JUDGMENT
Mueen-Uddin (Appellant) vSecretary of State for the Home Department (Respondent)
before
Lord Reed, President
Lord Sales
Lord Hamblen
Lord Burrows
Lord Richards
20 June 2024
Heard on 1 and 2 November 2023
Appellant
Jacob Dean
Lily Walker-Parr
(Instructed by Carter-Ruck)
Respondent
Anthony Hudson KC
Ben Silverstone
Robbie Stern
(Instructed by Government Legal Department)
LORD REED (with whom Lord Sales, Lord Hamblen, Lord Burrows and Lord Richards agree):
Introduction
The claimant in these proceedings, Chowdhury Mueen-Uddin, was born in East Bengal, which then formed part of the state of Pakistan. Since the conclusion of the war of independence in December 1971, it has been the state of Bangladesh. He has lived in the United Kingdom since 1973, and has been a British citizen since 1984. He has held a number of prominent public and charitable positions in British society. He served as Secretary General of the Council of Mosques in the United Kingdom and Eire from 1984 to 1988 and helped to set up the Muslim Council of Great Britain. He was Deputy Director of the Islamic Foundation from 1995 to 2005. He was Director of Muslim Spiritual Care Provision in the National Health Service from 2005 to 2012. He has served as vice-chairman of the East London Mosque and the London Muslim Centre, as chairman of Tottenham Mosque and Islamic Community Centre, and as chairman of Muslim Aid.
In October 2019 the Home Office published a report prepared by the Commission for Countering Extremism, a non-statutory committee of the Home Office, entitled “Challenging Hateful Extremism” (“the Report”). It was published in hard copy and online on the government’s website. Part One of the Report was entitled “What Extremism Looks Like in England and Wales”. Under the sub-heading “Ideological and Sectarian Violence”, it stated (p 54):
“We also heard about violence towards secular people from those of a similar faith background. Muslim bloggers described being physically attacked during a protest in East London [footnote 157]. The protest was to show support for the conviction of a senior Jamaat-e-Islami leader for war crimes committed during the 1971 War of Independence [footnote 158]. Some of those we spoke to are in hiding.”
Footnote 158 stated:
“Links between those responsible for the violence in 1971 and JI in the UK including community leadership in East London are well established. Chowdhury Mueen Uddin, former vice chair of the East London Mosque and who helped found the Muslim Council of Britain was found guilty of crimes against humanity following a trial in absentia. See: Channel 4. 2013. ‘British Muslim leader sentenced to death for war crimes’ 3 November 2013, (accessed: 4 September 2019) <https://www.channel4.com/news/chowdhury-mueen-uddin-war-crimes-london-muslim>’”.
It is agreed that around 80 hard copies of the Report were distributed and that the online version was downloaded about 5,000 times. It is not known how often it was read without being downloaded. It is agreed that it may have reached over 900,000 followers of the Home Office’s social media accounts. The claimant avers that he was contacted by numerous colleagues and associates who had read the Report and the allegations about him which it contained. He states that he was devastated that the government of the country of which he was a longstanding citizen and to which he had devoted the last 50 or so years of his life, through community work and public service, would treat him with such disregard, and that he was deeply concerned that a great many people would believe that the allegations made against him were true, because they were made by the Home Office.
A letter of claim against the Secretary of State was sent on behalf of the claimant in December 2019. In March 2020, five months after the Report had first been published, the online version was edited to remove the footnote which referred to the claimant. However, the allegations which had been made against the claimant were not (and have never been) retracted.
On 19 June 2020 the claimant issued proceedings against the Secretary of State for damages for libel, and for compensation for breach of statutory duty pursuant to Parliament and Council Regulation (EU) 2016/679 (the General Data Protection Regulation (“the GDPR”)), arising from the allegations made in the Report.
On 16 February 2021 Tipples J determined, as a preliminary issue in the proceedings, that the natural and ordinary meaning of the words used in the material part of the Report was that the claimant “(i) was one of those responsible for war crimes committed during a 1971 War of Independence in South Asia; and (ii) has committed crimes against humanity during a 1971 War of Independence in South Asia”, and that those meanings were allegations of fact which were defamatory at common law: [2021] EWHC 269 (QB), para 71. Tipples J accepted (para 70) that an accusation of crimes against humanity is plainly very grave. That is evidently correct: it is difficult to imagine a graver allegation than of guilt of war crimes and crimes against humanity. The allegation is especially grave when it is made by the government of this country against one of its own citizens.
On 15 November 2021 Sir Andrew Nicol, sitting as a judge of the High Court, struck out Mr Mueen-Uddin’s claim in limine as an abuse of process: [2021] EWHC 3026 (QB). On 28 July 2022 the Court of Appeal (Dame Victoria Sharp PQBD and Dingemans LJ, Phillips LJ dissenting) dismissed his appeal against that decision: [2022] EWCA Civ 1073; [2022] EMLR 23. The claimant now appeals against the decision of the Court of Appeal.
The factual background
In order to understand the reasoning of the courts below and the issues in the appeal, it is necessary to explain in outline the relevant factual background. Since the appeal concerns a strike out application, and the claim was struck out before a defence had been pleaded or the claimant had served a reply to the defence, this summary is based principally on the particulars of claim and the witness statements filed in support of it. No facts have yet been found.
In March 1971 a war of independence broke out in East Pakistan, during which large numbers of people died. Numerous atrocities were committed, including the abduction between 10 and 15 December 1971 of a number of prominent intellectuals, 18 of whom were murdered. On 16 December the Pakistan Army surrendered to India, which had invaded earlier that month, ensuring Bangladesh’s independence.
Until May 1971 the claimant belonged to Islami Chatra Sangha, an Islamic student organisation which was widely—but, according to the claimant, mistakenly—regarded as the student wing of the political party Jamaat-e-Islami, which was opposed to the creation of an independent Bangladesh. He denies having been involved in any violence.
Shortly after the killing of the intellectuals, the claimant became aware of allegations that he was a member of the militia which was said to be responsible for their deaths. A newspaper article was published in Bangladesh on 29 December 1971 which accused him of masterminding their murder. On 3 January 1972 the New York Times reported that he had been identified as the head of an organisation known as Al-Badar, which was said to have murdered several hundred prominent intellectuals.
The claimant left Bangladesh in late December 1971. He arrived in the United Kingdom in June 1973 and has lived here since then. He visited Bangladesh several times between 1982 and 2004.
In 1973 Bangladesh enacted the International Crimes (Tribunals) Act 1973 (“the 1973 Act”), which established the International Crimes Tribunal (“ICT”). It is not an international body but an institution of Bangladesh’s domestic legal system, with the power to try and punish individuals for a variety of crimes committed in Bangladesh. The 1973 Act (as amended) contains provisions which necessitated the amendment of the Constitution of Bangladesh, by the insertion of article 47A, so as to remove the rights of those accused under the 1973 Act from seeking constitutional remedies. It bars any challenge to the composition of a tribunal or the appointment of its members, disapplies the ordinary rules of evidence and procedure set out in the Criminal Procedure Code and the Evidence Act, provides that the tribunal is not bound by rules of evidence and can, in particular, rely on newspaper reports, and empowers the tribunal to impose the death penalty. Under its rules of procedure, the ICT’s decisions as to the admission of evidence are final and cannot be challenged. The ICT appears to have been dormant until about 2009, after which a new government was elected on a platform that the ICT would start to take cases.
In 1995 Channel 4 broadcast a programme which alleged that the claimant had been involved in the murder of the 18 intellectuals. He began libel proceedings against Channel 4, but did not have the financial resources to pursue the case to trial. The proceedings ended without either party paying the other’s costs.
In 2012 the claimant became aware from media reports that he was being investigated in Bangladesh. He was not contacted by the Bangladesh authorities during the investigation. He instructed his lawyer to make a public statement stressing his innocence and his concerns about how the investigation was being conducted.
On 28 April 2013 the chief prosecutor of Bangladesh submitted formal charges against the claimant. On 2 May 2013 the ICT issued an arrest warrant. Although the claimant’s location in the United Kingdom and the contact details of his lawyer were known to the Bangladesh authorities, he was not served with the arrest warrant or notified of the indictment. Nor was his extradition requested. On learning of the arrest warrant he issued a public statement on his website that he rejected the charges levelled against him and expressed his concerns that the ICT process was neither open nor fair.
On 14 and 15 May 2013 the ICT published a notice in two Bangladesh newspapers requiring the claimant to appear before it within ten days. He decided not to appear because of his inability to contribute to his defence without appearing in person, public criticisms of the fairness of the ICT and its procedures, and his fear of being executed if he was convicted, as another person had been earlier that year. There had been widespread international criticism of the ICT for failing to respect minimum fair trial guarantees and for lacking judicial independence. There had also been criticism of misconduct by prosecutors and judges of the ICT. By way of example:
On 11 July 2011 Human Rights Watch recorded that amendments to the ICT rules of procedure failed to bring the law and rules into compliance with international standards. Further amendments were said to be needed, including allowing the accused to question the impartiality of the tribunal, defining war crimes in accordance with international standards, ensuring that the defence had more than the current three weeks to prepare, allowing for interlocutory appeals and appointing an independent panel for appellate review.
On 28 November 2011 Stephen J Rapp, US Department of State Ambassador at Large for War Crimes Issues, identified four issues with the ICT: (i) the lack of definition of crimes against humanity; (ii) the need for restoration of constitutional rights, including rights to consult with counsel, to prepare and to challenge the process; (iii) the need for protection of witnesses from threats and intimidation; and (iv) the need for transparency of the proceedings.
