Takhar v Gracefield Developments Limited and others

Balber Kaur Takhar, the appellant, is the cousin of the third respondent, Parkash Kaur Krishan. For many years before 2004, they had not seen each other. In that year they became reacquainted. At the time, Mrs Takhar was suffering personal and financial problems. She had separated from her husband some five years previously. As part of the arrangements made between Mrs Takhar and her husband, she had acquired a number of properties in Coventry.

When Mrs Takhar and Mrs Krishan met again, according to Mrs Takhar, she confided in her cousin and grew increasingly to depend upon her. Mrs Takhar claims that Mrs Krishan exerted considerable influence over her.

The financial problems of Mrs Takhar arose mainly from the condition of the properties which she had acquired from her husband. Some were in a dilapidated condition. Payment for rates were in arrears. Bankruptcy for Mrs Takhar was in prospect.

The Krishans provided financial help to Mrs Takhar. Dr Krishan, the second respondent and the third respondent’s husband, took on responsibility for negotiating with Coventry City Council over the rates arrears and the dilapidated state of some of the buildings. Then, in November 2005 it was agreed that the legal title to the properties would be transferred to Gracefield Developments Ltd, a newly formed company, of which Mrs Takhar and the Krishans were to be the shareholders and directors.

Mrs Takhar claims that it had been agreed between her and the Krishans that the properties would be renovated and then let. The rent would be used to defray the cost of the renovation, which, in the short term, would be met by the Krishans. Mrs Takhar would remain the beneficial owner of the properties.

The Krishans present a very different account. They claim that Gracefield was set up as a joint venture company. The properties were to be sold after they had been renovated. They were to be given an agreed value and this would be paid to Mrs Takhar after they had been sold. Any profit over would be divided equally between Mrs Takhar and the Krishans. They explain that Mrs Takhar agreed to these arrangements because planning permission for development had to be obtained in order to realise the value of the properties and this was an area in which Dr Krishan had experience, having already successfully developed his own medical centre.

On 24 October 2008, Mrs Takhar, issued proceedings in the Birmingham District Registry of the Chancery Division. She claimed that the properties had been transferred to Gracefield as a result of undue influence or other unconscionable conduct on the part of Dr and Mrs Krishan. In a judgment delivered on 28 July 2010 [2010] EWHC 2872 (Ch), His Honour Judge Purle QC rejected that claim.

A significant item of evidence in the hearing before Judge Purle was a written profit share agreement dated 1 April 2006. It provided for an initial purchase price of £100,000 for the properties. This was to be placed on a loan account with Gracefield. Further sums totalling £200,000 as deferred consideration were also provided for. The total of £300,000 was to be paid to Mrs Takhar on completion of the sale of the properties. She was also to receive 50% of the profits on the sale of each property.

The circumstances in which this written agreement was discovered and Mrs Takhar’s evidence about it were described by Judge Purle in paras 21 and 22 of his judgment:

This was, therefore, powerful evidence in support of the Krishans’ case. And it is unsurprising that it was heavily relied on by the judge. As the quoted passage shows, he found that the written agreement represented what had earlier been agreed orally between Mrs Takhar and the Krishans. The judge therefore held that Mrs Takhar had transferred the properties to Gracefield for the sum of £300,000 and that she was to receive 50% of the profits when the properties were sold. He dismissed Mrs Takhar’s claim based on undue influence or unconscionable bargain and held that the properties had been transferred to Gracefield both legally and beneficially. That transfer was, he held, subject to the terms of the oral agreement made between the parties, as reflected in the written profit share agreement.

The original of the profit share agreement said to have been signed by Mrs Takhar has not been found. The Krishans claim that it was prepared by accountants at a time when Mrs Takhar was in India and then handed to her when she returned. She was asked to consider it and return it to the accountants. Mrs Takhar’s case is that she did not sign the document and had never seen it until the dispute arose. The authenticity of the document and whether it had been signed by Mrs Takhar are central issues in the dispute between the parties, therefore.

In advance of the trial before Judge Purle, Mrs Takhar had sought permission to obtain evidence from a handwriting expert to examine the signature on the profit share agreement which had been attributed to her. That application was refused because it had not been made until the trial was imminent. On the trial, Mrs Takhar gave evidence that she could not say that the signature on the profit share agreement was not hers, but she was unable to explain how it had got there.

After the trial Mrs Takhar instructed new solicitors and asked them to obtain a report from a handwriting expert. Robert Radley is such an expert and he was engaged to inspect and report on various documents. His subsequent report stated conclusively that the signature on the profit share agreement which purported to be that of Mrs Takhar had been transposed from a letter of 24 March 2006 which she had sent to the Krishans’ solicitors. He was also of the opinion that there was strong evidence that Mrs Takhar did not sign a 2006 bank inquiry form and that the signatures of both the Krishans and Mrs Takhar on later 2011 bank inquiry forms had also been transposed from previous forms.

On foot of this report, Mrs Takhar claims that she can now advance a case of fraud against the Krishans. She also claims that she was not in a position to do so until she had received Mr Radley’s report. The Krishans dispute both claims.

After receiving Mr Radley’s report, Mrs Takhar issued proceedings in which she sought to have Judge Purle’s judgment and order set aside. She claimed that she was entitled to this relief on the ground that it was obtained by fraud, the principal forgery relied upon being that of the copy of the profit share agreement. (Later Mrs Takhar applied for permission to amend her claim to allege an unlawful means conspiracy and deceit. That application was refused and no longer features in the appeal.)

The respondents served defences in which they pleaded that Mrs Takhar’s claim is an abuse of process, inter alia because the documents on which Mr Radley’s report was based were available to Mrs Takhar and her legal team since at least 12 July 2009 (approximately 12 months earlier than the trial before Judge Purle).

It was ordered that the question whether Mrs Takhar’s claim amounted to an abuse of process be tried as a preliminary issue. That trial took place before Newey J in February 2015. In his judgment, ([2015] EWHC 1276 (Ch)), Newey J held that a party who seeks to set aside a judgment on the basis that it was obtained by fraud did not have to demonstrate that he could not have discovered the fraud by the exercise of reasonable diligence. The present claim was therefore not an abuse of process.

The respondents appealed. The Court of Appeal (Patten, King and Simon LJJ) allowed the appeal in its judgment delivered on 21 March 2017 ([2018] Ch 1; [2017] 3 WLR 853; [2017] CP Rep 23). Patten LJ, delivering the leading judgment, said at para 30 that the appeal turned on whether Newey J was correct in holding that a due diligence condition did not need to be satisfied.

Patten LJ began his examination of that issue by a reference to the judgment of Wigram V-C in Henderson v Henderson (1843) 3 Hare 100. In that case the Vice Chancellor had said that where a matter had been the subject of litigation and adjudication by a court, it was required of the parties that they “bring forward their whole case” (at p 115). When litigation had taken place, it would only be in exceptional circumstances that parties could “open the same subject of litigation in respect of matter[s] which might have been brought forward as part of the subject in contest, but which [had not been] … brought forward” because of “negligence, inadvertence, or even accident”. This applied 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”.

