking to revoke the patent had no reason whatever to be aware of the facts giving rise to that ground.[31]The combined effect of these five features of the revocation of the Patent in this case persuades me that there is a real argument to the effect that the finding of validity and infringement in the first judgment should not prevent Stolt from relying on the subsequent revocation of the Patent on grounds of res judicata. The revocation is a fact which only came into existence after the first judgment, so that it could not have been relied on before the first judgment, and it is retrospective in its effect, and so, it may be said, no damage can have been suffered by Coflexip at the dates of the acts of infringement.[32]I believe that this argument receives some support from the fact, accepted by Mr Miller (and indeed by Parker J in Poulton), that the effect of the subsequent revocation is to put an end to the injunction granted by Laddie J, as varied by the Court of Appeal. At first sight, the fact that the injunction against infringement would cease to have effect once the Patent is revoked, takes matters no further so far as damages are concerned. There is obvious attraction in Parker J’s analysis that any inquiry is concerned with past breaches, in respect of which validity and infringement are res judicata as between the parties, whereas the injunction is concerned with future possible breaches.[33]However, on closer analysis, I do not find that distinction particularly persuasive in strict logic, although it has obvious force in terms of common sense. If the injunction ceases to have effect, it is only because Stolt can rely, in these very proceedings, on the effect of the subsequent revocation of the Patent in later proceedings, notwithstanding a judgment in these proceedings as to the validity of the Patent. That revocation is based on facts which existed at the date of the first judgment. Further, it is retrospective in its effect. In those circumstances, I do not find it easy to understand why, as a matter of strict logic, Stolt should not be entitled to rely on revocation in relation to an inquiry as to damages - especially a future inquiry - based on past infringements, if it is entitled to rely on revocation in relation to future infringements. This is not a case like Mills -v- Cooper [1967] 2 QB 489, where an earlier finding between two parties that someone was a gypsy was not binding in later proceedings, because a person could be a gypsy one time and not at another. The revocation of the Patent was based on facts as they existed at the time of its grant, and, therefore, ex hypothesi, at the time of Stolt’s alleged infringements.[34]The fact that Stolt would remain liable for the costs of the action heard by Laddie J, and the appeal heard by the Court of Appeal, does not, to my mind, call the argument into question, despite a contention to the contrary. Stolt fought Coflexip on validity and infringement, and they lost on both issues at first instance and on appeal. It is because of the fact that they lost and Coflexip won, that Stolt are liable for Coflexip’s costs. I do not see how that undermines Stolt’s argument for saying that they should be entitled to rely upon the subsequent revocation in the Inquiry. Indeed, it could be said that it would accord with most people’s notion of justice that, as a result of Stolt’s failure and Coflexip’s success in the proceedings, Stolt should pay Coflexip’s costs. However, that does not call into question the offensiveness of the notion of Stolt being liable for, and Coflexip being entitled to, tens of millions of pounds for infringement of a patent which is now shown to be invalid, with retrospective effect.[35]For these reasons, I think there are grounds for concluding that the decision in Poulton was wrong. However, even without considering the arguments the other way, most of which are expressed in clear terms in the judgments in Poulton, I do not believe that these grounds could justify this court departing from its earlier decision. The established circumstances in which the Court of Appeal can depart from one of its earlier decisions were identified in Young -v- Bristol Aeroplane Co Ltd [1944] 2 KB 718 at 729-730. Thus, in the recent case of Cave -v- Robinson, Jervis & Rolf [2002] 1 WLR 581, while the Court of Appeal clearly had very severe doubts as to the correctness of its earlier decision, it was not prepared to depart from it, on the basis that it was for the House of Lords to decide whether to over-rule it, as they did - see at [2002] 1 WLR 1107.[36]Accordingly, I consider that it would be wrong for this court to depart from an earlier decision merely because it would have arrived at a different result from that reached in the earlier decision. Quite apart from this, Poulton is a strongly expressed unanimous decision of the court which has stood for over 95 years, and which has been cited without disapproval (and, some might say, with implied approval) in the leading textbooks on patents, (Terrell on The Law of Patents 15th Edition: see at paragraphs 4.26 and 13.44) and on res judicata (Spencer Bower, Turner & Handley - The Doctrine of Res Judicata, 3rd Edition: see at paragraphs 168, 221 and 372).[37]In these circumstances, it is necessary to determine whether the decision in Poulton should not be followed for any of the three reasons advanced on behalf of Stolt.

The argument that the law of estoppel has been modified since Poulton

[38]If the conclusion reached by the Court of Appeal in Poulton is correct (which, for the reasons discussed above, must be assumed) this Court must proceed on the basis that, at least as the law was understood in 1908, Stolt would be precluded from raising the revocation of the Patent in the Inquiry on grounds of estoppel per res judicata. Stolt’s contention that, since 1908, and in particular over the last fifteen years, the strictness of that doctrine has been diluted, rests principally on the reasoning in two decisions of the House of Lords, Arnold -v- National Westminster Bank plc [1991] 2 AC 93 and Johnson -v- Gore Wood & Co [2002] 2 AC 1.