On 16 November 2012 the Bar Human Rights Committee of England and Wales expressed its concern that the ICT was failing to meet international fair trial standards. It expressed particular concern about interference in the defence; harassment, intimidation and surveillance of the defence team; denial of constitutional rights and perceived bias; and the lack of independence of tribunal members and of the chief prosecutor from both tribunal members and the government.
On 7 February 2013 two UN Special Rapporteurs appointed by the UN Human Rights Council referred to concerns as to the impartiality of the ICT judges and prosecution services, as well as their independence from the executive. They also referred to complaints by witnesses and lawyers for the defence of hostility, intimidation and harassment.
The ICT determined on 27 May 2013 that the claimant had “absconded or […] concealed” himself (para 22 of its judgment), although he had emigrated from Bangladesh 41 years earlier and had been living openly in this country during the intervening period at an address which was known to the Bangladesh authorities, with a high profile in British society. It proceeded to try the claimant in his absence. Court-appointed defence counsel was appointed on 4 June 2013 to represent him, but the claimant says that counsel never contacted him. The claimant understands that counsel were not permitted to take instructions from defendants who were being tried in absentia. The trial began on 15 July 2013. Judgment was given on 3 November 2013. The claimant was convicted of 11 charges of abetting, and complicity in, the commission of the offence of “extermination” as a crime against humanity, and sentenced to death by hanging.
It appears from its judgment that the tribunal heard oral evidence from around 20 witnesses, most of whom were young children at the time of the events in question. Most of the witnesses gave hearsay evidence, but some of them identified the claimant as being involved in the abduction of university professors who were subsequently murdered. The tribunal also received evidence of statements made by witnesses who had died before the trial. These included a filmed interview which formed part of the Channel 4 programme, in which a witness identified the claimant as being involved in the abduction of a professor. However, the tribunal relied primarily on newspaper reports. It stated that “the role and position of [the claimant] in 1971 needs to be categorized, chiefly on the basis of investigative reports mostly published immediately after the incidents in the news media” (judgment, para 108). It found that the Bangladesh newspaper report mentioned in para 11 above “proved that the ‘Al-Badar’ was a fascist organisation of JEI [Jamaat-e-Islami] and [the claimant] had acted as ‘operation-in-charge’ of Al-Badar in accomplishing the designed and calculated killing of intellectuals” (judgment, para 110). Further support was found in the New York Times report mentioned in para 11 above, and in a report published by the Observer. The claimant’s having left Bangladesh in 1971 was also treated as “a fair indicative of [his] guilty mind” (para 187).
United Kingdom media reported the conviction, and also reported that the claimant denied the charges and maintained that the trial was unfair and politically motivated. They also reported that the fairness of the ICT proceedings had been criticised by human rights organisations and other commentators. The claimant issued a statement on his website denouncing the conviction. He did not appeal against it within the 30-day time limit for doing so.
In December 2017 Interpol issued a “red notice” in respect of the claimant at the request of Bangladesh. The purpose of such a notice is to locate the person in question and to request their provisional arrest with a view to extradition. The red notice was withdrawn in 2018 after representations were made on behalf of the claimant. In its decision to withdraw the red notice, the Commission for the Control of Interpol’s Files stated (para 39):
“… the Commission paid particular attention to the numerous reports issued by reliable sources such as the United Nations (the High Commissioner for Human Rights, the Special Rapporteur for the Independence of Judges and Lawyers, the Special Rapporteur on extrajudicial, summary and arbitrary executions, the Special Rapporteur on Torture, the Working Group on Arbitrary Detention), diverse foreign governments and national entities (United States Special Ambassador for Global Criminal Justice, United States Congressional Tom Lantos Human Rights Commission, European Union Parliament, United Kingdom as well as various human rights organizations (Human Rights Watch, Amnesty International, International Commission of Jurists, International Centre for Transitional Justice), which all express serious concerns over the procedural safeguards before the ICT and document instances of witness abduction, intimidation of defence counsel, media censorship, pressure to convict from the government, lack of independence of judicial officers, amounting to gross violations of international fair trial standards.”
The Commission said (para 41) that it appeared from the judgment of the ICT that the counsel appointed to represent the claimant was “either unwilling or unable to contact [him] and prepare a robust defence”, resulting in the absence of any defence evidence at the trial. The Commission also said (para 42), in relation to the impartiality of the ICT judges, that the judgment contained “emotionally-charged wording, historical considerations or political innuendos unrelated to the legal reasoning”. The Commission concluded (para 49), on the basis of the information provided by the Bangladesh authorities themselves, that the ICT proceedings against the claimant were not compliant with Interpol’s constitution or with the Universal Declaration of Human Rights.
The proceedings below
Before Sir Andrew Nicol ([2021] EWHC 3026 (QB)), counsel for the Secretary of State submitted that both the claim in libel and the GDPR claim were an abuse of the process of the court. It was an abuse to seek to relitigate in subsequent civil proceedings whether a person had been rightly convicted by a criminal court. It made no difference that the criminal court in question was outside the jurisdiction. Such a claim brought the administration of justice into disrepute and was unfair to the defendant compelled to meet such a claim. The Secretary of State’s position was that that was so even if all the claimant’s complaints about the fairness of the ICT proceedings were justified and he was convicted after a gross miscarriage of justice. The claimant’s criticisms of the fairness of the trial were irrelevant. An alternative way of putting the Secretary of State’s case on abuse was that “the game was not worth the candle” and should be struck out on the ground recognised by the Court of Appeal in Jameel (Yousef) v Dow Jones & Co Inc [2005] EWCA Civ 75; [2005] QB 946 (“Jameel”).
The judge accepted those submissions (para 83). The proceedings were in his view an abuse of process, applying the approach laid down by Lord Diplock in Hunter v Chief Constable of the West Midlands Police [1982] AC 529 (“Hunter”). The courts were vigilant to guard against a collateral attack on a criminal conviction. The same considerations applied to a conviction by an overseas court. That was so even if the proceedings in Bangladesh were grossly unfair. The proper course was for the claimant to have appealed within the Bangladesh criminal process. It made no difference that that opportunity had passed. The claimant was trying to have the best of both worlds by remaining outside Bangladesh but challenging his conviction. Furthermore, it would be unfair to the Secretary of State for the claim to continue. In order to establish a defence of truth under section 2 of the Defamation Act 2013 (“the 2013 Act”), the Secretary of State would have the burden of proving the claimant’s guilt over 50 years after the events in question. Inevitably, witnesses must have died and documents been lost. The claim should therefore be struck out.
This conclusion rendered the Jameel point moot, but the judge considered it briefly. The proceedings would be lengthy and costly. While it was a serious matter to deprive a litigant of the opportunity to vindicate his reputation, the Jameel jurisdiction existed, and this would, if necessary, have been an appropriate case in which to exercise it.
The principal judgment in the Court of Appeal ([2022] EWCA Civ 1073; [2022] EMLR 23) was given by Dingemans LJ, with whose reasoning Dame Victoria Sharp PQBD agreed in a concurring judgment. He considered that the case was one “where Hunter abuse of process overlaps with Jameel abuse of process” (para 54). There were several factors which made the proceedings abusive.
The first was the fact that the claimant had been convicted before the ICT. There was no evidence that he had attempted to contact counsel appointed to represent him so that he might provide a written statement or give evidence from overseas. Although the fact that the claimant was in practice unable to appear before the ICT impacted on his opportunity to defend himself, that did not mean that the present proceedings could not be an abuse of process. If the claimant’s conviction had been the only basis for alleging an abuse of process, that might not have been sufficient, but it did not stand alone.
The second factor was that it was apparent that the claimant’s reputation since he left Bangladesh was that he was involved in the murder of the intellectuals. This was reported in Bangladesh and internationally. That, of itself, might not have been enough to affect his reputation. However, in 1995 Channel 4 had broadcast its programme in which the claimant was accused of being involved in the murder of the intellectuals. It was apparent that thereafter in England and Wales the claimant’s reputation was that he had been involved in those murders. That evidence was properly taken into account when considering whether the proceedings would serve any purpose in vindicating his reputation. This was relevant to a consideration of Jameel abuse, and whether the proceedings would be worth the candle of pursuing them.
The third factor was that the colleagues and associates of the claimant who read the allegations about him in the Report were likely to have an unusually developed interest in its subject, since there were not many people who would read footnotes in a government report without such an interest. They were therefore overwhelmingly likely to have been aware of the claimant’s conviction by the ICT and the contents of the Channel 4 programme. It was relevant that the footnote which referred to the claimant had subsequently been deleted from the online version of the Report, because this meant that there was no continuing publication of the allegations.
The fourth factor was that it would be impossible to obtain a fair hearing of the Secretary of State’s defence of truth, because of the length of time since the murder of the intellectuals in December 1971. If the proceedings continued, a judge would have to do his or her best to determine whether the allegations against the claimant were true without much live evidence. Whatever the judge concluded would not lead to a vindication of the claimant’s reputation, since those who did not accept the result would point to the absence of relevant witnesses and evidence. The costs of investigating such matters were likely to be very substantial.
The fifth factor was that in those circumstances the continuation of the proceedings would be manifestly unfair to the Secretary of State, because it would not be possible to have a fair hearing of the defence of truth. That also meant that the continuation of the proceedings would bring the administration of justice into disrepute. Further, given the evidence about the claimant’s reputation, the proceedings were not worth the candle of pursuing them.
In relation to the data protection claims, the claims for accuracy raised substantially the same matters as the claim for defamation and were an abuse for the same reasons. In considering whether the data had been processed for a lawful purpose, it would be relevant to determine whether the statements were accurate, since that would be relevant to the public interest in publishing.