Henderson is certainly authority for the general principle that parties must normally advance the totality of their case on the first bout of litigation. It is not open to them, save in exceptional circumstances, to bring up a point which should have been raised in that litigation and which could, with reasonable diligence, have been discovered and canvassed on the first trial. Henderson does not speak, however, on two subjects which are critical in the present case. The first of these is whether the rule applies where the new point was not in issue between the parties on the first trial and where, if it had been and evidence on the point had been led, a different outcome might have ensued. The second subject concerns the question whether the rule in Henderson requires modification or disapplication where the new issue raises an allegation of fraud by which, it is claimed, the original judgment was obtained.

The second case on this subject referred to by Patten LJ was Virgin Atlantic Airways Ltd v Zodiac Seats UK Ltd (formerly Contour Aerospace Ltd) [2014] AC 160. In his judgment in that case, Lord Sumption rejected the suggestion that the principle propounded in Henderson should be treated as applying only to the law governing abuse of process and not as an application of the doctrine of res judicata. At para 26, Lord Sumption had said that, “[w]here the existence or non-existence of a cause of action has been decided in earlier proceedings, to allow a direct challenge to the outcome, even in changed circumstances and with material not available before, offends the core policy against the re-litigation of identical claims.” I do not consider that Lord Sumption’s statement in this passage has any bearing on the issues which arise on this appeal. The “existence or non-existence” of fraud has not been decided in the proceedings before Judge Purle. It is a new issue. It does not involve the re-litigation of an identical claim.

In the Virgin Atlantic case, Lord Sumption had considered the House of Lords decision in Arnold v National Westminster Bank plc [1991] 2 AC 93. In that case, Lord Keith had drawn a distinction between cause of action estoppel and issue estoppel. At p 104, Lord Keith had described cause of action estoppel in this way:

He considered that a distinction between this and issue estoppel should be recognised and said, at p 109:

At para 35 of the judgment in the present appeal, Patten LJ said this about Lord Keith’s speech in Arnold:

It is important to note that Lord Keith, in the first passage quoted, at para 22 above, (dealing with cause of action estoppel), does not suggest that there is a reasonable diligence requirement in cases of fraud or collusion. And in the second passage, at para 23 above, (which concerns issue estoppel) he does not mention fraud at all. In my opinion, it is not to be assumed that Lord Keith was suggesting that due diligence was a prerequisite in cases of fraud.

Patten LJ had referred to Lord Sumption’s judgment in Virgin Atlantic, para 22, as supporting the proposition that a fresh point could not be relitigated unless the new material could not with reasonable diligence have been produced at the earlier hearing. It is important to note exactly what Lord Sumption said at para 22 in Virgin Atlantic:

The significance of this paragraph, so far as concerns the present case, lies in what it does not assert and the qualifications which it contains. It does not address the question of fraud at all. Moreover, the first proposition which Lord Sumption makes, that cause of action estoppel is an absolute bar in relation to such points as had to be and were decided in order to establish the existence or non-existence of a cause of action, is of no relevance at all in the present appeal. The case before Judge Purle did not involve an allegation of fraud on the part of the Krishans. The points which “had to be or were decided” in Mrs Takhar’s case before Judge Purle were not concerned with possible fraud.

The second proposition in Lord Sumption’s para 22, that cause of action estoppel prohibits the raising of points in subsequent proceedings which had not been decided because they were not raised in the earlier proceedings, must be read in context. Again, the issue of fraud is not mentioned. And Lord Sumption has expressly espoused the reasoning of Lord Keith in Arnold where reasonable diligence is not said to be a requirement in cases of fraud or collusion. Given that, I do not consider that the second proposition in para 22 of Virgin Atlantic should be interpreted as covering cases of fraud.

As to issue estoppel and the third proposition in para 22, it is clear that Lord Sumption did not prescribe a universally applicable rule. He was careful to qualify his statement (that issue estoppel bars the raising in subsequent proceedings of points which were not raised in the earlier proceedings or were raised but unsuccessfully), by saying that the bar will usually be absolute if it could with reasonable diligence and should in all the circumstances have been raised. And, of course, there is again no mention of fraud in this context. I do not consider, therefore, that Virgin Atlantic is authority for the proposition expressed by Patten LJ.

In any event, at para 37, Patten LJ accepted the submission of Mr Wardell QC (who appeared on behalf of Mrs Takhar before the Court of Appeal and this court) that the “obstacle faced by Mrs Takhar was not one of issue estoppel”. He also accepted that she was not challenging an earlier decision of the judge about the authenticity of the profit share agreement. As Patten LJ observed, “[t]he allegation about her signature being forged was not raised or decided at the trial of the 2008 action.”

But the Court of Appeal considered that, although it was not concerned with a question of res judicata in the strict sense of issue or cause of action estoppel, it had to deal with what it described, at para 37, as “the wider policy considerations embodied in the rule in Henderson v Henderson”. Those policy considerations were engaged, Patten LJ said, “whenever a litigant seeks to challenge an earlier decision of a competent court, whether directly or indirectly, by commencing new proceedings in which the same issues arise or seeks directly by way of appeal to challenge the judge’s decision on the basis of new evidence.”. On that account, Patten LJ concluded, “If the challenge … is by way of an action seeking to set the judge’s order aside on the ground that it was obtained by fraud the real question … is whether this amounts to an abuse of process if the success of the action depends upon evidence which could with reasonable diligence have been produced at the earlier trial.”

I do not agree with this conclusion. In the first instance, this is not a case of commencing new proceedings where the same issues arise. As Patten LJ had earlier said, the question of Mrs Takhar’s signature having been forged had not been raised or decided in the trial before Judge Purle. The appellant does not seek to set aside Judge Purle’s decision on any of the issues decided by him. Secondly, for the reasons that I have given, I do not consider that in Arnold or Virgin Atlantic there has been an unequivocal judicial statement that seeking to set aside a judgment on the basis that it was obtained by fraud constitutes an abuse of process, if evidence of the fraud could, with reasonable diligence, have been obtained and produced at the earlier trial.

I accept, however, that the question whether fraud should “unravel all” even where discovery of its existence was possible before the original trial does give rise to intensely relevant policy considerations. These are considered in the next section of this judgment.

In para 38 Patten LJ referred to the case of Phosphate Sewage Co Ltd v Molleson (1879) 4 App Cas 801. It had been argued in that case that evidence of fraud other than that presented at the original trial should be allowed to enable a company to claim repayment of a sum which it had paid to a bankrupt. The company had subsequently obtained further evidence of the fraud. It sought to advance a new case founded on that evidence. The House of Lords held that the new allegations of fraud were based on facts within the company’s knowledge at the time of the first trial. The plea of res judicata succeeded, therefore. Importantly, Earl Cairns LC said, at p 814, that it would be “intolerable” if a party who had been unsuccessful in litigation could re-open it merely because,

The contrast with the present case is immediately obvious. This is not an instance of the appellant seeking to adduce evidence of facts “going in the same direction” as facts previously stated, because Mrs Takhar had not asserted that the Krishans had been guilty of fraud, merely that she had no recollection of having signed the profit share agreement. The relief that she seeks now is quite different from that which she had earlier claimed. Previously, she sought to avoid the effect of the agreement because of undue influence and unconscionability on the part of the Krishans. Now she claims that the agreement on which they rely was, in its written form, a forgery.