The different types of estoppel

[39]In order to deal with Stolt’s contention, it is necessary first to examine the difference between various types of res judicata and their consequences. Three types of res judicata may identified through the cases, namely cause of action estoppel, issue estoppel, and abuse of process. Arnold was a case involving issue estoppel, and Johnson involved abuse of process.[40]At 104C-D in Arnold Lord Keith of Kinkel explained that cause of action estoppel:
“arises where the cause of action in the later proceedings is identical to that in the earlier proceedings, the latter having been between the same parties or their privies and having involved the same subject matter.”
[41]In relation to the effect of cause of action estoppel, Lord Keith said at 104D-E:
“[T]he bar is absolute in relation to all points decided unless fraud or collusion is alleged, such as to justify setting aside the earlier judgment. The discovery of new factual matter which could not have been found out by reasonable diligence for use in the earlier proceedings does not, according to the law of England, permit the latter to be re-opened.”
[42]At 105D-E, Lord Keith said that issue estoppel:
“may arise where a particular issue forming a necessary ingredient in a cause of action has been litigated and decided and in subsequent proceedings between the same parties involving a different cause of action to which the same issue is relevant one of the parties seeks to re-open that issue.”
[43]Estoppel per rem judicatam also extends to some types of abuse of process. Thus, at 104F, Lord Keith said:
“Cause of action estoppel extends also to points which might have been but were not raised and decided in the earlier proceedings for the purpose of establishing or negativing the existence of a cause of action.”
[44]Lord Keith went on to explain what he had in mind in a little more detail at 104F-105B:
“In Henderson v. Henderson (1843) 3 Hare 100, 114-115, Sir James Wigram V.-C. expressed the matter thus: ‘In trying this question, I believe I state the rule of the court correctly, when I say, that where a given matter becomes the subject of litigation in, and of adjudication by, a court of competent jurisdiction, the court requires the parties to that litigation to bring forward their whole case, and will not (except under special circumstances) permit the same parties to open the same subject of litigation in respect of matter which might have been brought forward as part of the subject in contest, but which was not brought forward, only because they have, from negligence, inadvertence, or even accident, omitted part of their case. The plea of res judicata applies, except in special cases, not only to points upon which the court was actually required by the parties to form an opinion and pronounce a judgment, but to every point which properly belonged to the subject of litigation, and which the parties, exercising reasonable diligence, might have brought forward at the time.’ It will be seen that this passage appears to have opened the door towards the possibility that cause of action estoppel may not apply in its full rigour where the earlier decision did not in terms decide, because they were not raised, points which might have been vital to the existence or non-existence of a cause of action. The passage has since frequently been treated as settled law, in particular by Lord Shaw, giving the advice of the Judicial Committee of the Privy Council, in Hoystead v. Commissioner of Taxation [1926] A.C. 155, 170. That particular part of it which admits the possible existence of exceptional cases was approved by Lord Kilbrandon in Yat Tung Investment Co. Ltd. v. Dao Heng Bank Ltd. [1975] A.C. 581, 590: ‘The shutting out of a 'subject of litigation' - a power which no court should exercise but after a scrupulous examination of all the circumstances - is limited to cases where reasonable diligence would have caused a matter to be earlier raised; moreover, although negligence, inadvertence or even accident will not suffice to excuse, nevertheless 'special circumstances' are reserved in case justice should be found to require the non-application of the rule.’”
[45]Lord Keith also explained at 107C, that “although Henderson … was a case of cause of action estoppel, the statement there by Wigram VC has been held to be applicable also to issue estoppel”. Immediately thereafter, he said this at 107D:
“That statement [sc. of Wigram VC in Henderson] includes the observation that there may be special circumstances where estoppel does not operate. The instant case is concerned with the nature of such special circumstances.”
[46]In relation to issue estoppel, Lord Keith said this at 109A-B “In my opinion your Lordships should affirm it to be the law that there may be an exception to issue estoppel in the special circumstance that there has become available to a party further material relevant to the correct determination of a point involved in the earlier proceedings, whether or not that point was specifically raised and decided, being material which could not by reasonable diligence have been adduced in those proceedings.” He then went on to consider the question:
“… whether the further relevant material which a party may be permitted to bring forward in the later proceedings is confined to matters of fact, or whether what may not entirely inappositely be described as a change in the law may result in, or be an element, in special circumstances enabling an issue to be re-opened.”
[47]In the passage that followed, Lord Keith concluded that a change in the law was a factor which could be taken into account, and that, on the facts of that case, the party seeking to challenge the earlier judgment on the basis of developments in the law which showed it to be wrong should be permitted to take that course.[48]So far as abuse of process is concerned, Lord Bingham described it in Johnson at 31A as “separate and distinct from cause of action estoppel and issue estoppel [although it] has much in common with them”. At 31B-D he continued:
“It is, however, wrong to hold that because a matter could have been raised in earlier proceedings it should have been, so as to render the raising of it in later proceedings necessarily abusive. That is to adopt too dogmatic an approach to what should in my opinion be a broad, merits-based judgment which takes account of the public and private interests involved and also takes account of all the facts of the case, focusing attention on the crucial question whether, in all the circumstances, a party is misusing or abusing the process of the court by seeking to raise before it the issue which could have been raised before.”
The special facts of Johnson, as emphasised by Lord Bingham’s statement at 31A, suggest that it may not be a cause of action estoppel case, or an issue estoppel case, but a case of pure abuse of process. It may therefore be unsafe to rely on it to any great extent in any case such as the present, which is concerned with cause of action estoppel and/or issue estoppel.[49]To the extent that Coflexip’s res judicata argument, based on Poulton, rests on strict cause of action estoppel, I do not consider that it is open to this court to justify departing from Poulton on the basis that the law relating to cause of action estoppel has developed, and become more flexible, since 1908. The law relating to strict cause of action estoppel was restated clearly, albeit obiter, by Lord Keith in Arnold at 104D-E, and nothing in the speech of Lord Bingham in Johnson can fairly be said to call it into question.[50]On the other hand, insofar as Coflexip’s case is based on strict issue estoppel, I consider that there is a much stronger argument to support the contention that the law has developed significantly since 1908. Although Lord Keith at 109A referred to the House of Lords “affirm[ing] it to be the law” that special circumstances could defeat an issue estoppel, unlike a strict cause of action estoppel, it appears that issue estoppel has become less strict over the last fifty years or so. By the 1960s Lord Reid was expressing the view that there was “room for a good deal more thought before we settle the limits of issue estoppel”, and Lord Upjohn was saying that “the principle of issue estoppel must be applied to the circumstances of the subsequent case with [the] overriding consideration” of “work[ing] justice and not injustice”. These observations, from Carl Zeiss Stiftung -v- Rayner & Keeler Limited (No 2) [1967] 1 AC 853, were cited with obvious approval in Arnold by Lord Keith at 107E.[51]The reference to “strict cause of action estoppel” is to the type of case where a party seeks to raise in a second action the “identical” issue to that decided in the first action. A relevant example would be a second attempt to invalidate a patent, after a first action had failed, relying on the same prior art and on the same grounds (eg obviousness and anticipation) as had been raised in the first action. As I read Lord Keith’s analysis of the law in Arnold, the type of abuse of process discussed in Henderson can apply in cases of cause of action estoppel and, although, in cases of strict cause of action estoppel, there is no escape other than through fraud or collusion, the somewhat more flexible approach in relation to issue estoppel applies to (indeed is to be derived from) Henderson abuse of process cases, even where those cases involve cause of action estoppel as Henderson itself did. That conclusion seems clear from the observations of Lord Keith which I have quoted at 107C-D. In other words, the law, as established in Arnold, appears to be this. In relation to strict cause of action estoppel, the law has not changed since 1908, and the estoppel can only be overridden on grounds of fraud or collusion. In relation to strict issue estoppel, and abuse of process estoppel (whether cause of action estoppel or issue estoppel), special circumstances, other than fraud or collusion, can be relied on to enable the court to disapply the estoppel, if the justice of the case requires it. (At the risk of apparently over-complicating matters, I would suggest that abuse of process cause of action estoppel may actually be a species of issue estoppel: it applies where the issue has already been litigated but a party wishes to raise new evidence or a new argument on that same issue. In the light of that, it is scarcely surprising that the House of Lords should have concluded that the same rules should apply to issue estoppel as apply to abuse of process cause of action estoppel.)[52]In my view, although it is fair to say that Lord Keith related his reasoning back to what was said in Henderson, which was, of course, decided before Poulton, the law has moved on since the decision in Poulton, so far as issue estoppel and abuse of process estoppel are concerned. Apart from earlier cases such as Carl Zeiss, the decision in Arnold itself represented a change, or at the very least a significant development, in the law. This appears to me to be recognised in Spencer Bower in the discussion of Arnold at paragraph 189. Furthermore, I consider that the more recent decision of the House of Lords, namely Johnson, which was concerned with abuse of process more widely, tends to support the proposition that the law on the topic of res judicata has developed in the past 100 years, and indeed, may still be developing.[53]Accordingly, to the extent that Coflexip is relying on issue estoppel or abuse of process cause of action estoppel, rather than strict cause of action estoppel, I am of the view that it is open to us to reach a different result from that reached in Poulton, provided, of course, that in so doing we are faithful to the principles and approach laid down by the House of Lords in Arnold (and, to the extent that it is relevant, in Johnson). In other words, while Poulton requires us to accept that there is a res judicata argument open to Coflexip, developments in the law would entitle us to hold that, provided the facts justify it, the estoppel is one to which we could refuse to give effect, unless it is strict cause of action estoppel.[54]That is not to say that issue estoppel or abuse of process cause of action estoppel are a dead letter, or are even to be more honoured in the breach than in the observance. Far from it. The commercial common sense, even, themoral strength, of the respondent’s case in Arnold was very strong. Yet the House of Lords (like the Court of Appeal and Sir Nicolas Browne-Wilkinson, V-C at first instance) were plainly concerned about over-riding the issue estoppel which stood in the respondent’s way. In the event, they did override the estoppel because of the special reasons in that case (as to which, see below).

The type of estoppel involved in this case

[55]The next question is whether this is a case of strict cause of action estoppel, strict issue estoppel, or abuse of process cause of action estoppel.[56]So far as the finding of validity in relation to the damages claimed in relation to the Magnus Swift contract is concerned, Mr Henderson accepted, to my mind rightly, that Coflexip was entitled to rely on cause of action estoppel. In light of the above analysis of Lord Keith’s speech in Arnold, it is necessary to consider whether it is strict cause of action estoppel or abuse of process cause of action estoppel.[57]In favour of the former view is the fact that the relevant issue which was determined between the parties in the first judgment in relation to the pleaded infringement, namely the Magnus Swift contract, was whether or not the Patent was valid, and that is the very cause of action which, in these very proceedings, it may be said that Stolt now wish to reopen.[58]On the other hand, the point which Stolt wish to raise, whether identified as the fact that the Patent has been revoked, or the basis upon which the Patent was revoked (namely the prior art relied on in the second judgment), was not in issue between the parties in the proceedings which led to the first judgment, which is the judgment upon which Coflexip relies as founding a res judicata. In those circumstances, it appears to me that this is not a case of strict cause of action estoppel but of abuse of process cause of action estoppel. It would, of course, be different if, rather than relying upon an actual revocation based on new prior art, Stolt were seeking to rely upon the same prior art which had actually been considered by the court in the first judgment.[59]My view that this case and Poulton are not cases of strict cause of action estoppel, but is a case of abuse of process cause of action estoppel, receives a little support from the reference in footnote 69 of paragraph 190 of Spencer Bower to the decision of Romer J in Shoe Machinery Company -v- Cutlan [1896] 1 Ch 667. In that case, the patentee had succeeded at trial in obtaining a declaration of validity and a determination of infringement, and, in subsequent proceedings, the infringer sought to challenge the validity of the patent by raising a fresh argument based on anticipation. Romer J held that he was not entitled to do so because the question was res judicata as between the parties. The significance of the location of the citation in Spencer Bower is that it is dealt with in a section headed “Questions not raised in earlier proceedings”, a section introduced by the reference to Henderson and the citation of the passage therefrom quoted by Lord Keith in Arnold.[60]It is true that the decision in Poulton itself is not cited in this part of Spencer Bower. Having looked at the four references to Poulton in Spencer Bower (footnote 91 of paragraph 168, footnote 38 of paragraph 221, footnote 43 to paragraph 246 and footnote 52 of paragraph 372) I can see nothing which suggests that the authors of Spencer Bower had any different view as to the nature of the estoppel in Poulton. It is also true that, in Shoe Machinery, the second attack on the patent was raised in separate proceedings, whereas in the present case Stolt seek to attack the Patent in the very proceedings in which Stolt’s earlier challenge failed. However, that difference does not appear to me to alter the fact that, as a matter of logic, if a point is subject to strict cause of action estoppel (or abuse of process cause of action estoppel) in one case, then it must be subject to the same type of estoppel in the other case. In Shoe Machinery, the infringer was seeking a second opportunity to attack the patent. To that extent Stolt’s position is stronger here, as they are trying to rely on an actual revocation.[61]It is right to acknowledge that my view may be inconsistent with the view reached by Aldous J in Chiron Corporation -v- Organon Teknika Limited (No 6) [1994] FSR 448 at 453-454. However, it was not a view necessary for his decision, as he made it clear that he would have reached the same conclusion if abuse of process estoppel applied.[62]It may be said that Stolt’s argument involves giving, at least in a case like this, strict cause of action estoppel a relatively narrow compass. It seems to me that the facts of this case (although not unique) are pretty exceptional in the run of cases concerned with cause of action estoppel. Furthermore, estoppels, particularly where they are almost absolute in their effect (as in the case of strict cause of action estoppel) should not, as I see it, be given a wide effect. After all, as Lord Millett recently said, such an estoppel “comes into its own only where the decision is wrong” (see Mulkerrin -v- PricewaterhouseCoopers [2003] 1 WLR 1937 at 1941C).[63]In relation to the fourteen other contracts, it appears to me that one is in the realm of issue estoppel, rather than cause of action estoppel. Although the parties have agreed that the fourteen other contracts are to be dealt with at the Inquiry in these proceedings, they were not pleaded in Coflexip’s Particulars of Infringement, and they were not therefore expressly dealt with in the first judgment. This might appear to be a very narrow point of distinction between the Magnus Swift contract and the fourteen other contracts, not least because the issue which Stolt wish to raise does not relate to infringement at all: it is an issue which would apply to any alleged infringement, namely the validity or the existence, of the Patent.[64]Nonetheless, as the facts of Arnold demonstrate, the extent of cause of action estoppel, as opposed to issue estoppel, is to be relatively narrowly confined. Arnold involved a lease of some 32 years with provisions for review of the rent, in line with the market, to be assessed by an arbitrator, every five years. On the first review, an issue relating to the interpretation of the rent review clause, was determined by the arbitrator in favour of the tenant, but he indicated the higher rent he would have fixed had he found in favour of the landlord. Walton J held the arbitrator was wrong in his interpretation of the rent review clause, and substituted the alternative higher figure as the rent from the first review date. On the second review, the tenant sought to reopen the question of interpretation, and the landlord contended that this was res judicata having been precisely the issue determined by Walton J. Although it was accepted that the facts did give rise to res judicata, the House of Lords held that the case was one of issue estoppel, and that, on the special facts, the issue of construction could (and on the facts should) be reconsidered on the second review, even though that would involve relitigating the very same dispute of construction determined by Walton J.[65]In these circumstances, if, for instance, the fourteen other contracts had been the subject of separate proceedings for infringement brought after the first judgment, the issue of the validity or enforceability of the Patent would undoubtedly been res judicata as between the parties, but Coflexip would be relying on issue estoppel, not cause of action estoppel. As already mentioned, the fact that damages are being claimed or assessed in relation to the fourteen other contracts in these very proceedings, rather than in separate or later proceedings, should not, as a matter of logic, principle or fairness, alter the nature of the estoppel, in my opinion.[66]For these reasons, I am of the view that Coflexip’s reliance on res judicata as to the enforceability of the Patent: i) is based on abuse of process cause of action estoppel, rather than strict cause of action estoppel so far as the Magnus Swift contract is concerned; ii) is based on issue estoppel so far as the fourteen other contracts are concerned.[67]Accordingly, I believe that it is open to this court, notwithstanding the reasoning and decision in Poulton, to hold that Stolt can rely in the Inquiry on the revocation of the Patent, which, as it seems to me, would result in the damages being assessed at nil, even though validity is res judicata between the parties. In other words, while remaining faithful to the reasoning and conclusion in Poulton, to the effect that the question of validity determined the first judgment is res judicata between the parties, the law, as it has been developed and explained in Arnold, would entitle this court to hold that it may nonetheless be open to Stolt to rely upon the revocation of the Patent as a result of the second decision (assuming, of course, that that decision is upheld on appeal).