In response to the dissenting judgment of Phillips LJ, Dingemans LJ observed (para 63) that the law of libel was critically concerned with damage to reputation. The evidence of the claimant’s general reputation before the publication of the Report was that he was a war criminal. He had left the Channel 4 programme uncontradicted. Further, his conviction by the ICT was admissible evidence of his general reputation. If the claimant wanted a fair trial of the issue of whether he was wrongly accused of being a war criminal, he had that opportunity in 1995.
Phillips LJ delivered a dissenting judgment. In his view, the proceedings did not constitute Hunter abuse of process or come anywhere close to doing so (para 79). The claimant had not had a fair opportunity, or indeed any opportunity, to defend himself before the ICT. In those circumstances the ICT conviction could not be a relevant factor in deciding whether the proceedings were an abuse of process. The proceedings were also to be contrasted with Jameel (para 81). The claimant had a clear and legitimate interest in vindicating his reputation against a most serious accusation made by the government of the state of which he was a citizen and where he resided. The publication was by no means minimal, and was from an authoritative and apparently trustworthy source. The publicity prior to the ICT conviction in 2013, including the Channel 4 programme, was of little relevance as it was some time ago and was superseded by the publicity given to the conviction. That publicity in respect of the conviction generally made plain that it was in the claimant’s absence, that he denied the charges, and that he maintained that the proceedings before the ICT were politically motivated and unfair. The Report had made the first unequivocal and unqualified assertion of the claimant’s guilt in this jurisdiction, and success in the present proceedings would be a major vindication. The fact that the statement complained of appeared in a footnote was of little or no significance: it was clear that it had been published relatively widely and that it was drawn to the claimant’s attention by a number of people, indicating its significant penetration. As for the supposed unfairness to the Secretary of State, if the evidence available was insufficient to support a defence of truth, that was not something of which the Secretary of State could legitimately complain, having chosen to endorse the ICT conviction long after the event. Furthermore, the approach of the majority, which relied on a combination of factors none of which amounted to an abuse on its own, was unprincipled. Finally, having been denied a fair trial in Bangladesh in 2013, it would be unfortunate if the claimant was now prevented from having access to the courts of this jurisdiction, at least in part because of his conviction by the ICT.
The grounds of appeal
The claimant challenges the decision of the Court of Appeal on the following grounds:
The Court of Appeal erred in law in its finding that a foreign criminal conviction which a claimant did not have a full opportunity to contest (and which therefore was not within the Hunterabuse doctrine) was a relevant factor in showing that English proceedings, which require determination of the same issues as those before the foreign criminal court, are an abuse of process.
The Court of Appeal erred in law in finding that prior press publications of defamatory allegations are admissible (and in this case conclusive) evidence of bad reputation if such publications have taken place some months prior to the publication complained of and are uncontradicted by a successful claim for libel.
The Court of Appeal erred in law in finding, as a relevant factor supporting a finding of abuse, potential difficulties which the Secretary of State may have in proving the truth of the allegations which the Secretary of State had published about the conduct of the claimant some 50 years ago.
The Court of Appeal erred in law and in principle in finding that an unidentified combination of partial aspects of the Hunter and Jameelabuse jurisdictions, none of which necessarily amounts to abuse on its own, can properly ground a finding of abuse of process.
No distinct points are raised in relation to the claim made under the GDPR, which is agreed to stand or fall with the claim in defamation.
Abuse of process
As Lord Bingham of Cornhill stated in Johnson v Gore Wood & Co [2002] 2 AC 1, 22:
“The rule of law depends upon the existence and availability of courts and tribunals to which citizens may resort for the determination of differences between them which they cannot otherwise resolve. Litigants are not without scrupulous examination of all the circumstances to be denied the right to bring a genuine subject of litigation before the court”.
As that dictum indicates, individuals have a fundamental right of access to the courts for the determination of their civil rights. That right has been recognised by the common law for many centuries, and has been protected by statute from Magna Carta (parts of which remain on the statute book, in the version issued in 1297) to the Human Rights Act 1998 (“the Human Rights Act”).
However, it is not an unqualified right. As Lord Bingham went on to make clear, it does not mean that the court must hear in full and rule on the merits of any claim or defence which a party to litigation may put forward. For there is, as Lord Diplock stated in Hunter [1982] AC 529, 536, in a speech with which the other members of the House of Lords agreed, an
“inherent power which any court of justice must possess to prevent misuse of its procedure in a way which, although not inconsistent with the literal application of its procedural rules, would nevertheless be manifestly unfair to a party to litigation before it, or would otherwise bring the administration of justice into disrepute among right-thinking people.”
That dictum is not to be treated as if it were a statutory definition. Nevertheless, Lord Diplock described the nature of the court’s power with characteristic clarity, and in terms which have stood the test of time over the past half century, both in this jurisdiction and overseas (see, for example, Walton v Gardiner [1993] HCA 77; (1993) 177 CLR 378 and City of Toronto v Canadian Union of Public Employees [2003] 3 SCR 77). Neither party to these proceedings has argued that this court should depart from Lord Diplock’s statement of principle or attempt to improve upon it.
Two aspects of Lord Diplock’s statement should be noted. First, the power in question is a power to prevent misuse of the court’s procedure. It follows that the power cannot be exercised if the claimant is making proper use of the civil jurisdiction of the court to protect his rights. Secondly, the court’s procedure must be being misused in a way which would be “manifestly unfair” to one or more of the parties or would otherwise “bring the administration of justice into disrepute among right-thinking people”. The primary purpose of the doctrine, in other words, is to preserve public confidence in the administration of justice.
The law in relation to abuse of process has developed in the manner characteristic of the common law. Relevant principles have emerged as the courts have considered the circumstances of cases in which the issue has arisen in different contexts. As Lord Diplock indicated in Hunter at p 536, it would be unwise to confine the concept of an abuse of process to fixed categories. Nevertheless, a number of categories have become well established. Examples include the unfair or oppressive treatment of an accused, the rule in Henderson v Henderson (1843) 3 Hare 100 that requires a party to bring its whole case in a single set of proceedings, and the power to stay or dismiss proceedings which are frivolous or vexatious.
In the present case, counsel for the Secretary of State argued that the proceedings fall within an established category, which counsel termed “Hunter abuse”, concerned with collateral attacks on criminal convictions. That argument was buttressed, as in the courts below, by a submission that it would be manifestly unfair for the Secretary of State to have the burden of establishing the claimant’s guilt so long after the events in question, in order to establish a defence of truth. A secondary argument, only faintly advanced, was that the claimant’s reputation was so badly damaged by his conviction by the ICT and by previous publicity, and the resources required to litigate the matter would be so great, that “the game was not worth the candle”, so that the case was an example of “Jameel abuse”. Each of those arguments will be considered in turn.
Before doing so, one other general principle should be noted. As Simon Brown LJ stated in Broxton v McClelland [1995] EMLR 485, 498, in a dictum which has been cited many times:
“Only in the most clear and obvious case will it be appropriate upon preliminary application to strike out proceedings as an abuse of process so as to prevent a plaintiff from bringing an apparently proper cause of action to trial.”
Hunter abuse
The decision in Hunter
The case of Hunter concerned an action for assault, brought against the police by one of six men who had previously been convicted of terrorist offences. Their convictions were based partly on their admissions to the police. At their criminal trial they contended that the admissions had been beaten out of them. The judge, after a voir dire, held that the men had not been assaulted. The jury’s verdict implied that they too had rejected the allegations of assault. The men applied for leave to appeal, but not on the ground that their confessions had been wrongly admitted. The subsequent civil action was accordingly an attempt to relitigate in a civil court an issue which had been in dispute at the criminal trial and had been decided against the plaintiff. The civil claim was held to be an abuse of process.
Lord Diplock laid down what has since been regarded as the governing rule on this subject ([1982] AC 529, 541):
“The abuse of process which the instant case exemplifies is the initiation of proceedings in a court of justice for the purpose of mounting a collateral attack upon a final decision against the intending plaintiff which has been made by another court of competent jurisdiction in previous proceedings in which the intending plaintiff had a full opportunity of contesting the decision in the court by which it was made.”
By “collateral” Lord Diplock meant an attack not made in the original proceedings which gave rise to the decision which the subsequent proceedings sought to impugn: an attack, in other words, which was not made by way of appeal in those original proceedings. Lord Diplock’s statement of the relevant principle was not confined to collateral attacks on criminal convictions, and the authorities on which he relied were concerned with collateral challenges to previous decisions in civil proceedings.
Lord Diplock also made it clear at p 545 that there could be exceptions to “the general rule of public policy that the use of civil actions to initiate collateral attacks on final decisions against the intending plaintiff by criminal courts of competent jurisdiction should be treated as an abuse of the process of the court”. In particular, he acknowledged that an exception would be justified where the plaintiff had come into possession of fresh evidence which was not available in the original proceedings and could not with reasonable diligence have been obtained then, but which entirely changed the aspect of the case. That exception was derived from the speech of Lord Cairns LC in Phosphate Sewage Co Ltd v Molleson (1879) 4 App Cas 801, 814.
It is to be noted that the convictions in issue in Hunter were subsequently quashed by the Court of Appeal in the light of fresh evidence casting doubt on scientific evidence given at the trial: (1991) 93 Cr App R 287; [1992] 2 All ER 417. That does not undermine the decision of the House of Lords to any extent. Safeguards are necessary for the correction of miscarriages of justice, but collateral challenges in a different forum are not the appropriate means. By contrast, review of a conviction by way of appeal affirms both the authority of the criminal process and the finality of the result.