Now, it is true that Earl Cairns had also said in the Phosphate Sewage case, at p 814, that “the only way in which [new evidence] could possibly be admitted would be if the litigant were prepared to say, I will shew you that this is a fact which entirely changes the aspect of the case, and I will shew you further that it was not, and could not by reasonable diligence have been, ascertained by me before.” But the essential context of this observation is set by the earlier passage quoted above. It is where precisely the same relief as had previously been claimed is sought again. In my view, it is not appropriate to lift the requirement of reasonable diligence out of the context in which it appears and to import it into a different scenario, namely, where a changed basis for success for the appellant is advanced.

Patten LJ acknowledged that Earl Cairns’ statement of principle was expressed in relation to a plea of res judicata and that that was how it had been treated by Lord Keith in Arnold (para 39, per Patten LJ). But he suggested that the rule expressed by Earl Cairns was said by Lord Diplock in Hunter v Chief Constable of the West Midlands Police [1982] AC 529 to be “relevant more generally as a statement of the conditions which must be satisfied to justify a departure from the policy of not permitting a party to challenge prior decisions by a court of competent jurisdiction in which case it is relevant to whether it would be an abuse of process to seek to set a judgment aside without satisfying the reasonable diligence condition.” (Patten LJ, also at para 39.)

Again, it is important to recognise the context in which Lord Diplock made his remarks about the application of the rule. Hunter was a case in which those who had been convicted of planting bombs in Birmingham sought to establish, by civil action, that they had been subjected to ill-treatment before they confessed to involvement in the bombings. This was described as a collateral attack on the correctness of their convictions. At p 545 Lord Diplock said:

The context of the issue in Hunter is therefore firmly set. It is whether a challenge to the correctness of a criminal conviction amounts to an abuse of process. This is, self-evidently, an entirely different situation from that which arises in the present case. And it immediately gives rise to concern about whether it is acceptable to apply a test fashioned for the circumstances of collateral challenge to a criminal conviction to the markedly dissimilar setting of seeking to set aside a civil judgment because it was obtained by a fraud which had not been alleged or adjudicated on at the original trial.

Patten LJ referred to the decision of the Privy Council in Owens Bank Ltd v Etoile Commerciale SA [1995] 1 WLR 44, (“Etoile”) where Lord Templeman had said:

Newey J’s criticism of Lord Templeman’s statement of principle was rejected by Patten LJ. He suggested that this was consistent with the earlier decision of the House of Lords in Owens Bank Ltd v Bracco [1992] 2 AC 443 (Bracco). In that case Lord Bridge, at p 483, had articulated “the common law rule” in this area as being:

The rule is therefore expressed to apply when there is a proposed re-litigation of the issue of fraud which had been determined in the earlier litigation. That is not the position here. Mrs Takhar seeks to raise the fraud of the Krishans for the first time. It should be noted that Lord Bridge referred to re-litigation of the issue which had been determined against the party seeking to reopen the case. The issue of the Krishans’ alleged fraud has not been determined. I do not consider, therefore, that Bracco forbids a challenge to a judgment which, it is claimed, was obtained by fraud, where that issue was not canvassed at the first trial. Likewise, it does not prohibit Mrs Takhar from pursuing her present action.

In HIH Casualty and General Insurance Ltd v Chase Manhattan Bank [2003] 2 Lloyds Rep 61, para 15, Lord Bingham of Cornhill said that:

This reflects the basic principle that the law does not expect people to arrange their affairs on the basis that others may commit fraud. It also carries echoes of what Lord Wilberforce said in The Ampthill Peerage [1977] AC 547, 569:

This passage from Lord Wilberforce’s speech resonates with earlier authority. In Hip Foong Hong v H Neotia & Co [1918] AC 888, 894, Lord Buckmaster said:

The clear implication from this statement is that in cases of fraud, unlike other instances of claimed miscarriages of justice, it is not necessary to show that the further evidence would have been a determining factor in the result. And, if it was not necessary to show that, it could hardly be said that it would have to be shown that evidence of the fraud could not have been obtained before the first trial by the exercise of reasonable diligence (a more rigorous requirement, by any standard).

A need to show reasonable diligence did not feature in Jonesco v Beard [1930] AC 298, where an application was made to set aside a judgment obtained by fraud. At p 300, Lord Buckmaster said

The special place occupied by fraud in the setting aside of judgments obtained by its use has been recognised in Australia and Canada. In McDonald v McDonald (1965) 113 CLR 529 the High Court of Australia applied Hip Foong Hong and Jonesco v Beard and rejected the notion that, to set aside a judgment obtained by fraud, it had to be shown that evidence of the fraud could not have been obtained by reasonable diligence before the trial which led to the judgment sought to be set aside. Barwick CJ contrasted the position where a verdict was impeached on the basis of fresh evidence with that where it was obtained by fraud. At p 533 he said:

This decision was followed in Toubia v Schwenke [2002] NSWCA 34; (2002) 54 NSWLR 46. As Newey J stated in para 33 of his judgment in this case, in Toubia Handley JA, with whom Heydon JA and Hodgson JA agreed, concluded (in para 41) that “[i]n an action for fraud, a plaintiff must prove that he was deceived but need not prove that he was diligent.” Handley JA continued:

Handley JA, in an earlier passage of his judgment, gave a powerful defence of this principle. Referring to the argument that the dicta in Owens Bank Ltd v Bracco [1992] AC 443 were to the effect that, if fraud was alleged (even for the first time) in an application to set aside a judgment, it had to be shown that it could not have been discovered with reasonable diligence, at paras 37 and 38 he said:

In Canada v Granitile Inc (2008) 302 DLR (4th) 40, the Ontario Superior Court of Justice reached the same conclusion. At para 299, Lederer J said:

Newey J found the reasoning in the Australian and Canadian cases compelling. I also. The idea that a fraudulent individual should profit from passivity or lack of reasonable diligence on the part of his or her opponent seems antithetical to any notion of justice. Quite apart from this, the defrauder, in obtaining a judgment, has perpetrated a deception not only on their opponent and the court but on the rule of law. Newey J put it well when he said, at para 37 of his judgment:

I agree with all of that. It appears to me that the policy arguments for permitting a litigant to apply to have judgment set aside where it can be shown that it has been obtained by fraud are overwhelming.

For the reasons that I have given, I do not consider that the Etoile and Bracco cases are authority for the proposition that, in cases where it is alleged that a judgment was obtained by fraud, it may only be set aside where the party who makes that application can demonstrate that the fraud could not have been uncovered with reasonable diligence in advance of the obtaining of the judgment. If, however, they have that effect, I consider that they should not be followed. In my view, it ought now to be recognised that where it can be shown that a judgment has been obtained by fraud, and where no allegation of fraud had been raised at the trial which led to that judgment, a requirement of reasonable diligence should not be imposed on the party seeking to set aside the judgment.

Two qualifications to that general conclusion should be made. Where fraud has been raised at the original trial and new evidence as to the existence of the fraud is prayed in aid to advance a case for setting aside the judgment, it seems to me that it can be argued that the court having to deal with that application should have a discretion as to whether to entertain the application. Since that question does not arise in the present appeal, I do not express any final view on it. The second relates to the possibility that, in some circumstances, a deliberate decision may have been taken not to investigate the possibility of fraud in advance of the first trial, even if that had been suspected. If that could be established, again, I believe that a discretion whether to allow an application to set aside the judgment would be appropriate but, once more, I express no final view on the question. In Mrs Takhar’s case, she did suspect that there may have been fraud but it is clear that she did not make a conscious decision not to investigate it. To the contrary, she sought permission to engage an expert but, as already explained, this application was refused.