Should effect be given to the estoppel in this case

[68]I now turn to consider the central question, namely whether Stolt should be allowed to raise in the Inquiry the fact that the Patent has been revoked. There are a number of reasons for allowing Stolt such an indulgence.[69]First, there is the unfairness of Coflexip receiving, and Stolt having to pay, substantial damages for infringement of a patent which, as a result of its revocation, has always been ineffective. This unfairness is similar to (if rather larger in scale than) that which would have arisen the issue estoppel had been enforced in Arnold. In that connection, Lord Keith said this, at 110D-E:
“I consider that anyone not possessed of a strictly legalistic turn of mind would think it most unjust that a tenant should be faced with a succession of rent reviews over a period of over 20 years all proceeding upon a construction of his lease which is highly unfavourable to him and is generally regarded as erroneous.”
To the same effect, he observed a little later at 110F:
“There is much force also in the view that the landlord, if the issue cannot be reopened, would most unfairly be receiving a very much higher rent than he would be entitled to on a proper construction of the lease.”
[70]The amount of money at stake is not, I think, irrelevant. Even allowing for the effect of inflation between 1908 and today, the sum involved in Poulton, namely £110, was of a very different order from that which appears to be involved here. In present values terms, the award in Poulton cannot have been much more than £10,000. The amount claimed by Coflexip on the Inquiry is, as I have mentioned, approximately £80 million.[71]Secondly, although the second judgment, resulting in the revocation of the Patent, was based on evidence, namely prior art, which Stolt could have obtained for the hearing which resulted in the first judgment, the second judgment itself and the fact of revocation were not available to Stolt until after the first judgment: indeed, they were not available until after the decision of the Court of Appeal upholding the first judgment on validity. Accordingly, to some extent like Arnold, this is a case where a subsequent, highly relevant decision of the court is being invoked. Indeed, it is a decision about the very subject matter of the first decision, it is a decision in rem, and it is a decision which is retrospective in its effect.[72]Thirdly, the public interest, which is one of the two bases upon which estoppel per rem judicatam rests, can be said to favour Stolt in this case. Because of the unfairness if the plaintiff could not have done so, Lord Keith said this at 110G in Arnold:
“I consider that abuse of process would be favoured rather than prevented by refusing the plaintiffs permission to reopen the disputed issue.”
There is no special public interest in whether a tenant pays too high a rent to his landlord under a lease. There is, however, rather more of a public interest in whether the grantee of a patent, which turns out to be invalid, is entitled to enforce that patent.[73]At paragraph 10, Spencer Bower refers to the “interest of the community in the termination of disputes, and in the finality and conclusiveness of judicial decision”. The present litigation between Stolt and Coflexip would finish much sooner, and much more cheaply, if Stolt were entitled to rely upon the revocation of the Patent at the Inquiry. It would put an end to the Inquiry, which is currently estimated to last around five weeks, and there would obviously be no room for a possible appeal.[74]Further, there is already a second, inconsistent, decision, which is a decision in rem. The law would, at least in some respects, be more certain and consistent, if that second decision applied in respect of everybody than if it did not. In any event, the public interest (and the patentee’s interest) in finality is a weaker factor in patent validity proceedings, partly because of the public interest against monopolies, and partly because a patent can be subject to attack for invalidity by “any person” - see s72(1) of the 1977 Act. Accordingly, a determination of validity is never capable of being final to the same extent as most other determinations.[75]Fourthly, I believe that Stolt’s case can be said to be at least as consistent with the private right policy behind the doctrine of res judicata as Coflexip’s case. That policy is explained in Spencer Bower, paragraph 10, namely “the right of the individual to be protected from vexatious multiplication of suits and prosecutions.”[76]Apart from the points already made in relation to the public policy, permitting Stolt to rely upon the fact of revocation would not enable Stolt to have another go at seeking to establish a ground of invalidity of the Patent, which was the issue litigated before Laddie J. It is true that, in that litigation, Stolt were arguing that the Patent should, as a result of invalidity, be revoked. However, if Stolt could rely upon the revocation in the Inquiry, it would not be a question of their arguing that the Patent was invalid, or even that it should be revoked, as they argued before Laddie J. It would be a question of Stolt relying on the undoubted and subsequent fact that the Patent has been revoked.[77]The terms of a lease are normally of concern only to the landlord and tenant (and their respective privies). Accordingly, once a provision in a lease has been given a particular construction by the court in relation to a particular rent review as in Arnold, the landlord can reasonably expect that to be the end of the construction issue in relation to all further reviews. On the other hand, as mentioned, a patent may be attacked by “any person”. Thus, a patentee who successfully fights off an attack of invalidity, must always be aware that another attack on the validity of his patent could take place at any time. Indeed, the facts of Poulton, and the remarks of Chadwick LJ on appeal from the first judgment in this case illustrate that point.[78]Fifthly, there is the fact that the revocation (if it occurs) actually will have a significant effect on the parties in this action and on the order made as a result of the first judgment. It will result in the injunction ceasing to have effect. Thus, unlike in Arnold, the fact which Stolt will seek to invoke against Coflexip in the Inquiry in this case, will have had an effect on Coflexip’s rights against Stolt in this case already. Further, there would be, at least in terms of strict logic, an arguable inconsistency if Stolt could rely on the revocation to say that the injunction has ceased to have effect but could not invoke it on the Inquiry, given that the revocation is retrospective. Particularly as the Inquiry has not taken place, the fact that the injunction is concerned with future infringements can be said to be a rather thin ground of distinction.[79]Indeed, if, after the Patent has been revoked, Coflexip were to start fresh infringement proceedings against Stolt for damages, it would not merely be a matter of Stolt being able to raise the revocation as a defence: the fresh proceedings would be summarily dismissed. This would, I think, be true even if Coflexip was relying on acts of infringement which predated the second judgment (or even if they predated the first judgment and Coflexip reasonably was unaware of the acts). It would seem odd if the Court had to give effect to the estoppel on the present claim for damages while it effectively could not do so in relation to a fresh claim for damages based on precisely the same facts.[80]On the other hand there is no doubt that Coflexip has substantial grounds for contending that, even if the rules relating to estoppel have been relaxed, they have not been and should not be sufficiently relaxed for Stolt to avoid the estoppel. First, there is the public interest and the private interest arising from any res judicata . The law as laid down in Poulton is certain, and it can be said that it should not be subject to an uncertain “merits-based” approach, to quote Lord Bingham in Johnson. Secondly, what is sought to be relied on by Stolt is the second decision, which resulted in the revocation of the Patent, which Stolt could have achieved if they had found the decisive prior art. Accordingly, Stolt can be said to be responsible for their own problems, in a way which could not be said of an unsuccessful litigant who lost an earlier action because of facts which he could not with reasonable diligence have found, or because of case law which was not available to him at the time of the first hearing. Thirdly, the special feature in Arnold of the appeal route being shut off by Walton J, rendering his wrong decision unassailable by refusing to grant a certificate under s1 of the Arbitration Act 1979 (see 110C-D), is not a point available to Stolt here. Indeed, that feature is suggested to be the decisive factor in Arnold in Spencer Bower at paragraph 190.[81]Fourthly, I am concerned, as was Jacob J, by Mr Miller’s argument that, if Stolt succeed here, unsuccessful defendants will search for better prior art and arrange for an attack on the patent by a third party. To a substantial extent, this sort of problem can be said to be an inherent feature of patent litigation, because a patent is good against the world, and therefore revocation is in rem, and “any person” can challenge a patent. Further, given the cost of patent litigation, the risk is unlikely to arise other than in relation to relatively valuable patents.[82]Nevertheless, the undesirability of an unsuccessful defendant to an infringement claim being able to raise new grounds of invalidity after he has lost at trial has been referred to in a number of cases, from Shoe Machinery (see per Romer J at 672) up to the present day (see for instance per Chadwick LJ at paragraphs 73 and 74 on the appeal from the first judgment). If Stolt were entitled to rely upon the revocation of the Patent at the Inquiry in the instant case, there must be a real risk in the future that a defendant, against whom an inquiry is ordered for infringement of a patent, will arrange for a third party to challenge the validity of the patent afresh on the basis of new evidence (or, even, I suppose on the basis of the same evidence in front of a different judge and with different experts), and will then try and rely on the existence of that challenge as a reason for deferring the assessment of damages, or putting off the enforcement of any award of damages.[83]Fifthly, Mr Miller suggested that Stolt’s point is one of timing, as much as of principle. If, for instance, the revocation of the patent had only occurred after the Inquiry as to damages had been completed, and the damages paid, there would be very substantial, possibly insuperable, difficulties in the way of Stolt getting those damages back. However, even if Coflexip succeeds on this application, there is still a potential timing aspect: if the order revoking the Patent had been made before the Court of Appeal gave judgment in this action, then it appears to me likely that the Court of Appeal would have allowed Stolt to rely on that fact on the appeal from the first judgment. In that connection, it appears to me that the main reason that the Court of Appeal refused Stolt permission to adduce the fresh prior art on which they wished to rely was because it would have required a retrial (see per Chadwick LJ at paragraphs 72 and 73). That particular problem would not arise if the new evidence was simply the fact of the revocation of the Patent.[84]Mr Miller’s point about timing also prompts this thought. If Stolt had not appealed the first decision to the Court of Appeal but all the other events had occurred as they did, they could now apply for permission to appeal out time, relying on the revocation of the Patent as new evidence for allowing the appeal on validity and on infringement. Given that allowing Stolt to rely on this new material would not involve the necessity for a new trial and (subject to the appeal against the second decision being allowed) would lead to a very short appeal, the Court of Appeal would probably allow the fresh material to be adduced. (Certainly the reasoning of Chadwick LJ on the appeal against the first decision would so suggest). It would seem a little odd if, by having unsuccessfully appealed, Stolt were in a far weaker position with regard to relying on the revocation to avoid their liability for damages, than if they had not appealed.