Application to foreign convictions
Hunter was concerned with proceedings impugning a decision taken in earlier criminal proceedings in this country. It is not in dispute on this appeal that the Hunter principle can also apply where a decision in foreign proceedings is impugned. However, whereas the risk that two inconsistent decisions of courts in the United Kingdom might bring the administration of justice into disrepute is evident, the risk may be less evident where the previous decision is that of an overseas court. Accordingly, it was submitted on behalf of the claimant that the Hunter principle could not be applied to foreign convictions without qualifications.
When Parliament made convictions admissible evidence in civil proceedings, by section 11 of the Civil Evidence Act 1968 (“the 1968 Act”), and made convictions conclusive evidence in defamation proceedings, by section 13, it did so in order to avoid the risk that conflicting decisions by criminal and civil courts would undermine public confidence in the administration of justice. That was explained in the Fifteenth Report of the Law Reform Committee, Cmnd 3391 (1967), which led to the enactment of the 1968 Act: see para 29. That is the same public policy as underpins the principle set out in Hunter. Parliament also confined the scope of those provisions to convictions by courts in the United Kingdom (and convictions of service offences): sections 11(1) and 13(3). The reasons for that limitation—which Parliament retained in force when it amended section 13 by the Defamation Act 1996—were also explained in the Report of the Law Reform Committee (para 17):
“The substantive criminal law varies widely in different countries. So does criminal procedure and the law of evidence. The relevance of the foreign conviction to the issues in the English civil action could not be ascertained without expert evidence of the substantive criminal law of the foreign country. Its weight could not be judged without expert evidence of the procedural law of the foreign country and reliable information as to the standards of its courts. There are, of course, many countries whose standard of administration of criminal justice is as high as our own, but there are others in which one cannot be assured of this. It would be invidious to draw distinctions between one foreign country and another in the legislation needed to give effect to our recommendations. It would be impracticable to leave the admissibility and weight of a foreign conviction to the discretion of an English judge unfamiliar with the legal system and standards of criminal justice of the foreign country concerned. Furthermore, the burden of showing that a foreign conviction was erroneous would be difficult, perhaps impossible, to sustain, since there would be no way of compelling the witnesses in the foreign criminal proceedings to attend to give evidence in the English courts. The practical effect of making foreign convictions admissible might well be to make them conclusive, and the remoter the country in which the conviction took place the more difficult it would be to dispute its correctness.”
Those are important considerations. It is because Parliament did not extend sections 11 and 13 to foreign convictions that the Secretary of State relies instead in these proceedings on Hunter abuse, in the hope that the courts will give the ICT conviction an unchallengeable status, in circumstances where Parliament has chosen not to do so.
The factors mentioned by the Law Reform Committee underline the importance, in relation to foreign convictions, of the requirement laid down by Lord Diplock that “the intending plaintiff had a full opportunity of contesting the decision in the court by which it was made” (Hunter, p 541): a full opportunity, in this context, including a fair opportunityfor the accused to put his or her case and to meet the case made against him or her. As Lord Woolf MR stated in Hamilton v Al Fayed [1999] 1 WLR 1569, 1582, the court must therefore judge the procedural quality of the earlier lis.
Cases since Hunter
Hunter abuse was considered in two subsequent decisions of the House of Lords. First, the case of Arthur J S Hall & Co v Simons [2002] 1 AC 615 concerned actions of damages brought against solicitors on the basis that they had negligently advised the plaintiffs to accept settlements of earlier claims which were then approved by the courts and reflected in court orders. The solicitors relied on the advocate’s immunity from suit. In response, it was argued that the immunity was no longer justified, since the principle enunciated in Hunter was sufficient to prevent any action being maintained which would bring the administration of justice into disrepute.
In the course of his speech (with which Lord Browne-Wilkinson agreed), Lord Steyn referred to the Hunter principle as applying to “defendants convicted after a full and fair trial” (p 679). He observed (ibid) that the principle underlying the Hunter case “must be maintained as a matter of high public policy”, but added, citing a dictum of Sir Thomas Bingham MR in Smith v Linskills [1996] 1 WLR 763, 769, that in Hunter “the House did not, however, ‘lay down an inflexible rule to be applied willy-nilly to all cases which might arguably be said to be within it’”.
Lord Browne-Wilkinson identified the policy which required collateral challenges to criminal convictions to be treated as an abuse of process as being that they brought the administration of justice into disrepute (p 685):
“… where the later civil action must, in order to succeed, establish that a subsisting conviction is wrong, in the overwhelming majority of cases to permit the action to continue would bring the administration of justice into disrepute.”
That was because the proceedings would, if they succeeded, result in two conflicting decisions of the court, in a context where the decision of the criminal court must be treated as authoritative for all the purposes of society unless and until it is set aside on appeal (p 684).
Lord Hoffmann, with whose speech Lord Browne-Wilkinson, Lord Millett and (on this issue) Lord Hutton agreed, identified the problem arising from collateral challenges to criminal convictions as being “the possibility of apparently conflicting judgments which could bring the administration of justice into disrepute” (p 687; see also p 706). Lord Hoffmann did not consider the difficulty of fairly retrying at a later date the issue which was before the court on the earlier occasion to be a substantial argument that it might be contrary to the public interest. He commented that, in principle, evidential difficulties have never been regarded as a reason for refusing to try a case (p 699).
Like Lord Browne-Wilkinson, Lord Hoffmann regarded criminal proceedings as being in a special category, because “although they are technically litigation between the Crown and the defendant, the Crown prosecutes on behalf of society as a whole” (p 702). However, he cautioned against exaggerating this argument, commenting that although a conviction has some of the quality of a judgment in rem which should be binding in favour of everyone, “it remains a judgment between the Crown and the accused and that is often the right way to consider it” (ibid).
Lord Hoffmann expressed his agreement with Lord Diplock (para 37 above) that the power to strike out proceedings as an abuse of process should be exercised only in cases in which re-litigation of an issue previously decided would be “manifestly unfair” or would otherwise bring the administration of justice into disrepute. He emphasised the need for flexibility and the exercise of judgment, observing (p 703):
“But I do not think that he meant that every case falling within this description was an abuse of process or even that there was a presumption to this effect which required the plaintiff to bring himself within some exception. That would be to adopt a scheme of categorisation which Lord Diplock deplored. As I shall explain, I think it is possible to make some generalisations about criminal proceedings. But each case depends upon an application of the fundamental principles. I think that Ralph Gibson LJ was right when, after quoting this passage, he said in Walpole v Partridge & Wilson [1994] QB 106, 116A that Hunter’s case [1982] AC 529 decides ‘not that the initiation of such proceedings is necessarily an abuse of process but that it may be’.”
In relation to criminal proceedings, Lord Hoffmann went on to say that although “it would ordinarily be an abuse of process for a civil court to be asked to decide that a subsisting conviction was wrong”, nevertheless “there are bound to be exceptional cases in which the issue can be tried without a risk that the conflict of judgments would bring the administration of justice into disrepute” (p 706). The case of Walpole v Partridge & Wilson [1994] QB 106 was cited as an example.
Lord Hobhouse of Woodborough stressed two important points. The first was that the essence of an abuse of process is “the use of civil litigation for an improper purpose, ie without a legitimate purpose” (p 742). The second was that “collateral attack” is a distinct concept from abuse of process: “challenging a previous decision does not necessarily connote an abuse of process” (ibid). In that regard, Lord Hobhouse observed at p 743 that “[t]here is no general rule preventing a party inviting a court to arrive at a decision inconsistent with that arrived at in another case”. Rather, as he pointed out, there are legal principles, such as those of res judicata and issue estoppel, which apply in particular circumstances. Lord Hobhouse also emphasised the different character of the criminal process from civil proceedings, and commented (p 745):
“It is of fundamental importance that the process by which the defendant is proved guilty shall have been fair and it is the public duty of all those concerned in the criminal justice system to see that this is the case. This is the public interest in the system.”
In other words, the public interest in upholding a criminal conviction is premised on its having been fairly obtained.
The second relevant decision of the House of Lords subsequent to Hunter illustrates some of the points just made. The case of In re Norris [2001] UKHL 34; [2001] 1 WLR 1388 (“Norris”) concerned a confiscation order made by the Crown Court following the defendant’s conviction of drug offences. At the hearing before the Crown Court, the defendant’s wife gave evidence that the matrimonial home belonged wholly or partly to her. The judge disbelieved her and made the confiscation order on the basis that the house formed part of the defendant’s realisable property. On a subsequent application to the High Court to enforce the order, the wife made an application to vary it on the ground that she had an interest in the house. The judge dismissed the application, and the Court of Appeal held that it was an abuse of process falling within the principle explained in Hunter. The court reached that conclusion on the basis that the issue had already been decided by the Crown Court, and the wife, although not a party to the criminal proceedings, had had a fair opportunity to put her case. That decision was unanimously reversed by the House of Lords.
Lord Hobhouse, with whose speech the other members of the House of Lords agreed, pointed out at para 23 that the wife was a mere witness in the criminal court, with no right of representation, no control of the proceedings, and no right of appeal. It was only in the High Court that she had the right to invoke the remedies of the court in defence of her civil rights. Lord Hobhouse observed at para 25:
“Judge Brown [the judge in the Crown Court] was not engaged in an exercise of determining her rights in accordance with the civil law. Because of this, she was never given, nor was it intended that she should be given, the right in the Crown Court to place before Judge Brown, through counsel representing her and supported by the documentary and other evidence which she chose, her civil case. … Mrs Norris is not ‘misusing’ the procedure of the High Court; she is making the proper use of the civil jurisdiction of the High Court to protect her proprietary rights”.