At para 26 of his judgment, Newey J said that the principles which govern applications to set aside judgments for fraud had been summarised by Aikens LJ in Royal Bank of Scotland plc v Highland Financial Partners lp [2013] 1 CLC 596, para 106. There, Aikens LJ said:

I agree that these are the relevant principles to be applied. I also agree with Newey J’s view (expressed at para 47 of his judgment) that Mrs Takhar’s application to set aside the judgment of Judge Purle has the potential to meet the requirements which Aikens LJ outlined. She should not be fixed with a further obligation to show that the fraud which she now alleges could not have been discovered before the original trial by reasonable diligence on her part.

I would therefore allow the appeal and restore the order of Newey J that Mrs Takhar’s case should be allowed to proceed to trial.

Subject to what follows, I agree with the judgment of Lord Kerr. I add some observations of my own only because the disorderly state of the authorities is apt to make the question before us appear more complicated than it really is. In my view, the basic principles on which this case falls to be decided are reasonably straightforward.

An action to set aside an earlier judgment for fraud is not a procedural application but a cause of action. As applied to judgments obtained by fraud, the historical background was explained by Sir George Jessell MR in Flower v Lloyd (1877) 6 Ch D 297, 299-300. Equity has always exercised a special jurisdiction to reverse transactions procured by fraud. A party to earlier litigation was entitled to bring an original bill in equity to set aside the judgment given in that litigation on the ground that it was obtained by fraud. Such a bill could be brought without leave, because it was brought in support of a substantive right. If the fact and materiality of the fraud were established, the party bringing the bill was absolutely entitled to have the earlier judgment set aside. In this respect, an original bill differed from a bill of review on the basis of further evidence, which was essentially procedural and did require leave. After the fusion of law and equity in the 1870, the procedure by way of original bill was superseded by a procedure by action on the same juridical basis.

The cause of action to set aside a judgment in earlier proceedings for fraud is independent of the cause of action asserted in the earlier proceedings. It relates to the conduct of the earlier proceedings, and not to the underlying dispute. There can therefore be no question of cause of action estoppel. Nor can there be any question of issue estoppel, because the basis of the action is that the decision of the issue in the earlier proceedings is vitiated by the fraud and cannot bind the parties: Director of Public Prosecutions v Humphrys [1977] AC 1, 21 (Viscount Dilhorne). If the claimant establishes his right to have the earlier judgment set aside, it will be of no further legal relevance qua judgment. It follows that res judicata cannot therefore arise in either of its classic forms.

The rule, originally stated by Wigram V-C in Henderson v Henderson (1843) 3 Hare 100, 115, that a party is precluded from raising in subsequent proceedings matters which were not, but could and should have been raised in the earlier ones, is commonly treated as a branch of the law of res judicata. It has the same policy objective and the same preclusive effect. But, it is better analysed as part of the juridically distinct but overlapping principle which empowers the court to restrain abuses of its process. The relationship between the two concepts was examined by this court in Virgin Atlantic Airways Ltd v Zodiac Seats UK Ltd [2014] AC 160, paras 22-25. Whereas res judicata is a rule of substantive law, abuse of process is a concept which informs the exercise of the court’s procedural powers. These are part of the wider jurisdiction of the court to protect its process from wasteful and potentially oppressive duplicative litigation even in cases where the relevant question was not raised or decided on the earlier occasion. Since the decisions of the House of Lords in Arnold v National Westminster Bank plc [1991] 2 AC 93 and Johnson v Gore Wood & Co [2002] 2 AC 1 it has been recognised that where a question was not raised or decided in the earlier proceedings but could have been, the jurisdiction to restrain abusive re-litigation is subject to a degree of flexibility which reflects its procedural character. This allows the court to give effect to the wider interests of justice raised by the circumstances of each case.

It is this flexibility which supplies the sole juridical basis on which the respondents can argue that the evidence of fraud must not only be new but such as could not with reasonable diligence have been deployed in the earlier proceedings. It is also the basis on which Lord Briggs, in his judgment on the present appeal, suggests a less absolute rule than that proposed by Lord Kerr. I cannot accept either the respondents’ argument, or Lord Briggs’ more moderate variant of it. The reason is that proceedings of this kind are abusive only where the point at issue and the evidence deployed in support of it not only could have been raised in the earlier proceedings but should have been: see Johnson v Gore-Wood & Co, at p 31 (Lord Bingham of Cornhill) and Virgin Atlantic Airways Ltd v Zodiac Seats UK Ltd, para 22 (Lord Sumption). As Lord Bingham observed in the former case, it is “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.” The “should” in this formulation refers to something which the law would expect a reasonable person to do in his own interest and in that of the efficient conduct of litigation. However, the basis on which the law unmakes transactions, including judgments, which have been procured by fraud is that a reasonable person is entitled to assume honesty in those with whom he deals. He is not expected to conduct himself or his affairs on the footing that other persons are dishonest unless he knows that they are. That is why it is not a defence to an action in deceit to say that the victim of the deceit was foolish or negligent to allow himself to be taken in: Central Railway Company of Venezuela v Kisch (1867) LR 2 HL 99, 120 (Lord Chelmsford); Redgrave v Hurd (1881) 20 Ch D 1, 13-17 (Jessell MR). It follows that unless on the earlier occasion the claimant deliberately decided not to investigate a suspected fraud or rely on a known one, it cannot be said that he “should” have raised it.

Nor do I accept Lord Briggs’ view that a more flexible and fact-sensitive approach may be required in order to distinguish between degrees of dishonesty. I think that this would introduce an unacceptable element of discretion into the enforcement of a substantive right. The standard of proof for fraud is high, and rightly so. But once it is satisfied, there are no degrees of fraud which can affect the right to have the judgment set aside.