[85]Quite apart from this, I consider that a timing problem can always be said to arise in relation to a litigant’s attempt to rely upon a new point (whether of fact or of law) after judgment. Thus, the development in the law which was so important in Arnold could have occurred just before the fourth or fifth review rather than just before the second review. It seems to me that the essential point, in a case such as this, is that so long as the matter is still under the aegis of the court, the court has the power, indeed the duty, to do what it considers right, in light of the facts of the case, and in accordance with principles of law and justice. Once litigation has concluded, on the other hand, and when the damages have been paid over, the matter is out of the hands of the court so that, save at least in exceptional cases, different considerations apply.[86]In these circumstances, the question whether or not to allow Stolt to rely in the Inquiry upon the revocation of the Patent is not easy to resolve. Although effect should normally be given to a res judicata, one must bear in mind Lord Upjohn’s observations in Carl Zeiss, and Lord Kilbrandon’s statement in Yat Tung, each judgment emphasised the overriding requirement of the justice of the case and each of which was cited with approval by Lord Keith in Arnold. Indeed, they are mirrored in the emphasis in Lord Keith’s speech in Arnold on the substantial financial unfairness if the point at issue in that casecould not be reconsidered. Unless there are unsatisfactory wider implications, it seems to me that anyone “not possessed of a strictly legalistic turn of mind would think it most unjust” that Coflexip should be entitled to receive, and Stolt required to pay, tens of millions of pounds forinfringements of a revokedpatent, simply because Stolt, through misfortune, poor judgment or ineptitude, failed to find the crucial prior art which has now been established as justifying the revocation of the Patent.[87]In this connection, the fact that Stolt can be said to be responsible for the existence of the res judicata (because they failed to rely on the vital prior art) is undoubtedly a powerful factor in favour of giving effect to the estoppel, but it is not necessarilya decisive factor. As Sir Nicolas Browne-Wilkinson said at first instance in Arnold in a passage quoted with apparent approval by Lord Keith at 103F-G:
“The question therefore is whether, given a subsequent change in the law indicating that the earlier decision was wrong, the injustice of holding the plaintiff in the second action bound by the erroneous decision in law in the first action outweighs the hardship to the other party in having to relitigate the matter and the public interest in the finality of legal proceedings.”
In a later passage, expressly approved by Lord Keith, Sir Nicolas referred to “the yardstick of whether issue estoppel should apply ‘as being’ the justice to the parties” (see 110H).[88]Blameworthiness is a relevant matter. However, I do not consider that it can be decisive. It is little short of fanciful to think that Arnold would have been decided differently if it had been shown that an important argument or a relevant authority had not been put before Walton J by the tenant. Even accepting Stolt’s blameworthiness, I consider that, subject to considering the wider implications, the various factors I have identified (in paragraphs 69 to 79 above) to justify over-riding the estoppel outweigh those which exist in favour of giving effect to a res judicata, even allowing for Stolt’s responsibility for their situation. Coflexip would not have the orders for costs already made in its favour revoked; indeed, as presently advised, I consider that it should recover the costs of the Inquiry to date.[89]The question, then, is whether the wider implications of such a decision, namely the over-riding of an estoppel, the risk of unsuccessful defendants organising revocation proceedings, and the uncertainty which would be thereby created lead to the conclusion that it would be a wrong decision.[90]So far as principle is concerned, I have already emphasised the fact that the relaxation of the strict enforcement of res judicata (other than in cases of strict cause of action estoppel) does not mean that such estoppels are to be easily over-ridden. It is an injustice to deprive a party of the benefit of a judgment. Nonetheless, the present case seems to me to be one where the combination of factors I have identified as being in Stolt’s favour justify the conclusion that this injustice to Coflexip is over-ridden by the injustice to Stolt of giving effect to the estoppel.[91]I turn to the risk of unsuccessful defendants arranging subsequent revocation proceedings. The court has to be wary of any “floodgates” argument, because the threatened flood may not occur if the gates are opened. Furthermore, the gates in this case are in any event partially open, at least where a valuable patent has a significant time to run. An unsuccessful defendant will have a substantial interest in having the patent revoked, so that the injunction granted against him is discharged (as in Poulton). Further, other parties would have an interest in relying on the new prior art so that fresh proceedings would be brought anyway (as in this case) - a point made by Chadwick LJ at paragraph 74 in the appeal from the first judgment.[92]I am far from convinced that there would be a risk of many cases of an unsuccessful defendant seeking to fend off the assessment of payment of damages for infringement by organising a fresh claim for invalidity. The fresh claim would be expensive, and it would have to involve fresh evidence (normally of new prior art) of a reasonably convincing nature before it could be seen to stand any chance of success. However, the fact that such fresh claims may only normally occur in cases where substantial sums of money are involved is by no means a wholly satisfactory answer, because claims for patent infringements involving large sums of money are not particularly infrequent. The amount of money which parties seem prepared to throw at patent litigation may be indicative of the reality of the risk of subsequent revocation proceedings being orchestrated by a defendant with a view to avoiding, or at least putting off, a liability to pay very substantial damages.[93]Any defendant, against whom an inquiry has been ordered, would have to invoke the assistance of the court before the inquiry could be stayed. Stolt’s case here is particularly strong. First, subject to the second decision being reversed, this is not a case where the Patent may be revoked: it has been ordered that the Patent be revoked (albeit that the order has been stayed pending appeal). Secondly, there is no suggestion of Stolt having in any way caused or orchestrated the revocation proceedings brought by Rockwater. Thirdly, there is no indication of Stolt having in any way dragged their feet or sought to hold up or put off the Inquiry in the hope or expectation of the Patent being revoked. Fourthly, there is no question of Stolt having decided to suppress the prior art eg as it may have invalidated one of their own patents.[94]While it is idle to pretend that this is by any means a universally satisfactory answer, I consider that, where a patentee has obtained an order for assessment of damages for infringement, attempts by the infringer to put off the assessment or the payment of the damages, on the ground that the patent may be subsequently revoked, would usually fail. If there were no revocation proceedings on foot, the attempt would be doomed, save in very unusual circumstances. Even where there were revocation proceedings on foot, the attempt would normally face serious difficulties. The alleged injustice of paying damages for infringement when the patent has been revoked is very different from that of paying damages when the patent may be revoked. After all, as explained above, in every case where damages for infringement are ordered there is always a risk that someone else may seek to revoke the patent.[95]There is a substantial distinction in this connection between an application to revoke a patent and its actual revocation. As Chadwick LJ made clear when considering the appeal from the first judgment, there is all the difference between, for instance, prior art which may invalidate a patent, and prior art which will invalidate, or has invalidated, a patent.[96]It is one thing to deprive a party, such as Coflexip, of the fruits of a successful judgment in circumstances where one can say with certainty that, in light of what has happened since judgment and before the assessment of those fruits, the whole basis for the judgment was flawed. It is quite another to hold that the assessment and enjoyment of those fruits should be held up because of other proceedings which might produce such a result.[97]As to uncertainty, it appears to me that this is a general point which may be made against the modern attitude to res judicata, which places rather more weight on justice in the particular case, as opposed to certainty of outcome, than in the past. Further, in a case such as the present, with an actual revocation, the Inquiry not even started and no criticism of the defendant save failure to find the crucial prior art, I believe that justice would normally require the estoppel to be over-ridden. In other cases, there may be more room for argument. However, the possibility of difficulties which may arise on different facts does not appear to be to be a good enough reason for not giving effect to what would otherwise be the right result as between the parties in this litigation.[98]I have considered whether this conclusion involves in reality, doing that which I have said is impermissible, namely not following Poulton. In terms of practical result, such a suggestion would be fair. I think it would follow from my reasoning that, in the absence of special facts, the revocation of a patent prior to the assessment of damages for its infringement could be relied on by the infringer in the Inquiry. However, in terms of principle, it appears to me that Poulton requires this court to accept that there is a res judicata, but subsequent developments in the law entitle the court to hold that, where justice between the parties requires it, and principle permits it, effect need not be given to the res judicata (save where it is a strict cause of action estoppel).[99]In these circumstances, while acknowledging that the point involves difficult issues of law and a difficult assessment of merits and implications, I have come to the conclusion that, in light of recent developments in the law of res judicata, it is open to this court to allow Stolt to rely upon the revocation of the Patent at the Inquiry, and that, in all the circumstances, we should take that course.[100]Damages here have not yet been assessed: indeed, the assessment hearing has not even started. Accordingly, the issue of what sum Stolt should pay Coflexip is very much still under the aegis of the court. It has been ordered that the Patent be revoked (although the order has been stayed pending appeal). So this is not a case where Stolt’s argument raises a mere hope or expectation. There is no suggestion of Stolt’s behaviour being worthy of any criticism, other than poor judgment, ineptitude or hard luck in failing to identify the decisive prior art which would have defeated the Patent. The amount of money for which Stolt would be liable, and which Coflexip would receive, would be wholly disproportionate to the beneficial effect of enforcing the estoppel. Over-riding the estoppel would avoid a five week hearing and would merely involve giving effect to a second decision on which everyone else could rely for all purposes.[101]This conclusion renders it unnecessary to consider the other two grounds upon which Stolt contend that Poulton should not apply, but, I shall deal with those points as they were fully argued, and as I understand that Peter Gibson LJ and Sir Martin Nourse take a different view from me on the issue I have been discussing.