Lord Hobhouse distinguished the situation in Hunter, statingat para 26:
“In Hunter [1982] AC 529 the plaintiff was engaged in trying to relitigate in a civil court a factual issue which had already been decided against him in a criminal case in which he had been a party. It involved a collateral attack upon a decision in previous proceedings to which he had been a party, fully represented and with complete control over the evidence he wished to put before the court. The plaintiff had ‘had a full opportunity of contesting the decision in the court by which it was made’: per Lord Diplock, at p 541. The present case does not have those features.”
In relation to the principle of abuse of process, he added (ibid):
“Any such abuse must involve something which amounts to a misuse of the litigational process. Clear cases of litigating without any honest belief in any basis for doing so or litigating without having any legitimate interest in the litigation are simple cases of abuse. Attempts to relitigate issues which have already been the subject of judicial decision may or may not amount to an abuse of process. Ordinarily such situations fall to be governed by the principle of estoppel per rem judicatam or of issue estoppel (admitted not to be applicable in the present case). It will be a rare case where the litigation of an issue which has not previously been decided between the same parties or their privies will amount to an abuse.”
This decision underlines how important it is, if proceedings are to be characterised as an abuse of process, that the claimant should have had what Lord Diplock described in Hunter, at p 541,as “a full opportunity of contesting the decision [in the previous proceedings] in the court by which it was made”. It also makes it clear that challenging a previous decision, even of a criminal court, does not necessarily connote an abuse of process. Not all re-litigation of the same issue will bring the administration of justice into disrepute.
Application to the present case
In the light of these authorities, it is plain that the present proceedings cannot be regarded as an abuse of process of the kind described in Hunter. As I have explained, Lord Diplock’s statement of the relevant principle, cited at para 44 above, included as an essential element that the plaintiff “had a full opportunity of contesting the decision [in the previous proceedings] in the court by which it was made”. That requirement was met in Hunter:the plaintiff had participated fully in the criminal proceedings against him and had an unrestricted ability to appeal against his conviction. The fairness of the proceedings was unimpeached. The same was true in subsequent cases where the principle laid down in Hunter was applied. In the present case, on the other hand, the claimant was tried in absence before the ICT. He could not realistically be expected to attend the trial or any subsequent appeal, since he faced a real risk of execution. Counsel appointed to represent him was seemingly either unwilling or unable to obtain his instructions. It is unclear what grounds, if any, he might have advanced on appeal, since he was debarred by the 1973 Act from relying on the procedural guarantees contained in the Bangladesh Constitution or on the Criminal Procedure Code or the Evidence Act, and was unable to challenge the ICT’s decisions as to the admission of evidence (para 13 above). Furthermore, it is said on behalf of the claimant, without contradiction, that success in any appeal would in all likelihood have resulted in a retrial, at which he would again have been unable to appear because of the risk of execution, and at which the rules of evidence would again have been disapplied. In these circumstances, he did not have a full opportunity of contesting the decision of the ICT in the proceedings in which it was made. Accordingly, the fact that the present proceedings involve a challenge to the outcome of the trial before the ICT cannot render them an abuse of process.
I need only add in relation to this aspect of the case that it is surprising that the Secretary of State should submit that even if the proceedings before the ICT were unfair—even, indeed, if there was a gross miscarriage of justice—a challenge to the conviction should nevertheless be struck out as an abuse of process. As Lord Hobhouse stated in Arthur J S Hall & Co v Simons (para 58 above), it is of fundamental importance that the process by which the defendant is proved guilty shall have been fair. That reflects the fact that the doctrine of abuse of processis designed to prevent the court’s procedure being misused in a way which, in Lord Diplock’s words, would bring the administration of justice into disrepute among right-thinking people. In other words, the doctrine is designed to preserve public confidence in the administration of justice. The submission advanced on behalf of the Secretary of State would have precisely the opposite effect.
Manifest unfairness
As explained earlier, Lord Diplock’s description of abuse of process is of the misuse of the court’s procedure in a way which would be manifestly unfair to a party to litigation before it or would otherwise bring the administration of justice into disrepute among right-thinking people. The focus of the doctrine is on public confidence in the administration of justice rather than on the interests of a party, as Lord Diplock’s description implied (and as has been more fully explained in later cases such as City of Toronto, cited in para 38 above), but the two are directly connected where the court’s procedure is being misused in a manner which is manifestly unfair.
In the present case, the majority of the Court of Appeal considered that it was manifestly unfair for the Secretary of State to have to prove the claimant’s guilt, so long after the events in question, in order to establish a defence of truth. I am unable to agree. But for the fact that the courts below reached a different conclusion, I should have regarded the Secretary of State’s submission that the claimant’s action is an abuse of process because it is difficult for him to establish his proposed defence as unarguable. It is difficult to accept that, if the Secretary of State is unable to establish the truth of his allegations against the claimant, therefore he can defame the claimant with impunity.
There are a number of other reasons for rejecting the submission. In the first place, this is not in any sense a stale claim. The relevant part of the Report is concerned, as its heading indicates, with “What Extremism Looks Like in England and Wales” at the time of publication in 2019. The footnote which is the subject of the claim is said to show links between those responsible for the violence in Bangladesh in 1971 and current community leadership in East London. The claimant issued his claim promptly following the publication of the Report and has prosecuted it expeditiously. The long interval of time between the events in question and the date of any trial of the action, and any evidential difficulties which may result from the length of that period, are not matters for which the claimant bears any responsibility. They are entirely the result of the Secretary of State’s decision to publish defamatory allegations about the claimant more than 50 years after the events in question.
Secondly, where truth is pleaded as a defence, the burden lies on the defendant to prove the truth of the defamatory imputation. The burden of proof allocates the risk of an insufficiency of evidence to the defendant. If the Secretary of State is unable to establish the truth of the accusation which the Home Office chose to publish concerning the claimant’s conduct more than 50 years ago, that should have been considered before the Report was published.
A further point is that the Court of Appeal ([2022] EMLR 23) had no information before it (as Phillips LJ recorded at para 82(iii)) as to the availability of witnesses or other evidence. There was no pleaded defence, nor even an indication in outline of the evidence which the Secretary of State might seek to rely on. That was because no investigations had yet been carried out. However, as mentioned earlier (para 19 above), the ICT heard evidence in 2013 from around 20 witnesses, and statements made by deceased witnesses prior to the trial were also available. The supposed evidential difficulties are, in the circumstances, a matter of speculation.
It is in any event relatively common, in both civil and criminal proceedings, for cases to be fairly tried notwithstanding a substantial interval of time between the proceedings and the events in question. Familiar examples include, in the civil context, trespass claims where there is a defence of adverse possession, and nuisance claims where there is a defence of prescription; and, in the criminal context, cases of war crimes (eg R v Sawoniuk [2000] 2 Cr App R 220) and of historical sexual abuse. As Lady Hale stated in A v Hoare [2008] UKHL 6; [2008] 1 AC 844, para 60, “[a] fair trial can be possible long after the event and sometimes the law has no choice”. Fairness is guaranteed by the trial process itself, which will involve the careful weighing of the available evidence in the light of all relevant circumstances, including the passage of time.
In the context of the law of defamation, in particular, there are numerous examples of libel claims brought over allegations concerning conduct many years earlier. Well-known examples include Plato Films Ltd v Speidel [1961] AC 1090, where there was a 25 year interval between the events in question and the publication; Broome v Cassell & Co Ltd [1972] AC 1027, where the interval was about 28 years; Tolstoy Miloslavsky v United Kingdom (1995) 20 EHRR 442, where the interval was 44 years; Irving v Penguin Books Ltd [2001] EWCA Civ 1197, where the interval was over 50 years; and Polanski v Condé Nast Publications Ltd [2005] UKHL 10; [2005] 1 WLR 637, where the interval was 33 years.
Jameel abuse
The relevant case law and legislation
In the case of Jameel [2005] QB 946, a Saudi Arabian claimant sued the US publisher of the Wall Street Journal over an article published online. The article was shown to have reached just five people in England and Wales, three of whom had been acting on behalf of the claimant in relation to his claim. Any harm to the claimant’s reputation in England and Wales was minimal. In reality, this was a case of forum shopping. Minimal publication in England was being used to found jurisdiction to litigate a dispute with no real connection to this country, but which the claimant could not bring in the US because of the protection of freedom of speech under the First Amendment. However, under English law, at common law (prior to the legislative reform described in para 77 below) libel is one of the categories of defamation which are actionable per se: in other words, at common law libel is actionable without proof of damage. In such cases, the common law presumes that the claimant’s reputation has been damaged (Ratcliffe v Evans [1892] 2 QB 524, 529), and awards damages that are not only compensatory but also serve to vindicate the claimant’s reputation. That rule of the common law is long established, and supported by authority at the highest level: see, for example, Shevill v Presse Alliance SA [1996] AC 959, 982-983 and Berezovsky v Michaels [2000] 1 WLR 1004, 1012.
In Jameel, the Court of Appeal rejected a submission that for the court to apply the common law rule that damage is presumed would be contrary to its duty under section 6 of the Human Rights Act, as it would be incompatible with the right to freedom of expression under article 10 of the European Convention on Human Rights (“the Convention”). Lord Phillips of Worth Matravers MR, delivering the judgment of the court, observed (para 37) that “English law has been well served by a principle under which liability turns on the objective question of whether the publication is one which tends to injure the claimant’s reputation”. However, the court accepted “that in the rare case where a claimant brings an action for defamation in circumstances where his reputation has suffered no or minimal actual damage, this may constitute an interference with freedom of expression that is not necessary for the protection of the claimant’s reputation” (para 40). The appropriate remedy for the defendant in such circumstances might well be to challenge the claimant’s resort to English jurisdiction on the ground that no real and substantial tort had been committed within the jurisdiction (which, in Jameel, the defendant had failed to do), to seek to strike out the action as an abuse of process, or to ask the court to subject the claimant to costs sanctions (ibid).