Dicta apart, the only direct authority for a requirement that the new evidence should be such as could not with reasonable diligence have been deployed in the earlier proceedings is McIlkenny v Chief Constable of the West Midlands [1980] 1 QB 283, affirmed sub nom. Hunter v Chief Constable of the West Midlands Police [1982] AC 529. This was an action for damages for assault committed in the course of a police investigation into the Birmingham bombing of November 1974, which had ultimately led to the plaintiffs’ prosecution and conviction for murder. The same allegation of assault had been made at the criminal trial as an objection to the admission of a confession. The trial judge had rejected it on a voir dire. The subsequent civil action was struck out as an abuse of process, because it was a collateral attack on the trial judge’s ruling on the admissibility of the confession and thus on the conviction. The actual decision is distinguishable on a number of counts. The earlier proceedings were criminal. Moreover, the result of the civil action, if it had succeeded, would have been to discredit a subsisting conviction without setting it aside. The reasoning, however, is not so readily distinguishable. In the Court of Appeal Goff LJ, at pp 333-335, relied on the authorities on setting aside civil judgments for fraud. These supported the proposition that decisive new evidence must be available in the later proceedings, but in summarising them Goff LJ added a requirement, which they did not support, that the evidence could not with reasonable diligence have been deployed in the earlier proceedings. The sole authority for that refinement was Goff LJ himself. His formulation of the test was endorsed by Lord Diplock, delivering the only reasoned speech in the House of Lords, [1982] AC 529, 545. There are dicta to the same effect in cases arising out of actions to set aside judgments in civil proceedings: see Lord Bridge in Owens Bank Ltd v Bracco [1992] 2 AC 443, 483 and Lord Templeman in Owens Bank Ltd v Etoile Commerciale SA [1995] 1 WLR 44, 48. It may well be that policy considerations justify such a principle where a collateral attack is mounted on a criminal conviction following a trial in which the same issue was decided. There are other procedures for reopening unsafe criminal convictions in such cases. As is well known, these were ultimately invoked and resulted in the convictions of the plaintiffs and McIlkenny and Hunter being quashed. I would respectfully decline to treat the statements in those cases as applying to proceedings to set aside a civil judgment and would hold that the dicta in the two Owens Bank cases were mistaken. None of these judicial statements sufficiently distinguishes between (i) the proposition that an action to set aside a civil judgment must be based on new evidence not before the court in the earlier proceedings, and (ii) the proposition that that evidence must not have been obtainable by reasonable diligence for the earlier proceedings. The first proposition is well established. The second is not supported by any authority earlier than McIlkenny and appears to me to be an insufficient answer to an allegation that a civil judgment has been obtained against the claimant by the deliberate fraud of another party. This is the effect of the decision of the High Court of Australia in McDonald v McDonald (1965) 13 CLR 529, in which the two propositions were carefully and separately considered. It was also the reason why in Toubia v Schwenke (2002) NSWLR 46 the New South Wales Court of Appeal observed that even in the absence of binding Australian authority they would have regarded the dicta in the two Owens Bank cases as being contrary to principle and declined to follow them. In my opinion the Australian cases on this point are correct.

I would leave open the question whether the position as I have summarised it is any different where the fraud was raised in the earlier proceedings but unsuccessfully. My provisional view is that the position is the same, for the same reasons. If decisive new evidence is deployed to establish the fraud, an action to set aside the judgment will lie irrespective of whether it could reasonably have been deployed on the earlier occasion unless a deliberate decision was then taken not to investigate or rely on the material.

I recognise the risk of frivolous or extravagant litigation to set aside judgments on the ground of fraud, but like other members of the court, I think that the stringent conditions set out by Aikens LJ in Royal Bank of Scotland plc v Highland Financial Partners lp [2013] 1 CLC 596, para 106, combined with the professional duties of counsel, are enough keep it within acceptable limits. I do not think that the imposition of further conditions would be consistent with the long-standing policy of equity of reversing transactions procured by fraud.

This appeal turns on the outcome of a bare-knuckle fight between two important and long-established principles of public policy. The first is that fraud unravels all. The second is that there must come an end to litigation. I will call them the fraud principle and the finality principle. On the facts of this case I agree with Lord Kerr that the fraud principle should prevail. As will appear I also agree with most of his reasoning. But I have been unable to follow him all the way down a path which seeks to erect a reliable bright-line boundary between types of case where one principle or the other should clearly prevail. There will be too many cases where that supposed bright line is either invisible, or so technical that it fails to afford a basis for choosing between the two principles which accords with justice, common sense or the duty of the court to retain control over its own process, and thereby protect it from abuse. I would have preferred a more flexible basis upon which, recognising that many cases will straddle any bright line, the court can apply a fact-intensive evaluative approach to the question whether lack of diligence in pursuing a case in fraud during the first proceedings ought to render a particular claim to set aside the judgment in those proceedings for fraud an abuse of process. This approach would in particular seek to weigh the gravity of the alleged fraud against the seriousness of the lack of due diligence, always mindful of the principle that victims of a fraud should not be deprived of a remedy merely because they are careless.

Basing himself on that obvious principle of justice, Lord Kerr proposes a clear rule that a judgment may always be set aside for fraud without regard either to the gravity of the fraud or to any lack of reasonable diligence by the alleged victim, unless either (i) fraud was actually alleged in the earlier proceedings, or (ii) there was a deliberate decision not to investigate a suspected fraud. Only in those cases should the finality principle either prevail, or at least give rise to a judicial discretion to apply it.

As will appear I agree that there should be such a power (although I doubt whether discretion is the right word) in those two types of case, but there will be numerous other types where the absence of any such power will in my view be an unacceptable fetter upon the court’s duty to control its own process, and to protect itself and the parties from abuse. The problems arise mainly from the wide range in the gravity of the alleged frauds, the low threshold of the summary judgment test which the fraud allegation must pass to enable the allegation to be tried, and the almost infinite levels of seriousness of the shortfall in the victim’s application of reasonable diligence. It makes no sense to me either that a serious, pre-meditated, skilfully executed and successful fraud should go without remedy merely because the victim fell short of reasonable diligence by a narrow margin (as the rule propounded by the Court of Appeal would ensure) or that something falling just on the wrong side of honesty should expose the successful litigant to the full rigour of a second trial, where the fraud allegation itself was only just arguable, and the alleged victim was guilty of the most basic failure to test the other party’s case (as I fear that Lord Kerr’s proposed rule would permit).

The allegation of fraud in the present case is a telling example of a grave fraud, at the most serious end of the range. I emphasise that it is at this stage only an untested allegation. It is said that the Krishans planned and implemented a clever forgery of a document vital (if genuine) to their case, and that it was instrumental in their victory. By using a genuine signature of Mrs Takhar, superimposing it on the joint venture agreement, disposing of the original and using a copy to conceal the superimposition, they made it as hard as they possibly could to prevent Mrs Takhar from dealing with it. On looking at the copy document it appeared to her to be, and indeed it was, her signature. Furthermore the alleged forgery was designed from the outset to deceive not only Mrs Takhar but also the court, in litigation which must have been pending when the forgery was planned and committed. By contrast the failure in reasonable diligence may be said to have been at the less serious end of the range. Although the relevant facts have not yet been investigated, it appears likely that Mrs Takhar and her then legal team did take steps to investigate a forgery which they suspected, but they left it too late.

A much more familiar example might go like this. Party A tells a spur of the moment deliberate half-truth (ie a lie) about a fact in issue when cornered during cross-examination at trial, and the advocate for party B fails to put it to A that the statement was a lie (ie dishonest) rather than merely a mistaken recollection, when there were disclosed documents which plainly justified putting it as a lie, absent from the trial bundle due to a serious failure in preparation for trial. Let it be supposed that the deliberate concealment of the whole truth was just on the wrong side of honesty. The trial judge gives credit to A’s evidence, in particular because the offending half-truth was not challenged as a deliberate lie, and this materially contributed to A’s success. A new legal team then does the necessary work on the disclosed documents and B seeks to set aside the judgment for fraud. Neither of Lord Kerr’s exceptions would apply. B would be able to seek a retrial of substantially the same case, if successful in getting the judgment set aside. The court would be powerless to stop the process as an abuse. Strictly B would be entitled to say that the judgment had been obtained by fraud, but a retrial in such circumstances would strike at the heart of the finality principle. On the other hand, if B’s advocate had just said “that’s a lie Mr A”, the court would, on Lord Kerr’s analysis, although not on Lord Sumption’s, have the requisite discretion because fraud would have been put in issue at the trial.