The argument based on the amendment provisions of the 1977 Act

[102]Stolt contend that there are provisions, primarily relating to amendment, in the 1977 Act, which demonstrate that the law as laid down in Poulton is no longer correct.[103]The power to revoke a patent is contained in s72, and subsection (1) thereof sets out five grounds upon which a patent can be revoked. Section 72(4) provides:
“An order under this section may be an order for the unconditional revocation of the patent or, where the court … determines that one of the grounds mentioned in subsection (1) above has been established, but only so as to invalidate the patent to a limited extent, an order that the patent should be revoked unless within a specified period the specification is amended under section 75 below to the satisfaction of the court ….”
[104]Section 75 is headed “Amendment of Patent in Infringement or Revocation Proceedings”. Subsection (3) is in these terms:
“An amendment of the specification of a patent under this section shall have effect and be deemed always to have had effect from grant of the patent.”
[105]Finally, it is appropriate to refer to s62(3) , which is in these terms:
“Where an amendment of the specification of a patent has been allowed under any of the provisions of this Act, no damages shall be awarded in proceedings for an infringement of the patent committed before the decision to allow the amendment unless the court … is satisfied that the specification of the patent as published was framed in good faith and with reasonable skill and knowledge.”
[106]Stolt’s argument on this aspect is as follows. First, it is clear, both as a matter of common sense and in light of the provisions of s72(4) , that the effect of revocation and amendment, so far as infringers are concerned, should, at least in general, be consistent with each other. Secondly, it is clear from s75(3) that an amendment is retrospective in its effect to the date of the grant of the patent, and accordingly it follows that revocation of a patent should be retrospective in its effect. Thirdly, it follows from s75(3), and indeed from s 62(3), that, once an amendment is granted, the court is precluded from awarding damages for any infringement prior to the amendment being ordered. Fourthly, it therefore follows that if, in the present case, the second decision had resulted in the Patent being amended rather than revoked, damages would not have been recoverable from Stolt unless the conditions set out at the end of s62(3) were satisfied. Fifthly, consistency demands that, if, as was the case, the second judgment resulted in the revocation of the Patent, no damages should be recoverable from Stolt.[107]I am inclined to accept all but the fourth step of this five-step argument. If the fourth step in the argument is correct, then it would mean that, had the Patent been successfully amended as a result of the second judgment, Coflexip would have been unable to recover any damages as a result of Stolt’s infringement unless the court could be satisfied that the Patent “as published was framed in good faith and with reasonable skill and knowledge”. To put the point another way, if the court had not been so satisfied, then, even though (on this assumption) Stolt would have infringed the Patent as amended, Coflexip would not be entitled to recover any damages. If that is right, then, subject to one point, it would seem extraordinary, indeed perverse, that Coflexip should be able to recover damages for infringement from Stolt if the Patent were revoked. It would mean that a patentee was in a better position, and an infringer in a worse position, in relation to a patent which had been revoked than in relation to a patent which had been amended. Although I think the point is a good one, it is not quite as extraordinary as it at first appears, because the effect of s62(3) can only be to disallow damages: it does not disallow the alternative relief of an account of profits - see s61(1)(c) & (d) and (2).[108]However, as I have indicated, it appears to me that the problem in this line of argument is in the fourth step. The important point in this case, as I see it, is that, before the Patent was revoked, judgment was entered for damages. Thus, although the revocation of the Patent - like the amendment of a patent pursuant to s75(3) - is retrospective in its effect, the revocation - or amendment - does not impinge on an award of the damages already made by the court (unless, of course, the court is prepared to re-open the issue). The statement that “no damages shall be awarded” is concerned with an order for damages such as that made by Laddie J following the first judgment, and not with the subsequent machinery for carrying out that order or award, namely the assessment and any enforcement proceedings.[109]Furthermore, since this is said to be a ground for refusing to follow Poulton, it is legitimate to refer to what was said in that case by Fletcher-Moulton LJ at 437-8. “As regards the issues of validity and infringement the case had … terminated [with judgment], but there had to be an inquiry as to the amount of the damages. In point of fact, such an inquiry takes time, but as regards all legal consequences, it may be supposed to take place at the same instant as the determination of the other issue.”[110]Accordingly, I do not consider that there is anything in this ground for distinguishing, or not following, Poulton.