In relation to the second of those possibilities, Lord Phillips MR referred to two developments which, he said, had rendered the court more ready to entertain a submission that pursuit of a libel action was an abuse of process. The first was the introduction of the Civil Procedure Rules (“CPR”), which required the court to adopt a more flexible and proactive approach to litigation (para 55). The second was the coming into effect of the Human Rights Act, section 6 of which required the court to administer the law in a manner which was compatible with Convention rights, so far as it was possible to do so. Lord Phillips MR continued (ibid):
“Keeping a proper balance between the article 10 right of freedom of expression and the protection of individual reputation must, so it seems to us, require the court to bring to a stop as an abuse of process defamation proceedings that are not serving the legitimate purpose of protecting the claimant’s reputation, which includes compensating the claimant only if that reputation has been unlawfully damaged.”
Article 10 of the Convention, as given effect by the Human Rights Act, therefore required the court to stay or dismiss proceedings, as an abuse of process, where the proceedings were not serving the legitimate purpose of protecting the claimant’s reputation. It was well established before Jameel that an action which was not being pursued for a proper purpose might be an abuse of process (paras 57–58). On the facts of the case, it was clear that, if the claimant succeeded in his action, it could “perhaps be said that he will have achieved vindication for the damage done to his reputation in this country, but both the damage and the vindication [would] be minimal” (para 69). Lord Phillips MR added (ibid) that “[t]he game will not merely not have been worth the candle, it will not have been worth the wick”. In other words, the action could not achieve anything of value.
The judgment in Jameel was considered in Thornton v Telegraph Media Group Ltd [2010] EWHC 1414 (QB); [2011] 1 WLR 1985 (“Thornton”), a decision of Tugendhat J. The judge treated the reasoning in Jameel as being based on article 10 of the Convention (para 61). He went on to hold that there was a threshold of seriousness to be surmounted before a statement could be regarded as meeting the legal definition of “defamatory”, so as to exclude trivial claims (para 90). This was considered to be in accordance with the true interpretation of Lord Atkin’s speech in Sim v Stretch [1936] 2 All ER 1237, and to be “required by the development of the law recognised in [Jameel] as arising from the passing of the Human Rights Act: regard for article 10 and the principle of proportionality both require it” (para 90).The judge explained (para 94) that this also explained why in libel the law presumed that damage had been suffered by a claimant:
“If the likelihood of adverse consequences for a claimant is part of the definition of what is defamatory, then the presumption of damage is the logical corollary of what is already included in the definition. And conversely, the fact that in law damage is presumed is itself an argument why an imputation should not be held to be defamatory unless it has a tendency to have adverse effects upon the claimant.”
The judge added that if there was not already a threshold of seriousness in the definition of “defamatory”, Jameel required that the definition included such a threshold (para 95). The judge’s definition was that a statement “may be defamatory of him [ie the claimant] because it substantially affects in an adverse manner the attitude of other people towards him, or has a tendency so to do” (para 96).
Jameel was considered by the Court of Appeal in Lait v Evening Standard Ltd [2011] EWCA Civ 859; [2011] 1 WLR 2973, where Laws LJ stated (para 42):
“The principle identified in [Jameel] consists in the need to put a stop to defamation proceedings that do not serve the legitimate purpose of protecting the claimant’s reputation. Such proceedings are an abuse of the process. The focus in the cases has been on the value of the claim to the claimant; but the principle is not, in my judgment, to be categorised merely as a variety of the de minimis rule tailored for defamation actions. Its engine is not only the overriding objective of the Civil Procedure Rules but also, in Lord Phillips MR’s words, the need to keep ‘a proper balance between the article 10 right of freedom of expression and the protection of individual reputation’.”
This dictum again focuses on whether the action is being pursued for a legitimate purpose, and on the need to comply with article 10 of the Convention.
Following these decisions, Parliament enacted the 2013 Act, section 1 of which provides:
“Serious harm
(1) A statement is not defamatory unless its publication has caused or is likely to cause serious harm to the reputation of the claimant.
(2) For the purposes of this section, harm to the reputation of a body that trades for profit is not ‘serious harm’ unless it has caused or is likely to cause the body serious financial loss.”
These developments were considered by this court in Lachaux vIndependent Print Ltd (Media Lawyers Association intervening) [2019] UKSC 27; [2020] AC 612 (“Lachaux”). Lord Sumption explained, in a judgment with which the other members of the court agreed, that a working definition of what makes a statement defamatory, derived from the speech of Lord Atkin in Sim v Stretch, at p 1240, is that “the words tend to lower the plaintiff in the estimation of right-thinking members of society generally” (para 6). The “important” decisions in Jameel and Thornton “added a further requirement, namely that the damage to reputation in a case actionable per se must pass a minimum threshold of seriousness” (para 7). In Jameel, the Court of Appeal “held that the presumption [of general damage] could not be applied consistently with the Convention in those cases, said to be rare, where damage was shown to be so trivial that the interference with freedom of expression could not be said to be necessary for the protection of the claimant’s reputation” (para 8). The effect of the decision was “to introduce a procedural threshold of seriousness to be applied to the damage to the claimant’s reputation” (para 8). Two things were clear from the judgment in Jameel (ibid):
“One is that the threshold was low. The damage must be more than minimal. That is all. Secondly, the Court of Appeal must have thought that the operation of the threshold might depend, as it did in the case before them, on the evidence of actual damage and not just on the inherently injurious character of the statement in question.”
Lord Sumption went on to say that the effect of the decision in Thornton was that “in addition to the procedural threshold of seriousness recognised in Jameel, there was a substantive threshold of seriousness to be surmounted before a statement could be regarded as meeting the legal definition of ‘defamatory’” (para 9). Citing the dictum quoted at para 75 above, Lord Sumption observed that it “suggested that (unlike the Jameel test) the application of the threshold depended on the inherent propensity of the words to injure the claimant’s reputation”(ibid). Lord Sumption noted that section 1 of the 2013 Act raised the threshold of seriousness above that envisaged in Jameel and Thornton, and required its application to be determined by reference to the actual facts about the impact of the publication and not justto the meaning of the words (para 12).
Given the decision in Thornton, and the enactment of section 1 of the 2013 Act, one might have thought that cases in which the principle laid down in Jameel, as interpreted by this court in Lachaux, might continue to be relevant were likely to be few and far between. However, in this case it is Jameel, rather than the 2013 Act or Thornton, on which reliance has been placed, as the Secretary of State seeks to have the claim struck out as an abuse of process, rather than seeking summary judgment on the ground that the claimant has no real prospect of establishing an essential ingredient of his cause of action.
The first important point about Jameel, as was noted in Thornton and Lachaux, is that the problem which gave rise to the need to strike out the claim was that for the court to allow the proceedings to continue in the absence of more than minimal damage would have been incompatible with the defendant’s rights under article 10 of the Convention, and therefore contrary to the court’s duty under section 6 of the Human Rights Act. The second point to note is that the dicta of Lord Phillips MR cited in paras 73 and 74 above, and the explanation of Jameel given by Lord Sumption in Lachaux, did not import a test based upon weighing the value of the claim against the cost of the proceedings. Although Lord Phillips MR’s use of the metaphor that “the game was not worth the candle” seems to have given rise to that misapprehension, the point of the metaphor is to underline the trivial nature of what is at stake. The dicta focused upon the question whether the damage to the claimant’s reputation was shown to be so trivial that the proceedings could not be regarded as serving the legitimate purpose of protecting the claimant’s reputation, with the consequence that the interference with the defendant’s freedom of expression could not be said to be necessary. The game was not worth the candle, if one wants to use that metaphor, because the action could not achieve, to any significant extent, the legitimate objective of protecting the claimant’s reputation in this jurisdiction.
The application of Jameel in the present case
As explained earlier, Jameel abuse was only faintly argued in this appeal. However, as it was found to be established by the judge, and the majority of the Court of the Appeal treated it as a factor which established abuse when considered along with other matters, it is necessary to address it, while bearing in mind that the court has not heard full argument.
Since the Secretary of State is a public authority, he has no right to freedom of expression which is protected by article 10 of the Convention. However, Dingemans LJ considered that article 10 nevertheless applied, on the basis that the publication of the Report engaged the article 10 rights of those who read it (para 56). It was therefore necessary, on his approach, for the court to balance the claimant’s right to the protection of his reputation against the article 10 rights of the readers of the Report. This approach was challenged on this appeal on a number of grounds. First, it was argued that the interference with article 10 rights with which Jameel was concerned was the fact of being sued; and the readers of the Report are not being sued. Secondly, it was argued that since the Secretary of State does not seek to be free to repeat the allegation against the claimant, the public’s right to be informed is not in jeopardy. Thirdly, it was argued that the public’s right under article 10 includes the right not to be misinformed; and the law protects that right by permitting libel claims to be brought.