I would suggest that, standing back from the legal technicalities, the real reason why most reasonable observers would say that the application to set aside the judgment in the first example should not be stayed as an abuse, but that in the second example it should, is not because of the brief putting of fraud in issue (if it had been) in the second, but because of the obvious disparity between them, when weighing the gravity of the fraud against the extent of the failure in due diligence. Both examples plainly engage both the fraud and the finality principles. In both of them judgment is alleged to have been obtained by fraud but, equally, in both examples the applicant is seeking, as the overall objective, completely to re-litigate the first case, albeit as the second stage in a two-stage process.

The low level of the summary judgment threshold contributes to the problem in this way. Applications to set aside a judgment for fraud present a potential double whammy to the finality principle. Subject of course to appeals, the first judgment should ordinarily be the end of the matter. But an application to set it aside for fraud will itself involve a trial which, because of the seriousness of the allegation, will be likely to be litigated with bell, book and candle, no stone being left unturned. If the application succeeds, there will then be a third trial, namely the re-trial of the original claim. Of course the third trial will be avoided if the application to set aside fails on the merits, but the only protection against the multiple litigation constituted by the trial of the fraud allegation will be if it fails to raise a triable issue.

Nor therefore is it appropriate to address the extent to which the fraud principle should prevail as against the finality principle on the basis that the fraud has actually occurred. In particular cases the fraud allegation may be a weak one, just passing the summary judgment test, whereas the invasion of the finality principle in such a case will not merely be a risk but an expensive and time-consuming actuality. If these considerations can be weighed in an evaluative balance wherever the two principles are at loggerheads, well and good, but they would only feature in the regime proposed by Lord Kerr if one or the other of his two exceptions applied.

I agree that the dicta of Aikens LJ in in Royal Bank of Scotland plc v Highland Financial Partners lp [2013] 1 CLC 596, para 106, cited by Lord Kerr, provide some protection against the abusive use of fraud allegations as a way of re-opening decided cases. But they would be unlikely to prevent the AB example from withstanding an application for defendant’s summary judgment. A deliberate lie in the witness box is no less fraudulent because it is committed on the spur of the moment. If it contributed materially to the outcome, then the requirement for causation is likely to be satisfied, at least at the level necessary to give rise to a triable issue. But those dicta are concerned with delimiting the cause of action, not the varied circumstances in which its pursuit may amount to an abuse.

I have already described by example the way in which there can be a wide range of seriousness in a failure to conduct litigation with reasonable diligence. An important part of the finality principle is that a party is expected to bring his whole case about the relevant dispute to bear when it is first litigated. That is the foundation of what used to be called the rule in Henderson v Henderson (1843) 3 Hare 100. If a matter relevant to a dispute could have been raised in the earlier case, then it should have been, and to litigate it even for the first time in a second case used automatically to be prohibited. But the rigour and inflexibility of the old rule has been completely transformed by its re-evaluation in Johnson v Gore Wood & Co [2002] 2 AC 1. Now, even if the new matter could have been raised in the earlier proceedings, it no longer follows that it necessarily should have been. Rather, the court conducts an open-ended fact intensive evaluation of the question whether to raise the new matter in a second claim is or is not an abuse. Lord Bingham said, at p 31:

Like Lord Kerr I do not consider that these problems are satisfactorily resolved by the existing authorities. My main reason is that the question whether the making of a particular type of application to a court does or does not involve an abuse of process is not one which is to be regarded as set in stone for all time, once resolved (if it has been) by the highest court. The abuse of process doctrine is one which both needs to be, and has conspicuously been, adaptive to changes in litigation culture over time, during which the pressures upon the courts and the readiness of the courts to conduct evaluative assessments in place of the mechanical application of bright-line rules is constantly changing. Johnson v Gore Wood is a prime but not sole example of the courts’ capacity to adapt to changing circumstances, in an area which is at least in part concerned with procedure.

In any event I agree with Lord Kerr, Lady Arden and Newey J that the existing authorities do not upon analysis provide a reliable guide, even though there are statements in some of them which, taken on their own, do appear to suggest that it is settled law that a lack of reasonable diligence will always be fatal to an application to set aside a judgment for fraud, as the Court of Appeal held. They do not, either because the foundations of earlier authority upon which they are expressed to be based prove upon minute analysis to be much less clear than the later dicta which rely upon them suggest, or because there are factors about the context in which the statements are made which are clearly distinguishable from the present context. Owens Bank v Etoile is an example of the first, while Hunter v Chief Constable is a clear example of the second. To the extent that the English authorities do appear at least superficially to espouse the rule that lack of reasonable diligence will always defeat an application to set aside a judgment for fraud, they simply fail to face up to the invasion thereby caused into the principle that a knave should not escape liability because he is dealing with a fool, as powerfully explained in the Australian and Canadian cases referred to by Lord Kerr.

This is, therefore, an opportunity for this court to put upon a proper modern basis the principles which ought to regulate the court’s power to give full effect to the right of victims of fraud to obtain relief while at the same time exercising an evaluative power to recognise the abuse of that right, and to deal with it where it occurs. This would have enabled the courts to maintain supervision of a type of claim which, if uncontrolled, could turn into a flood of attempts by dissatisfied court users aimed at re-litigation of their disputes, based upon merely arguable assertions that they lost because an opposing party lied in the witness box, or even encouraged a witness to do so.

The starting point is clearly to recognise that the right to have a judgment set aside for fraud is a distinct cause of action recognised by the common law, like a right to set aside a contract obtained by fraud, which is not inherently conditional upon any requirement for the exercise of reasonable diligence in the proceedings leading to the impugned judgment or, for that matter, the making of the impugned contract. In short, the exercise of reasonable diligence is not in any way part of the cause of action. It is for example fully applicable to a case in which judgment followed upon a defendant simply letting the case go by default, if fraud was involved in the obtaining of the judgment. It is a claim for relief obtainable as of right (ex debito justitiae), rather than only by the invocation of a judicial discretion. Nor does the right depend upon the seriousness of the fraud.

Two consequences flow from that starting point. The first is that, if no allegation of fraud was made in the proceedings leading to the impugned judgment, there is no question of cause of action or issue estoppel, for the reasons given by Lord Kerr and Lord Sumption. The second is that the absence of reasonable diligence is not of itself a reason for staying the claim to set aside the judgment. This would be to deny relief to foolish victims of a knave, merely because of their foolishness. The only reason to stay the application to set aside is if the lack of reasonable diligence is so serious, in the context of all other relevant factors, that the application can really be categorised as abusive.

But by contrast with claims to set aside agreements for fraud, applications to set aside a judgment for fraud will usually engage the finality principle, because of the re-litigation objective which normally lies at the heart of them. Leaving aside default judgments (where there has as yet been little real litigation), the objective of the applicant is not merely to have the impugned judgment set aside, but also to clear the way to have the original dispute relitigated. This may be thought to be obvious where the applicant is (like Mrs Takhar) the unsuccessful claimant in the earlier case, but it is in substance also true of a defendant’s application. Defendants will not of course wish thereby to bring the same claim again, but their objective is to keep the fraudulent claimant away from their door unless the claimant undertakes the burden of a completely new case, all the way to trial.