The Community Patent Convention

[111]The long title to the 1977 Act is in these terms, so far as relevant:
“An Act to establish a new law of patents …; to amend the law of patents …; to give effect to certain international conventions on patents; and for connected purposes.”
[112]The Conventions referred to are identified in s130(7) which is in the following terms:
“Whereas by resolution made on the signature of the Community Patent Convention the governments of the member states of the European Economic Community resolved to adjust their laws relating to patents so as (among other things) to bring those laws into conformity with the corresponding provisions of the European Patent Convention, the Community Patent Convention, and the Patent Co-operation Treaty, it is hereby declared that the following provisions of this Act that is to say, sections … 72(1) and (2), 74(4) …, are so framed as to have, as nearly as practical, the same effect in the United Kingdom as the corresponding provisions of the European Patent Convention, the Community Patent Convention, and the Patent Co-operation Treaty have in the territories to which those conventions apply.”
[113]The Community Patent Convention (“the Convention”) was finalised in its original form in 1975, but it was amended in 1989. Although it does not affect the arguments raised in relation to the Convention in this case, it appears to me that it is the 1989 version which applied in the present case, in light of s130(6) of the 1977 Act. This provides that any reference to the three conventions mentioned in s130(7) is to be treated as a reference to the relevant convention “as amended or supplemented”.[114]Stolt’s case hinges on the contention that Article 33 of the Convention is brought into English law by s130(7). Article 33 is in these terms:
“(1) A European patent application in which the Contracting States are designated and the resulting Community patent shall be deemed not to have had, as from the outset, the effects specified in this Chapter to the extent to the patent has been revoked. (2) Subject to the national provisions relating either to claims for compensation for damage caused by negligence or lack of good faith on the part of the proprietor of the patent or to unjust enrichment, the retroactive effect of the revocation of a patent as a result of opposition or revocation shall not effect [sic]: ‘(a) any decision on infringement which has acquired the authority of a final decision and been enforced prior to the revocation decision; (b) any contract concluded prior to the revocation decision, insofar as it has been performed before that decision ….”
[115]Although the point was not subject to any significant argument, I would, as at present advised, be inclined to accept Stolt’s submission that, if article 33 is part of English law, then its effect, on the present facts, would be to disentitle Coflexip from recovering damages, even if, in the absence of article 33, Coflexip would be entitled to recover such damages. Although it could be argued that the “decision on infringement” embodied in the first judgment not only “has acquired the authority of a final decision” (which must be right in light of the House of Lords dismissing the petition to appeal from the decision of the Court of Appeal), but has also “been enforced prior to the revocation decision”, such enforcement being through the means of the award of damages and the injunction. However, albeit in the absence of any significant argument, and without having been referred to authority, it seems to me that the concept of enforcement in article 33(2)(a), at least in the case of an award of damages, means actual recovery of damages. I believe that that view is supported by the natural meaning of the words, the reference to “unjust enrichment” in the opening part of article 33(2), and the purpose of article 33 read in its context.[116]On behalf of Coflexip, Mr Miller contends that Stolt’s case on article 33 of the Convention is misconceived, essentially for two reasons. The first reason, which found favour with Jacob J, is that article 33 was not within the scope of s130(7). The second argument, which Jacob J rejected, was that, as the Convention even now, has not been brought into effect in the United Kingdom, or in any other signatory country, s130(7) cannot have the effect, for which Mr Henderson contends. In my judgment, the first argument is correct, and, while I am more impressed with it than was Jacob J, the second argument need not be resolved.[117]So far as the first argument is concerned, although I accept that s130(7) is somewhat ambiguously worded, I consider that it is reading too much into the word “to have … the same effects” to contend that article 33 of the Convention is effectively carried into English law.[118]Section 72(1) sets out, as I have mentioned, five grounds upon which a patent may be revoked, and it begins in these terms:
“Subject to the following provisions of this Act, the court or the comptroller may on the application of any person by order revoke a patent for an invention on (but only on) any of the following grounds …”
[119]It is common ground that the grounds set out in s72(1) are the same as the grounds set out, albeit in slightly different language, and in a slightly different order, in article 56(1) of the Convention. It is also true that s72(4), which I have also set out above, and which is concerned with amendment, reflects the provisions of article 56(2) of the Convention. However, I do not read s130(7) as having the effect of saying that, because s72(1) has been “so framed as to have … the same effects … as the corresponding provisions” of the Convention, it must follow that revocation pursuant to s72 must have the same consequences under the Convention as revocation under article 56, and that one of those consequences is the application of article 33.[120]It appears to me that, when s130(7) provides that a section such as s72(1), is “so framed as to have … the same effects .. as the corresponding provisions” of the Convention, it is concerned with the meaning of the relevant section, and not with its consequences. In other words, although (for reasons best known to itself) the legislature, when framing s72(1), has used slightly different language from the Convention when describing the grounds of revocation, and has set out those grounds in a slightly different order, from article 56(1) of the Convention, s130(7) provides that, when interpreting s72(1), one should strive to do so in such a way as to achieve the same meaning as would be given to article 56(1).[121]In my view, this conclusion is supported by the fact that the legislature enacted s72(4). If s130(7) has the effect for which Stolt contend, then it would seem to me to follow that it would have been unnecessary to enact s72(4). Article 56(2) of the Convention provides:
“If the grounds for revocation affect the patent only partially, revocation shall be pronounced in the form of a corresponding limitation of the patent. The limitation may be effected in the form of an amendment to the claim, the description or the drawings.”
[122]If, as Stolt argue, the inclusion of s72(1) in s130(7) results in “the effect” of revocation engaging article 33, it would seem to me to follow that its “effect” would also be to engage article 56(2), in which case there was no need for the legislature to have enacted s72(4).[123]To my mind, the 1977 Act was intended to be a relatively complete code, and s130(7) was not intended to require parties, their lawyers, or the court to cross-refer to various conventions in order to discover, and argue about, whether certain consequences, not specifically provided for in the Act, were intended nonetheless to apply, simply because they were referred to in the Convention. It would have been only too easy for the legislature to have included a provision reflecting article 33, if it had thought it right to do so.[124]I turn to Mr Miller’s second argument. It is common ground that the Convention, even now, has not been ratified, and that it has never come into effect in any country. It also appears that, if it had come into effect, it would have been implemented pursuant to s86, which, in subsection (2), empowered the Secretary of State to make provision, through regulations, for implementing and giving effect to the Convention “and all remedies and procedures” provided for by or under it. As Mr Miller argued, if the Convention comes into force, it will give rise to a patent system “separate and distinct from the national system”. However, at the moment, there is no such supra-national system in existence.[125]In these circumstances, if the effect of s130(7) would be as Stolt contend, so that revocation of a patent pursuant to s72(1) would engage article 33 of the Convention, I consider that there is real force in the contention that could only arise if and when the Convention comes into force. It is one thing to say that provisions such as s72(1) should be construed consistently with the corresponding provision in the Convention, even if the Convention is not in force. It is quite another to suggest that a provision such as s72(1) should have the same consequences provided for in the Convention, as its equivalent provision in the Convention would have, even at a time when the Convention is not yet in force.[126]I revert to the centrally relevant words of s130(7) which provides that certain “provisions [in that Act] are so framed as to have … the same effects in the United Kingdom as the corresponding provisions of the [Convention] have in the territories to which [that] Convention … appl[ies].”[127]As a matter of ordinary language, given that the Convention has not been ratified, and has never come into effect, there are no “territories to which the Convention appl[ies].” There is therefore, in my judgment, real force in the point that unless and until the Convention is brought into force, it is scarcely possible to talk about a provision such as article 56(1) having “the same effects” in the sense for which Stolt contend, namely the consequences set out in article 33. As article 56(1) or its equivalent has “in the territories to which [the] Convention … appl[ies]”, in circumstances where the Convention does not apply in any territory at all.[128]In this connection, it should be mentioned that the effect of article 6(2) of the Convention (mirroring article 98 of the 1975 form) is that it can only come into effect when all the signatory states have ratified it.[129]I appreciate that the point found no favour with Jacob J on the basis that s130(7) “is an interpretation section”, and that the same view appears to have been taken by this court in Kimberley-Clark Worldwide Inc -v- Procter & Gamble Limited [2000] RPC 422 and Menashe Business Mercantile Limited -v- William Hill Organisation Limited [2003] RPC 575. However, it appears to me that, particularly if the word “effects” is given the wide meaning for which Stolt contends in the present case, the closing words of s130(7) indicate that, so long as the Convention is not in force, the subsection may well not apply. However, if the narrower meaning, which I believe to be correct, is given to the word “effects”, then it would appear to me to strengthen the contrary view reached by Jacob J, and, indeed, apparently assumed to be correct by the Court of Appeal. I should add that it appears clear that there was no argument on the point in either of the two cases before the Court of Appeal.[130]In all the circumstances, I do not consider that we should decide the second ground raised by Mr Miller for rejecting Stolt’s case based on the Convention. It is unnecessary to decide it, and the reason I have discussed it is to only explain why it impressed me rather more than it impressed Jacob J.[131]In these circumstances, I am of the view that the Convention does not assist Stolt in its argument that this court should not follow Poulton.

Conclusion

[132]In the event, I would allow Stolt’s appeal, on the basis that, although the decision in Poulton, is binding on this court as to the existence of an estoppel, it is open to us to decide, and we should decide, that effect should not be given to the estoppel which would prevent Stolt raising the revocation of the Patent at the Inquiry.[133]Between argument and judgment in this appeal, Coflexip’s appeal against the second judgment has been heard. If that appeal fails, I would have stayed the hearing of the assessment of damages in this action, pending the outcome of any application to the House of Lords in the other proceedings. If Coflexip’s appeal against the second judgment succeeds, then, in the light of the possibility of an application to the House of Lords, I would have invited submissions from the parties as to the appropriate course to take as to the Inquiry. In the event, however, in light of the view taken by Peter Gibson LJ and Sir Martin Nourse, whose judgments I have had the advantage of reading in draft, the issue with regard to the stay is a little different, and I would agree with the course they propose in relation to Stolt’s contention that there should be a stay of the Inquiry.[134]The following issues appear to us to be raised by this appeal and the Respondent’s Notice:(1) Are the Appellants (“Stolt”) estopped from relying on the revocation of the patent as a defence to the claim for damages on the inquiry as to damages? That issue gives rise to a number of sub-issues: (i) Is the decision of this court in Poulton v Adjustable Cover and Boiler Block Co. [1908] 2 Ch 430 one from which, subject to sub-issue (ii), we can and should depart? (ii) If Poulton was correctly decided at the time, (a) has the law then applicable been superseded by the Patents Act 1977 (“the 1977 Act”), and/or (b) has the approach to estoppel in 1908 been discarded in favour of a broad merits-based approach as suggested by the House of Lords in Johnson v Gore Wood & Co. [2002] AC1 (“Johnson”)? (iii) In relation to the 15 contracts made by Stolt, which are the subject of the inquiry, (a) does cause of action estoppel or issue estoppel apply, and, if the latter, (b) should Stolt be permitted to rely on the revocation as a defence to the claim for damages?(2) Do the provisions in the 1977 Act relating to revocation of a patent have the same effects by reason of s. 130 (7) of the 1977 Act as those provided for in Art. 33 of the Community Patent Convention (“the CPC”)? That issue gives rise to three sub-issues: (i) Is Art. 33 imported into the provisions in the 1977 Act relating to revocation by the words of s. 130 (7) “it is hereby declared that …. sections 72 (1) and (2) …. are so framed as to have, as nearly as practicable, the same effects in the United Kingdom as the corresponding provisions of …. [the CPC]”? (ii) Does the fact that the CPC has never been ratified and has never come into effect in any territory affect the applicability of Art. 33? (iii) Are the terms of Art. 33 (2)(a) satisfied in the present case so as to prevent the revocation having retrospective effect?(3) In any event should a stay of the inquiry be ordered until the validity of the patent is finally determined in the proceedings (“the Rockwater action”) brought by Rockwater Ltd. (“Rockwater”) against the Respondents Coflexip? (i) Is the decision of this court in Poulton v Adjustable Cover and Boiler Block Co. [1908] 2 Ch 430 one from which, subject to sub-issue (ii), we can and should depart? (ii) If Poulton was correctly decided at the time, (a) has the law then applicable been superseded by the Patents Act 1977 (“the 1977 Act”), and/or (b) has the approach to estoppel in 1908 been discarded in favour of a broad merits-based approach as suggested by the House of Lords in Johnson v Gore Wood & Co. [2002] AC1 (“Johnson”)? (iii) In relation to the 15 contracts made by Stolt, which are the subject of the inquiry, (a) does cause of action estoppel or issue estoppel apply, and, if the latter, (b) should Stolt be permitted to rely on the revocation as a defence to the claim for damages? (i) Is Art. 33 imported into the provisions in the 1977 Act relating to revocation by the words of s. 130 (7) “it is hereby declared that …. sections 72 (1) and (2) …. are so framed as to have, as nearly as practicable, the same effects in the United Kingdom as the corresponding provisions of …. [the CPC]”? (ii) Does the fact that the CPC has never been ratified and has never come into effect in any territory affect the applicability of Art. 33? (iii) Are the terms of Art. 33 (2)(a) satisfied in the present case so as to prevent the revocation having retrospective effect?[135]We consider these issues in turn.