In relation to the first of these arguments, while it is unnecessary to express a concluded view, I am not at present persuaded that the potential for article 10 to be infringed by trivial defamation claims is necessarily restricted to cases where the Convention rights in question are those of the defendant. I am therefore inclined to agree with Dingemans LJ that Jameel abuse may be relevant even in a case where the defendant is a public authority. In relation to the second argument, it is not only where the repetition of the libel is in issue that article 10 is engaged: see, for example, Tolstoy Miloslavsky v United Kingdom, where article 10 was held to be violated by the disproportionate amount of a damages award. In relation to the third argument, if the doctrine of Jameel abuse is confined to cases where the damage to the claimant’s reputation is so trivial that the proceedings cannot be regarded as serving the legitimate purpose of protecting the claimant’s reputation, consistently with the approach of Lord Phillips MR in Jameel (para 74 above) and Lord Sumption in Lachaux (para 78 above), then it will not exclude claims which genuinely serve a public interest in not being misinformed.
In their discussion of Jameel in relation to the present case, the majority of the Court of Appeal disclaimed any reliance on the principle that the pursuit of litigation for an improper purpose may constitute an abuse of process (para 42). They applied an approach which Dingemans LJ described as follows (para 45):
“It is necessary to assess what the litigation is worth to the claimant and the cost (in every sense of the word) of the litigation. Such cases are to be distinguished from valid claims of small value and cases where vindication is of importance to the claimant. In such cases the court should only conclude that continued litigation would be abusive where a way cannot be found to adjudicate the claim proportionately, and courts have power to control costs to attempt to ensure that they are proportionate.”
Since Dingemans LJ excluded from the scope of this approach “valid claims of small value”, it seems that this approach was considered to be confined to cases where the claim is not “valid”. If that means that this approach is restricted to claims where the statement of case discloses no reasonable grounds for bringing the claim, then such claims can be struck out on that ground, under CPR rule 3.4(2)(a). Dingemans LJ also excluded from the scope of this approach “cases where vindication is of importance to the claimant”. Since the proper purpose of a claim in defamation is to protect or vindicate the claimant’s reputation, it seems to follow that this approach is confined to claims which are not being pursued for a proper purpose. That takes one back to the well-established principle which Lord Phillips MR invoked in Jameel (para 74 above), and which Laws LJ treated as underpinning Jameel in Lait v Evening Standard Ltd (para 76 above), but which Dingemans LJ had earlier disclaimed.
It is also to be noted that, as explained earlier (para 81 above), the approach to abuse of process which was applied in Jameel and accepted in Lachaux was not based upon a comparison between the value of the claim and the cost of the litigation: an approach which could potentially have implications extending far beyond trivial claims or cases of defamation. As explained earlier, Lord Phillips MR spoke in Jameel of cases where the claimant’s reputation “has suffered no or minimal actual damage” (para 73 above). Consistently with that test, he also spoke of the need to stop as an abuse of process “defamation proceedings that are not serving the legitimate purpose of protecting the claimant’s reputation, which includes compensating the claimant only if that reputation has been unlawfully damaged” (para 74 above). To the same effect was Lord Sumption’s statement in Lachaux that the threshold introduced by Jameel is that “[t]he damage must be more than minimal” (para 78 above).
However, applying the approach which they adopted, the majority of the Court of Appeal found that an abuse of process was established when they concluded that the value of the claim was greatly exceeded by the cost of the litigation—that “the proceedings [were] not worth the candle of pursuing them” (para 60)—and considered that fact together with the fact that the claim involved a collateral attack on the ICT conviction, and the supposed unfairness to the Secretary of State of the likely shortage of evidence to support a defence of truth.
The factors leading the majority to conclude that the value of the claim was much less than the cost of the litigation were (a) that “Mr Mueen-Uddin’s reputation since he left Bangladesh at the end of December 1971 has been that he was involved in the murder of the intellectuals” (para 56), (b) that persons who read the relevant part of the Report were “overwhelmingly likely to have been aware of Mr Mueen-Uddin’s conviction by the [ICT] in Bangladesh and the contents of the Channel 4 Dispatches programme” (para 57), and (c) that “[w]hatever the judge concluded would not lead to a vindication of Mr Mueen-Uddin’s reputation, and this is because those who do not accept the result will point to the absence of relevant witnesses and evidence” (para 58). I shall consider each of those factors in turn. Before doing so, however, I would observe that by reason of section 1(1) of the 2013 Act, it is an essential element of the claimant’s cause of action that the publication of the Report has caused or is likely to cause serious harm to his reputation. If he had no real prospect of establishing this, as the majority of the Court of Appeal considered, the appropriate response would be to grant summary judgment in favour of the Secretary of State, rather than to strike out the claim as an abuse of process.
The claimant’s reputation
In relation to the claimant’s reputation prior to the publication of the Report, the first point to make is that this is a question of fact. At the stage of an application to strike out a statement of case prior to trial, the court is not generally in a position to make findings of fact. Its focus is on the claimant’s case. In this case, the particulars of claim narrate that the claimant is a senior and prominent member of the Muslim community in the United Kingdom. A number of positions which he has held are detailed in the amended particulars of claim and in his witness statement. They include the roles mentioned in para 1 above. According to his witness statement, he has met HRH the Prince of Wales, as he then was, in the course of his work and has attended a number of events at Buckingham Palace at the invitation of members of the Royal family, including a Garden Party in 2006. He states that he felt obliged to offer his resignation from a number of the charitable organisations he was involved with following his conviction by the ICT, but that many of them were reluctant to accept it. He states that two of them refused altogether to accept his resignation, making it clear that they did not believe the allegations made against him. He states:
“I was very pleased and relieved that the fact of the conviction had caused me limited reputational damage, and had not significantly harmed my ability to continue playing a part in British public life. I continued to be made welcome at my local mosque and other community groups to which I belonged such as the East London Mosque and the organisations I name above, and to enjoy the company of my friends and acquaintances much as I had before the decision of the ICT was announced, in a way which would have been simply inconceivable had my reputation been that of a war criminal.”
He adds that United Kingdom media coverage of the ICT conviction in most instances made clear his strong denial of the allegations and also included commentary about the criticisms of the ICT process.
Accordingly, on the face of the particulars of claim and the witness statement, the claimant is offering to prove that his reputation was not significantly damaged by the newspaper articles in 1971 or the Channel 4 programme in 1995 and that, on the contrary, he remained in good standing in society. On his averments, the damage caused to his reputation by media coverage of the ICT conviction also appears to have been limited. What is said in the particulars of claim and the witness statement is inconsistent with Dingemans LJ’s statements (para 56) that “it is apparent … that Mr Mueen-Uddin’s reputation since he left Bangladesh at the end of December 1971 has been that he was involved in the murder of the intellectuals”, and that, following the Channel 4 programme in 1995, “in England and Wales, Mr Mueen-Uddin’s reputation was that he had been involved in the murder of the intellectuals in Bangladesh in 1971”.
The second point to be made about the majority of the Court of Appeal’s approach to this issue is that they relied on the publication of the same allegation against the claimant on earlier occasions, treating it as not only admissible but conclusive evidence. In relation to this reasoning, it is necessary to consider the decision of the House of Lords in Dingle v Associated Newspapers Ltd [1964] AC 371 (“Dingle”) that other publications to the same effect as the words complained of are inadmissible in mitigation of damages.
It was argued on behalf of the Secretary of State that the rule in Dingle is irrelevant in the present context, because the issue in the appeal is abuse of process, not mitigation of damages. That contention cannot be accepted. The impact on the claimant’s reputation of the 1971 newspaper reports and, more particularly, the Channel 4 programme, was said by Dingemans LJ to be “relevant to a consideration of Jameel abuse, and whether the proceedings will be worth the candle” (para 56): in other words, it was relevant because of its effect on the value of the claim. The only way in which it could affect the value of the claim, where the claimant had been accused of the most serious crimes, would be by operating in mitigation of damages. The rule in Dingle therefore needs to be considered. Indeed, that was recognised by Dingemans LJ, who discussed the principles governing the mitigation of damages (paras 47–49), and explained why in his view the general rule laid down in Dingle did not apply in the present case. That explanation will be considered below.
The reasons for the rule which was reaffirmed in Dingle were explained by Lord Radcliffe (p 396):
“It is, I think, a well understood rule of law that a defendant who has not justified his defamatory statements cannot mitigate the damages for which he is liable by producing evidence of other publications to the same effect as his … A defamed man would only qualify for his full damages if he managed to sue the first defamer who set the ball rolling: and that, I think, is not and ought not to be the law.”
As Lord Radcliffe went on to explain, this view of the law has been accepted since the case of Saunders v Mills (1829) 6 Bing 213; 130 ER 1262, which, as he said, “has long been recognised and treated as an authority for the proposition that publication of the same libel by other persons on other occasions is irrelevant in mitigation of damages” (p 397).
The continuing validity of the Dingle rule was made clear by this court in Lachaux, para 24. As was said there, the effect of the rule “is to treat evidence of damage to the claimant’s reputation done by earlier publications of the same matter as legally irrelevant to the question what damage was done by the particular publication complained of.”
The last sentence in the passage from Lord Radcliffe’s speech quoted in para 94 above is inconsistent with the view adopted by the majority of the Court of Appeal in the present case: that because the claimant did not recover damages from Channel 4 in respect of its 1995 programme, he therefore cannot recover damages from the Secretary of State in respect of the allegations published in the Report.
The majority of the Court of Appeal considered that the Dingle rule did not apply in the present case. Dingemans LJ, at para 48, referred to a passage in the speech of Lord Cohen in Dingle, at p 406, which he treated as authority for the proposition that “if months had gone by after an uncontradicted report, especially if succeeded by other publications, that could be proved against the claimant” (para 48). It is not entirely clear whether Dingemans LJ meant that the uncontradicted report could be proved, but his next sentence (“It is apparent that defendants have been permitted to plead and rely on, to mitigate damages, unchallenged reports affecting the claimant’s reputation in the relevant sector”) suggests that he did. If so, that was a misreading. Nothing in Dingle supports the proposition that a previously published report becomes admissible evidence of bad reputation if it is months old and uncontradicted. One might add that the Channel 4 programme was not, in any event, uncontradicted: the claimant challenged the programme by issuing proceedings for libel: see para 14 above.