It is in my view no answer to this analysis to say that the application to set aside a judgment for a fraud not previously alleged will not of itself re-litigate anything, so that the finality principle is not thereby really engaged at all. The overall objective is re-litigation. The forensic pursuit and defence of the fraud allegation may well travel over, or at least overlap with, ground trampled on in the original proceedings. The present case is an obvious example, since the question whether Mrs Takhar made the agreement in issue was central to the original trial, and would plainly be relevant to the issue whether the document apparently recording the agreement was a forgery. Even if it does not, the application will inevitably involve at least the risk of further expensive and time-consuming proceedings about the entitlement of the opposing party to relief already obtained by a judgment which, subject to appeal, ought (in the public interest) to have put an end to the underlying dispute. To that extent I respectfully disagree with the thrust of this part of Lord Kerr’s analysis about the separateness of the original proceedings and the application to set aside. In my view the contest between the fraud and finality principles is inherent in applications of this kind, rather than only in Lord Kerr’s two exceptions, where fraud was either alleged, or suspected but deliberately not investigated, although it is of course present a fortiori in such cases.

That being so, it is in my view wrong in principle to say that the generality of applications to set aside judgments for fraud are entirely unaffected by questions about lack of reasonable diligence (or other factors pointing towards abuse) subject only to the identification of specific exceptional types where a judicial discretion may be engaged. The true principle should be to recognise that such applications constitute the assertion of a legal right with which the court will only interfere if satisfied that the exercise of the right is abusive, but that all such applications by their nature give rise to a risk of abuse with which the court is duty bound to engage, because they undermine finality by their mere pursuit, regardless of outcome. Re-litigation is always unfortunate, but it by no means follows that the reason for it is an abuse of process by the applicant. It may well be the respondent who is the real culprit.

I would not describe the court’s exercise of this duty to guard against abuse as a discretion. There should be no judicial interference with the exercise of the right to set aside a judgment for fraud unless the court is satisfied that it involves an abuse of process. If it does, then the proceedings should be stayed. That will not be discretionary, but it will involve the evaluation of a potentially wide range of factors. I have already mentioned the gravity of the fraud and the extent of the shortfall from the exercise of reasonable diligence. Those would almost always be relevant. But the categories of potentially relevant factors are in principle unlimited. They might include, in particular cases, the centrality (or otherwise) of the fraudulent conduct to the outcome of the case, the extent to which (as here) the alleged fraud was specifically aimed at taking advantage of a lack of care in the preparation of the case by the alleged victim, the resources of the alleged victim during the first proceedings, the amount of toil, treasure and court time which would be thrown away by the setting aside of the judgment, the amount of the same which would be likely to be consumed by the trial of the fraud allegation and, if successful, the re-trial of the original claim, and even the apparent strength (or otherwise) of the allegation of fraud. But from start to finish, the question is whether the application really is an abuse of process.

Nor would I expect this evaluative approach frequently to come down in favour of a stay. The principle that fraud unravels all is deeply rooted in the common law, and its continued application is an important contributor to honesty within society, to the rule of law and to the ability of the courts to adjudicate disputes justly. Fraud of this kind is all the more serious because it is aimed at deceiving the court itself. But the court must arm itself against always having to allow re-litigation, and potentially two further trials between the same parties, wherever the unsuccessful party wants to allege, for the first time, that the case was lost because an opposing party was lying about, rather than just mis-recalling, the facts in issue, and can demonstrate an arguable case that this is what happened at trial.

Lord Sumption equates the setting aside a judgment for fraud with the setting aside of any transaction for fraud on the basis that a reasonable person is entitled to assume honesty in those with whom he deals, and is not expected to conduct himself or his affairs on the footing that other persons are dishonest unless he knows that they are. I fully agree with that approach to ordinary transactions out of court such as the making of a contract. But I cannot agree that it is reasonable for a litigant always to assume that an opponent’s case (and evidence in support of it) is honest, with a concomitant right to conduct litigation on that basis, and to re-litigate the same dispute whenever he can show, after judgment, an arguable case that his assumption was wrong.

It will be apparent from the foregoing why I regard the present case as lying very clearly on the non-abusive side of the line, without forming any view at all about whether the allegation of this fraud will in fact be proved. If proven it was a very serious, pre-meditated, carefully planned and executed fraud which was instrumental in the defeat of Mrs Takhar’s claim, and plainly aimed from start to finish at deceiving the court about the central issue in the case. For the reasons given Mrs Takhar’s failure to use reasonable diligence was by no means at the more serious end of the scale. The expert evidence which she has now obtained, although thus far untested or opposed, plainly gives her a real (rather than merely arguable) prospect of success. Her application comes nowhere near being categorised as an abuse. I would therefore allow the appeal.

This appeal concerns Mrs Takhar, who has brought an action (“a rescission action”) to rescind a judgment against her and made available to the court fresh evidence which, if proved at the trial of the rescission action, will demonstrate that the winning party obtained judgment against her by fraud. Preventing a person from prosecuting a rescission action in these circumstances amounts to restricting her right to pursue her cause of action in fraud, and to have access to justice for that cause of action. Therefore, the law should only impose a restriction on such a claimant in a rescission action where there is justification for doing so.

I agree with Lord Kerr that there is no authority which binds this court to hold that failure to act diligently in searching for this evidence before the original trial is, of itself, a bar. It is easy to see how it came to be thought that reasonable diligence in this regard had to be shown as in Boswell v Coaks (No 2) (1894) 86 LT 356n (using the fuller report in the footnotes to Birch v Birch (1902) 86 LT 364, which is also the report used by Lord Templeman in the passage cited by Lord Kerr at para 40 above), the Earl of Selborne, giving the judgment of the House of Lords, held obiter, having emphasised the importance of the finality of judgments, that the rules which applied to a bill of review before 1875 should continue to apply “in their full force, and even with greater freedom than before”. Those rules included a threshold condition in a rescission action that the fresh evidence “could not possibly have been used when the decree was made” (see Thomas v Rawlings (1865) 11 LT NS 721, 722, a decision of the Court of Appeal in Chancery). Moreover, the Court of Appeal in Birch v Birch appears obiter to have concluded that this earlier rule continued to apply (see (1902) 86 LT 367). This may have been the reason why Goff LJ held in McIlkenny v Chief Constable of the West Midlands [1980] QB 283, 335 that there was such a rule. In the House of Lords, in the same case but under the name of Hunter v Chief Constable of the West Midlands Police [1982] AC 529, Lord Diplock approved Goff LJ’s holding, but, as Lord Kerr has explained, in a different context. Furthermore, it may be, as the Federal Court of Australia suggested in Monroe Schneider Associates Inc v No 1 Raberem Pty Ltd (No 2) (1992) 37 FCR 234, 239, that prior to 1875 the Chancery Court had to be somewhat circumspect when asked to rescind a judgment given in a common law court: this might explain such threshold conditions.

It is only right that in the generality of cases a judgment obtained by the fraud of the winning party should be rescinded because it is wrong in principle that a person who is proved to be a fraudster should obtain and retain the fruits of his fraud. Clearly, however, a restriction can be justified in some special situations. In other words, there are occasionally exceptions to the principle that “fraud unravels all”. There are cases where both parties have colluded to deceive the court - in those cases it would be an abuse of process to ask the court later to rescind the judgment. As Lord Mansfield held in Montefiori v Montefiori (1762) 1 Black W 363, 96 ER 203 “no man shall set up his own iniquity as a defence, any more than as a cause of action”.