(i) Poulton

[136]Subject to sub-issue (ii) next considered, it is not for this court to say that Poulton was wrongly decided.[137]In any event we have no reason to think that Poulton was wrongly decided, despite Mr. Henderson’s submissions to the contrary. The fact that Fletcher Moulton L.J. was wrong, as Jacob J. pointed out in para. 16 of his judgment, in expressing the view that an order for revocation only takes effect from the time of its pronouncement does not affect the ratio of this court’s decision, and his error was not shared by the other members of the court. Mr. Henderson sought to rely on what he described as the anomaly that once the patent is revoked, that puts an end to an injunction granted in the earlier proceedings but does not undermine the claim for damages in an inquiry held after the revocation. However the logic of the distinction seems to us defensible. It is that the final decision in the earlier proceedings established, in a way which the defendant could not be allowed to challenge in subsequent proceedings, the fact of the infringement of a valid patent before the decision and the consequent loss to the owner of the patent, the inquiry doing no more than quantify the damages that had been suffered up to the time of the judgment. As Fletcher Moulton L.J. said ([1908] 2 Ch. 430 at p. 439):
“In point of fact such an inquiry takes time, but, as regards all legal consequences, it may be supposed to take place at the same instant as the determination of the other issues.”
An injunction, whilst granted because of the findings as to past infringements which the defendant cannot subsequently challenge, is expressed to restrain the party enjoined from future infringements. When the patent is subsequently revoked the court will not allow the injunction to continue because there is no patent then subsisting capable of being infringed. That does not undermine the correctness of the decision.[138]Further, as Neuberger L.J. has pointed out, the correctness of Poulton does not appear to have been challenged in nearly a century, and it is noteworthy that in the 3rd ed. (1996) of Spencer Bower, Turner and Handley on Res Judicata it continues to be cited with approval. See, in particular, para. 168, where the remarks of Fletcher Moulton L.J. [1908] 2 Ch at p. 439 are cited:
“…. the inquiry as to damage in this case must …. proceed on the basis of the judgment already given upon the issue as to the validity of the patent and its infringement by the defendants, a judgment which binds the parties and which is not affected by the subsequent proceedings for revocation.”
See also para. 372 which firmly rejects the notion that a res judicata estoppel can be defeated by another estoppel or a cross-estoppel.

(a) Statutory changes

[139]Mr. Henderson draws attention to the long title to the 1977 Act, stating, as it does, that it is an Act to establish a new law of patents applicable to future patents and to amend the law of patents applicable to existing patents. He points to the fact that the 1977 Act provides expressly for the retrospective operation of an amendment of a patent. He says that the statutory provisions operative in 1908 have been superseded. No doubt they have, but it has not been shown that there has been a change in the statute law which materially affects the basis of the decision in Poulton. In this context we respectfully agree with what Neuberger L.J. has said in paras. 102 - 110 of his judgment.

(b) Johnson

[140]The real thrust of Mr. Henderson’s argument is that recent authorities show that estoppel, whether res judicata or the extension to it recognised in Henderson v Henderson (1843) 3 Hare 100, should only be allowed where to do so would prevent abuse of process, and should not be allowed where injustice would result.[141]The classic statement of res judicata and the two species of estoppel embraced by that term is that of Diplock L.J. in Thoday v Thoday [1964] P 181 (at pp. 197 – 8):
“The particular type of estoppel relied upon by the husband is estoppel per rem judicatam. This is a generic term which in modern law has two species. The first species, which I will call “cause of action estoppel”, is that which prevents a party to an action from asserting or denying, as against the other party, the existence of a particular cause of action, the non-existence or existence of which has been determined by a court of competent jurisdiction in previous litigation between the same parties. If the cause of action was determined to exist, i.e. judgment was given upon it, it is said to be merged in the judgment, or, for those who prefer Latin, transit in rem judicatam. If it was determined not to exist, the unsuccessful plaintiff can no longer assert that it does; he is estopped per rem judicatam. This is simply an application of the rule of public policy expressed in the Latin maxim “Nemo debet bis vexari pro una et eadem causa.”
In this application of the maxim “causa” bears its literal Latin meaning. The second species, which I will call “issue estoppel”, is an extension of the same rule of public policy. There are many causes of action which can only be established by proving that two or more different conditions are fulfilled. Such causes of action involve as many separate issues between the parties as there are conditions to be fulfilled by the plaintiff in order to establish his cause of action; and there may be cases where the fulfilment of an identical condition is a requirement common to two or more different causes of action. If in litigation upon one such cause of action any of such separate issues as to whether a particular condition has been fulfilled is determined by a court of competent jurisdiction, either upon evidence or upon admission by a party to the litigation, neither party can, in subsequent litigation between one another upon any cause of action which depends upon the fulfilment of the identical condition, assert that the condition was fulfilled if the court has in the first litigation determined that it was not, or deny that it was fulfilled if the court in the first litigation determined that it was.”[142]Under the rule in Henderson the doctrine of res judicata is extended to cover not only matters which have been the subject of judgment between the parties but also “to every point which properly belonged to the subject of litigation and which the parties, exercising reasonable diligence, might have brought forward at the time” (per Sir James Wigram V.-C. 3 Hare at p. 115).[143]The statement of the rule in Henderson was expressly qualified by exceptions for special circumstances and special cases. Other subsequent cases recognise that this extension of the rule of res judicata operates only where it becomes an abuse of process to raise in subsequent proceedings what should have been litigated in earlier proceedings (see, for example, Greenhalgh v Mallard [1947] 2 All ER 255 at p. 257 per Somervell L.J.).[144]In Arnold v National Westminster Bank plc [1991] 2 AC 93 the House of Lords held that issue estoppel was subject to a limited exception. As Lord Keith, with whom Lord Griffiths, Lord Oliver and Lord Jauncey agreed, said at p. 109:
“In my opinion your Lordships should affirm it to be the law that there may be an exception to issue estoppel in the special circumstance that there has become available to a party further material relevant to the correct determination of a point involved in the earlier proceedings, whether or not that point was specifically raised and decided, being material which could not by reasonable diligence have been adduced in those proceedings. One of the purposes of estoppel being to work justice between the parties, it is open to courts to recognise that in special circumstances inflexible application of it may have the opposite result.”
[145]However it is noticeable that Lord Keith was careful to confine his reasoning to issue estoppel. He said at p. 104:
“It is appropriate to commence by noticing the distinction between cause of action estoppel and issue estoppel. Cause of action estoppel arises where the cause of action in the later proceedings is identical to that in the earlier proceedings, the latter having been between the same parties or their privies and having involved the same subject matter. In such a case the bar is absolute in relation to all points decided unless fraud or collusion is alleged, such as to justify setting aside the earlier judgment. The discovery of new factual matter which could not have been found out by reasonable diligence for use in the earlier proceedings does not, according to the law of England, permit the latter to be re-opened.”
[146]As we see it, therefore, in Arnold the House of Lords was carefully distinguishing between cause of action estoppel and issue estoppel, the former allowing of no exceptions save for fraud or collusion, the latter allowing a limited exception when further material is found which could not have been adduced by reasonable diligence in the earlier proceedings.[147]In Johnson the House of Lords considered whether an action should be struck out as an abuse of process under the rule in Henderson. Lord Bingham referred to Arnold as a case of issue estoppel and cited the passage from Lord Keith’s judgment which we have already quoted in para. 145 above. He accepted that Henderson abuse of process was separate and distinct from cause of action estoppel and issue estoppel, but said that it had much in common with them in that there should be finality in litigation and that a party should not be twice vexed in the same matter. He continued at p. 31:
“The bringing of a claim or the raising of a defence in later proceedings may, without more, amount to abuse if the court is satisfied (the onus being on the party alleging abuse) that the claim or defence should have been raised in the earlier proceedings if it was to be raised at all…. It is, however, wrong to hold that because a matter could have been raised in earlier proceedings it should have been, so as to render the raising of it in later proceedings necessarily abusive. That is to adopt too dogmatic an approach to what should in my opinion be a broad, merits-based judgment which takes account of the public and private interests involved and also takes account of all the facts of the case, focusing attention on the crucial question whether, in all the circumstances, a party is misusing or abusing the process of the court by seeking to raise before it the issue which could have been raised before.”
[148]Mr. Henderson fastens on Lord Bingham’s recognition of what the two types of estoppel have in common with Henderson abuse of process and submits that the same flexible approach should apply to all of them. He argues that both the policy justifications for res judicata as well as for Henderson abuse of process told in his favour on the facts of this case because if Stolt are allowed to rely on the revocation of the patent, that would bring the litigation to an end, as the revocation, being retrospective, would be a complete defence to Coflexip’s claim for damages. Further, he says, to rely on the revocation would not be to reargue a point raised at the trial and determined against Stolt, and so Coflexip would not be vexed twice in the same matter.[149]We do not think it open to this court to accept these submissions, given Lord Keith’s careful distinction in Arnold between cause of action estoppel and issue estoppel, in the absence of an allegation of fraud or collusion. The House of Lords in Johnson, whilst fully alive to all that was said in Arnold, did not cast doubt on what Lord Keith said on cause of action estoppel or on the limited exception to issue estoppel.[150]We would add that we do not agree that the policy justifications tell in Stolt’s favour. The desideratum of finality in litigation is not achieved by allowing a final decision on the validity of a patent and its infringement to be, in effect, reopened if another person obtains a decision on the invalidity of the patent, with a consequent order for revocation, and if that revocation can be relied on in an inquiry as to damages in the first action. The patent owner would be vexed twice. In the first action he had to establish that the patent was valid and subsisting at the time of the infringement and had to resist any counterclaim for revocation. On the inquiry he would be faced with the claim that the patent was not valid and subsisting at the time of the infringement because of the in rem effect of the revocation of the patent. True it is that the order (albeit stayed) for revocation in the present case was obtained because of prior art not relied on by Stolt, but in a real sense Coflexip would be vexed twice in the same matter in relation to the validity and subsistence of the patent. Furthermore it is a double vexation arising from Stolt’s failure to find and adduce the prior art. Mr. Miller told us that the Recalde prior art which defeated the patent in the Rockwater action was always there to be found, that evidence was adduced by Coflexip before this court on the appeal from Laddie J.’s judgment of 29 January 1999 that an independent professional searcher instructed by Coflexip and searching for prior art against the patent had produced a list of prior art items which included Recalde, that a copy of that list, which was put before this court by Stolt on that appeal, showed by the markings thereon that Stolt had noted the inclusion of Recalde on that list, and that Stolt, whilst applying to amend their pleadings and to adduce further evidence relating to further prior art, chose not to rely on Recalde. Mr. Miller submitted that it was as a consequence of Stolt choosing not to search further that Recalde was not found. He drew attention to the fact that it is not asserted by Stolt on this application that they could not have found Recalde by reasonable diligence for use at the trial. Mr. Henderson in response said that Stolt had employed experienced and qualified agents to effect a search but that unfortunately they failed to find Recalde. That does not show that Stolt could not have found Recalde with reasonable diligence, and in this context we refer to what Aldous L.J. said in his judgment, cited by Neuberger L.J. at para. 5 of his judgment, that whilst Stolt carried out searches which they believed were appropriate and reasonable, it was their decision not to search more widely and for that reason it was not right to allow Stolt to reopen their case. Chadwick L.J. noted (in para. 72 of his judgment) that Stolt accepted that more extensive searches would have revealed the additional prior art on which they then sought to rely and that it could not be disputed that the additional material was there to be found. In our opinion Stolt have failed to show on this appeal that they could not with reasonable diligence have found Recalde and adduced it at the trial before Laddie J.

(a) Type of estoppel in this case

[151]We did not understand Mr. Henderson to dispute that, subject to his other arguments, cause of action estoppel applied to stop Stolt from asserting that no loss resulted from the infringement of the patent when Stolt performed the contract relating to the Magnus Swift field, that contract alone being the subject of the pleadings in this case and of the determination by Laddie J. and this court in the this action.[152]In accordance with what we understand to be the usual and convenient practice of the Patents Court of including in an inquiry other indistinguishable claims for damages for infringement of the patent against the same infringer even though the infringements were in circumstances which have not been pleaded, the inquiry will also consider what damages were suffered by Coflexip through Stolt using the same process and product in performing 14 other contracts. In all of them the same equipment and the same vessel, the Seaway Falcon, were used. Mr. Miller submits that cause of action estoppel applies in respect of all of the contracts, judgment having been entered against Stolt for keeping the equipment. Mr. Henderson says that there is no decision on infringement in the performance of those contracts and that there is a different cause of action in relation to each.[153]We agree with Mr. Henderson on this point, technical though it is. The 14 contracts other than that relating to the Magnus Swift field give rise to different causes of action. However, by issue estoppel Stolt would be estopped from asserting in the inquiry that the patent was not valid and subsisting at the time of the infringement unless Stolt are to be permitted to rely on the revocation.

(b) Should Stolt be permitted to rely on the revocation

[154]In relation to the contract relating to the Magnus Swift field, Poulton, which in our judgment is indistinguishable, compels a negative answer to that question. We respectfully agree with Jacob J.[155]The judge did not advert to the fact that there were 14 contracts other than that which was the subject of the pleadings. He treated the case as raising only a question of cause of action estoppel.[156]Mr. Henderson submitted that it would be an affront to justice and an abuse if in the inquiry the court were to give effect to the revoked patent and complete the incomplete judicial business of the quantification of loss on the false premise that the patent was not revoked and if up to £80 million of damages were awarded for infringement of a patent which neither is nor was valid. He argued that he should be allowed to say in the inquiry what Stolt had not said before, that the patent is invalid because of its revocation in other proceedings. Looking at the matter in the round, he said, and not applying the law of res judicata in a mechanistic way, the court should hold Stolt not to be estopped from relying on the revocation.[157]Mr. Miller submits that in respect of the 14 contracts there are no special circumstances entitling Stolt to reopen the issues of validity and subsistence of the patent which were decided at the trial. He says that as between Coflexip and Stolt the patent has been found valid and subsisting and that finding binds both Coflexip and Stolt in personam for all time. It is, he submits, irrelevant whether the decision, being final, was right or wrong, and the subsequent finding of invalidity and the order of revocation in other proceedings not involving Stolt do not displace the in personam estoppel which is mutual between Coflexip and Stolt. Further, Mr. Miller points out that, not only have Stolt had their opportunity to challenge the validity of the patent and failed, but also the prior art, which founded the finding of invalidity in the Rockwater action, was available to be discovered if Stolt had searched for it fully at or before the trial. As Aldous L.J. indicated in para. 55 of his judgment of 31 July 2000 (which Neuberger L.J. has quoted at para. 5 of his judgment), Stolt took the deliberate decision not to search more widely.[158]We agree with Mr. Miller. It has been repeatedly stated that where issue estoppel applies, the parties to a suit cannot advance argument or adduce further evidence directed to showing that the issue was wrongly determined where the fresh material could with reasonable diligence have been adduced when the issue was determined. We have already cited the statements in Arnold to that effect, which were based on earlier decisions of this court such as Fidelitas Shipping Co. Ltd. v V/O Exportchleb [1966] 1 QB 630 and Mills v Cooper [1967] 2 QB 459. Arnold was a case decided on exceptional facts. Walton J., in holding that in a dispute on a rent review in a lease an arbitrator was wrong in his construction of the rent review provisions, had himself erred, as decisions of this court later showed, but had not permitted the point to be taken to an appeal. The House of Lords held that justice required that the tenants be permitted to reopen the question of construction in respect of later rent reviews. The circumstances of the present case are very different and there is no question but that the decision of Laddie J. in this action was correct on the material put before him.[159]Johnson was not concerned with an issue which had been raised and decided in a previous action, still less with an issue raised and decided in the same action, as in the present case. It was concerned with whether an issue which could have been raised in an earlier action should be allowed to be raised in a later action. We do not see that what was said in Johnson was addressed to a case such as the present. However, even on the more flexible, merits-based approach suggested by Johnson we see no sufficient reason for saying that issue estoppel should not bind Stolt in respect of infringements already committed by them by the time of the order directing the inquiry, given that Stolt had failed to show that they could not with reasonable diligence have adduced Recalde. In the circumstances it is an abuse of process, after the validity of the patent was finally decided as between Coflexip and Stolt, to seek to go back on that decision.[160]We, of course, take note of the very substantial sum claimed by Coflexip in the inquiry, but against the private interest of Stolt must be set the public interest in finality of litigation. Mr. Henderson sought to rely on the extension by the 1977 Act of the classes of person who could seek an order for revocation, s. 72 now permitting any person to apply (subject to a limited exception in s. 72 (2)(a) in relation to an application for revocation on the ground that the patent was granted to a person not entitled to be granted the patent). He argued therefrom that the policy must be that there could be multiple litigation in relation to a patent, that the revocation of a patent operates against the patentee in favour of the whole world and that the validity of a patent can be repeatedly challenged by different parties. We do not see that these points seriously detract from the public interest in the finality of litigation between the parties to it, nor from the public policy that a party should not be vexed twice for the same cause.[161]We also agree with the policy point taken by Jacob J. in para. 35 of his judgment:
“Moreover there is this consideration concerning policy. If a defendant, having finally lost a patent action, knew that an application for successful revocation by another might get him off the hook of damages, he would have every motive for digging up better prior art and encouraging another to attack the patent. So although he himself could not re-litigate validity, one of the heads of public policy (re-litigation) behind the rule could be subverted indirectly.”
[162]For these reasons we conclude that Stolt should not be permitted to rely on the revocation of the patent in the inquiry as to damages. The CPC (i) s. 130 (7)[163]We respectfully agree with what Neuberger L.J. has said in paras. 117 – 123 of his judgment. It seems to us that the draftsman, by s. 130 (7), has explained why the particular provisions to which he refers in the 1977 Act “are so framed”, viz. to have, as nearly as practicable, the same effects in the U.K. as “the corresponding provisions” of the CPC have in the territories to which it applies. Those particular provisions must therefore be construed to have the same meaning as the corresponding provisions in the CPC (if it does apply in any other territory). It does not mean that because s. 72(1) of the 1977 Act deals with the grounds of revocation and s. 72(2) and s. 74 (4) with who may apply for revocation, any provision of the CPC which provides for consequences under the CPC of revocation has effect, regardless of whether that provision corresponds to s. 72 (1) and (2) or s. 74 (4). That would go beyond what is stated in s. 130 (7). (ii) Non-ratification of CPC[164]We, like Neuberger L.J., see considerable force in the contention of Mr. Miller that as there is no territory to which the CPC applies, s. 130 (7) does not in this respect have application; but we need not decide that point in view of our conclusion on sub-issue (i). (iii) Art.

(a) of the CPC

[165]Again, it is unnecessary for us to decide whether Mr. Miller is right to argue that the terms of Art. 33 (2)(a) of the CPC are in any event satisfied in the present case so as to present the revocation having retrospective effect under the CPC. (3) Stay[166]Mr. Henderson boldly argued that even if this court decided the other issues against him, we should order a stay. He says that it is likely that whatever is decided by the constitution of this court hearing the appeal in the Rockwater action and whatever we decide on this appeal, the loser in each case will seek to appeal to the House of Lords. He asks that until these matters are finally decided, the inquiry should be stayed, though he accepts that the preparations for the inquiry, which are now in an advanced state, should be completed.[167]We must decide whether to order a stay in the light of the conclusions reached on this appeal. Like Jacob J. we are of the opinion that there is no point in ordering a stay. If the appeal in the Rockwater action is dismissed, the position on the inquiry is left unaffected. If the appeal is allowed, the basis for Stolt’s argument disappears. If Stolt apply for permission to appeal to the House of Lords and seek a stay, pending the determination of the appeal, we will deal with that application then. In the meantime, we have not looked at the further evidence which Coflexip wish to adduce on this point nor have we decided whether to admit that evidence. Conclusion[168]For these reasons we would dismiss this appeal.

Order

Appeal dismissed; consequential matters to be dealt with at a hearing on 19th March 2004. (Order does not form part of the approved judgment)

Cited in 17 later judgments

and 8 more of the most senior