There are other situations: for instance, where the fraud is not material to the outcome: see, for example, Boswell v Coaks. Likewise, a claim to set aside a decree absolute made on divorce, which is equivalent to an order in rem, was not set aside in Callaghan v Hanson-Fox [1992] Fam 1 because the parties knew about the evidence and its significance. There must be actual fraud: constructive fraud is not enough. The fraud must be one for which the defendant is responsible. There may well be other situations which I have not mentioned. There may for instance (I express no view) be exceptions where the judgment in the original action was obtained by perjury during the trial or where the fraud was in fact pleaded in the first action.

Greater difficulty lies in situations where at the time of the original action a party suspects a fraud but does not investigate it or decides not to investigate it. The justice in this situation may not be so easily answered by allowing an unfettered right to bring a rescission action. I would treat this as a case where the Ashingdane principles found in the jurisprudence of the European Court of Human Rights apply (see Ashingdane v United Kingdom (1985) 7 EHRR 528). Any restriction would have to be derived from a rule which serves the legitimate aim of proving a just solution, thus striking a fair balance between the relevant considerations and going no further than necessary, and which does not defeat the core right of access to court.

There are factors which favour some restriction on the victim’s right in this situation. The judgment in the original action will be final and conclusive (subject to any appeal, and it is to be noted that on any appeal lack of reasonable diligence in obtaining the new evidence for the trial would be relevant). Finality in judgments leads to certainty, and hopefully to the social benefits of dispute resolution. Where property is in issue (eg the ownership of a business), the owner following a final judgment can develop it, invest in it and use it as security to raise money to develop other businesses free from the risk that it might be claimed by someone else. That also is for the economic and social benefit of the community, and there is a social and economic cost if that process is delayed.

Furthermore, a litigant has plenty of opportunities to challenge the other party’s case under the Civil Procedure Rules (“CPR”). Mrs Takhar could have served notice on the respondents to prove the disputed agreement, sought further information about it from the respondents and appealed from the denial of permission to adduce expert handwriting evidence. A party is expected to co-operate with the court and the other parties in ensuring that so far as possible all issues are dealt with efficiently at a single trial. Under the overriding objective in CPR rule 1.1(1), the court must deal with cases just and proportionately and this includes allotting to each case “an appropriate share of the court’s resources, while taking into account the need to allot resources to other cases”. Litigation resulting in rescission actions may involve not just one but three actions in all. There is also always the risk that the defendant if successful will incur costs which he cannot recover. In assessing reasonable diligence, however, it must be recalled that some litigants do not have legal representation and also that a party is entitled to conduct herself on the basis that the other side is not fraudulent, and that no investigations are needed, until she has real grounds for suspecting fraud.

It is of course important for the efficient despatch of litigation that a court insists on compliance with its procedural rules: failure to do so in appropriate cases would provide an incentive to non-compliance.

The rule in contention on this appeal, however, held by the Court of Appeal to be derived from case law, selects just one consideration - and one based on the victim’s conduct. It is illogical to the extent that it automatically imposes a sanction (a ban on bringing the second action) that may be wholly disproportionate to the lack of diligence. Moreover, it would leave all other factors out of account, including the defendant’s allegedly fraudulent conduct. In any event, a restriction on a person pursuing a cause of action should in principle only be imposed where it is necessary to do so to protect the rights and freedoms of others.

In addition, such a rule would take no account of the protections that can be provided to the defendant in appropriate cases by the exercise of other procedural powers, such as the power to strike out actions which had no real prospect of success. Where the defendant to the rescission action considers that it is clearly not well-founded, he can apply to strike out the action on the grounds of abuse of process or obtain summary judgment in his favour. He can be protected in costs and also by the strict rule of procedure that a fraud must be particularised with exactitude: see Jonesco v Beard [1930] AC 298.

Where the defendant is prejudiced, and his position cannot for some reason be safeguarded, or the rights of innocent third parties have intervened, it may be relevant to take into account any remedies that the victim may have against her professional advisers in the original action. Where innocent third parties have obtained rights, there may indeed be no point in rescission of the original judgment and the victim may have to resort to other remedies.

Too robust a rule favouring finality might encourage litigants to attempt to deceive the other parties or the court. Where deceit is practised on the court, the integrity of the legal system is put at risk and that is an important consideration against the rule contended for.

But the short point is that there is as of now at least no procedural rule about any restriction on a person bringing a rescission action in the CPR. The drafters of those rules may wish to consider whether the position should be changed following these judgments. The jurisprudence of the courts of Australia and Canada cited by Lord Kerr is instructive. It is worth noting that in Clone Pty Ltd v Players Pty Ltd (in liquidation) (Receivers & Managers Appointed) [2018] HCA 12 the High Court of Australia was only dealing with lack of reasonable diligence before the fraud was discovered, but the position reached in these cases is not universal in common law jurisdictions. For instance, a rule requiring the claimant in the rescission action to show that he made a “reasonable effort in the original trial to ascertain the truth of the matter” can be found in the Restatement of Judgments (Second) (1982), para 70 (American Law Institute). There is also legislative precedent for the loss of a right of action as a result of a failure to exercise reasonable diligence in discovering fraud in Limitation Act 1980, section 32.

It might be salutary if the CPR were to require a party bringing a rescission action to provide an explanation as to his state of knowledge at the time of the first action in his pleading in the second action, or if they enabled the court to award some security for costs or impose other conditions if reasonable diligence was not taken in the first action. I agree with much that Lord Sumption has said in his judgment, but the question whether conditions should be imposed in this context may involve considerations apart from the imperative of reopening judgments procured by fraud and fall well short of preventing actions for that purpose being brought. The Civil Procedure Rule Committee is empowered to introduce changes to the CPR, and importantly the process involves consultation.

The statement of principles set out by Aikens LJ in Royal Bank of Scotland plc v Highland Financial Partners lp [2013] 1 CLC 596 at para 106 approved by Lord Kerr deals with the position at the trial of a rescission action, not with threshold conditions on bringing such an action.

In the absence of any special provision in the CPR, and assuming the claim is properly pleaded, the only remedy available to the defendant in a rescission action is to apply to the court for an order barring the claim as an abuse of process. It must be a matter for the respondents in this case whether to pursue that course. They may wish to bring an application or even to contend that some exception to the general rule which this court has found applies. There are matters which cause me some concern on the limited evidence before this court: the fact that Mrs Takhar had concerns about the authenticity of her signature on the agreement from 2008 (two years before the trial), the fact that she had no other explanation in the original trial for the fact that the agreement appeared to bear her signature, her failure to challenge the authenticity of the agreement or to appeal the judge’s order denying permission to adduce handwriting evidence, the fact that she has received a settlement already from her solicitors suggesting that there was a lack of reasonable diligence (and clearly it cannot matter that it was the solicitors and not the client who showed lack of diligence), and the fact that two years passed between obtaining the report of the handwriting expert and bringing these proceedings.

I do not consider that it is for this court to determine the precise state of knowledge at any time of Mrs Takhar or those representing her. It is enough to say that in my judgment, there is not, and should not be, a rule that want of reasonable diligence in the first action of itself leads to a blanket ban on bringing an action to rescind a judgment which the claimant can properly allege the respondents obtained by fraud. This appeal should be allowed and the order of Newey J restored.