Dingle concerned proceedings in defamation which were brought against a newspaper in respect of an article it published. Some of the subject matter of the article had previously been published in a Parliamentary Select Committee report. It was held that evidence of the report was inadmissible in mitigation of damage. In the passage cited by Dingemans LJ, at para 48, Lord Cohen quoted the judgment of Holroyd Pearce LJ in the Court of Appeal ([1961] 2 QB 162, 180–181), as follows (p 406):
“Can it here be said that there was some evidence of the plaintiff’s bad or partially bad reputation? In my judgment there was none. Evidence of bad reputation must be properly proved. It is a grave matter. One is not entitled to assume without evidence that the plaintiff’s reputation must have changed in the four weeks following the publication of the report. As the months went by such report if uncontradicted would, no doubt, have its effect on his reputation, especially if it was succeeded by other similar matters; and the plaintiff’s reputation might in due course become bad and be proved against him … But when the defendants’ libel was published the plaintiff was a man of good reputation who had recently had something damaging attributed to him. A man’s reputation in the sense in which the word is used in civil or criminal courts does not alter daily as good or bad deeds are ascribed to him. It is the judgment of his fellows on his general life over a period.” (emphasis added)
I have italicised the passage cited by Dingemans LJ. Its import is not that an uncontradicted report might be proved: the whole point of the decision in Dingle was that the report could not be proved in mitigation of damage. As Holroyd Pearce LJ said earlier in his judgment (at p 179):
“The fact that other persons on previous occasions have published the same libel has been held irrelevant: Saunders vMills (1829) 6 Bing 213. Such evidence is inadmissible, even when coupled with evidence that the plaintiff took no steps to contradict the libel: Pankhurst vHamilton (1887) 3 TLR 500.”
That view was endorsed by Lord Cohen (p 406), and by all the other members of the Appellate Committee: see Lord Radcliffe (with whom Lord Morton of Henryton and Lord Cohen agreed) at p 396, cited at para 94 above; Lord Denning at pp 410–411; and Lord Morris of Borth-y-Gest at pp 416–417.
What Holroyd Pearce LJ meant in the dictum cited by Lord Cohen was that such a report, if uncontradicted, would over time affect the plaintiff’s reputation; and his reputation could, of course, be proved. That would be done in the conventional way, by the testimony of those who knew him, or knew of him, and could speak to his reputation. As Lord Denning said in Dingle at p 412:
“In order to show that a man has a bad reputation, you should call those who know him and have had dealings with him.” (Emphasis in original.)
The majority of the Court of Appeal in the present case also considered that the Dingle rule did not apply to the ICT conviction, on the basis of the decision in Goody v Odhams Press Ltd [1967] 1 QB 333 (“Goody”). In order to understand that case, it is necessary to mention first the decision of the Court of Appeal in Hollington v F Hewthorn & Co Ltd [1943] KB 587 (“Hollington”), where it was held that evidence of a previous conviction of careless driving was inadmissible in subsequent civil proceedings. The decision has attracted some criticism in relation to criminal convictions, reflecting the fact that a conviction by a United Kingdom court can be regarded as possessing some of the quality, within this country, of a judgment in rem, as Lord Hoffmann observed in Arthur J S Hall & Co v Simons (para 55 above). However, the decision is justifiable in principle on the basis that it is the duty of the court to form its own conclusion on the matter in question, on the basis of the evidence and submissions of the parties to the proceedings before it. The decision in Hollington remains authoritative under the common law: see, for example, Hui Chi-Ming v The Queen [1992] 1 AC 34, 43; Calyon v Michailaidis [2009] UKPC 34, paras 23–33, and the authorities cited there; and Hurnam v Bholah [2010] UKPC 12, para 41.
The position in relation to convictions by courts in the United Kingdom was amended by the 1968 Act. Under section 11, such a conviction is prima facie evidence that the person convicted committed the offence of which he was found guilty. Under section 13, the conviction is conclusive evidence in defamation proceedings. Those provisions do not apply in the present case since the ICT conviction was not a conviction by a court in the United Kingdom. No other statutory exception is in point. There is no dispute that the rule in Hollington therefore applies.
In Goody, the plaintiff had been convicted of taking part in the Great Train Robbery. After a newspaper published a story describing the part he had played in the robbery, he brought proceedings for libel.The newspaper did not seek to establish that he was guilty of the robbery, but pleaded in mitigation of damages that he was of bad reputation as a thief and robber given to violence. In that regard, they sought to lead evidence of his criminal record: he had been convicted six times in the previous six years of crimes of violence and dishonesty, and had spent most of that time in prison. Lord Denning MR accepted that where a defendant seeks to prove bad reputation in mitigation of damage, the law distinguishes between evidence of a plaintiff’s actual reputation and evidence of facts tending to prove that the plaintiff ought not to have a good reputation (Plato Films Ltd v Speidel). However, he considered that the evidence in question was admissible (pp 340–341):
“I think that previous convictions are admissible. They stand in a class by themselves. They are the raw material upon which bad reputation is built up. They have taken place in open court. They are matters of public knowledge. They are accepted by people generally as giving the best guide to his reputation and standing. … They are very different from previous instances of misconduct, for those have not been tried out or resulted in convictions or come before a court of law. To introduce those might lead to endless disputes. Whereas previous convictions are virtually indisputable. … what better guide can there be to his character and reputation than his previous convictions?”
Salmon LJ accepted that the language used by the majority of the House of Lords in Plato Films Ltd v Speidel could be interpreted as excluding evidence of previous convictions, but emphasised the specific context in which the question arose in the case before him (pp 343–344):
“… criminal trials are held in public and the convictions and sentences are all pronounced in public and given some notoriety. A man would be very fortunate indeed if he were convicted of a crime of violence or dishonesty without his conviction coming to the attention of those who knew him. For a man to be convicted six times in six years of such crimes without acquiring a bad reputation is to my mind impossible. Had those convictions not taken place the fact that the man was reputed to be a thief given to violence could only have been hearsay and surmise. The fact that he was so convicted is to my mind the best evidence of his bad reputation”.
Significantly, Salmon LJ made it clear that the position would not be the same in all cases where the plaintiff in defamation proceedings had previous convictions (p 344):
“It must not be thought that because of our decision in this case it follows that evidence of a conviction can always be given in mitigation of damages in an action for libel. Everything turns upon the nature and the date of the conviction. In the present case the convictions are all recent and highly relevant to the imputation in the article complained of. There may be cases, however, eg, where the convictions occurred long ago and are irrelevant, in which it would be manifestly unjust and impermissible in law to make any reference to them in mitigation of damages.”
There has been no argument on this appeal as to the correctness of the decision in Goody, or as to whether the ratio applies to convictions by courts outside the United Kingdom. This is therefore not the occasion to consider those questions. However, a number of material points of distinction from the circumstances of Goody are apparent in the present case.
First, since Goody was concerned with convictions by the English courts, it was possible for Lord Denning MR to say that they were “accepted by people generally as giving the best guide to [the plaintiff’s] reputation and standing” and were “virtually indisputable”, and to ask rhetorically, “what better guide can there be to his character and reputation?” This reasoning cannot apply to foreign convictions, at least without qualification, for the reasons set out by the Law Reform Committee in the report which led to the 1968 Act (para 48 above). The ordinary person is aware that criminal prosecutions and trials in foreign jurisdictions are not necessarily subject to the same requirements of independence, impartiality and procedural fairness as apply in the United Kingdom, and that capital sentences may be imposed which render submission to such a jurisdiction practically impossible. It follows that the reasoning in Goody cannot automatically be applied to all foreign convictions. If it can be applied, it has to be in a more nuanced manner.
In the present case, the claimant’s case, as set out in his pleadings and witness statements, is that the ICT conviction is not accepted by people generally as giving the best guide to his reputation and standing. Far from being virtually indisputable, the fairness of the conviction appears not to have been accepted by Interpol, and reputable organisations in this country have also been highly critical of the fairness of the ICT’s procedures (paras 17 and 21–22 above). Those averments are plainly material. If the conviction was the result of an unfair process, then no weight can be attached to it.
Secondly, this case is in any event distinguishable from Goody. In that case, the plaintiff’s previous criminal record was relevant to the assessment of damages if he succeeded on the question whether he had been defamed by the report of his involvement in the Great Train Robbery. Even if he was not proved to have been one of the great train robbers, the fact that he had a bad reputation as a thief and robber given to violence was material to the assessment of damages, and his previous record of convictions for such crimes was treated as admissible evidence of that reputation. In the present case, on the other hand, if the claimant succeeds in establishing that he was defamed by the Report, then it follows that the libel that he is guilty of war crimes and crimes against humanity has not been proved. In those circumstances, it would be perverse to treat the fact that he was convicted of those very crimes by the ICT as diminishing the damages to which he is entitled. Consistently with the logic of Dingle, if it is established that the claimant has been defamed by a report that he is guilty of war crimes and crimes against humanity, it cannot be relevant to prove in mitigation of damages that he has the reputation of being guilty of the very same war crimes and crimes against humanity. Lord Radcliffe’s explanation in Dingle (p 396) is precisely in point:
“A libel action is fundamentally an action to vindicate a man’s reputation on some point as to which he has been falsely defamed, and the damages awarded have to be regarded as the demonstrative mark of that vindication. If they could be whittled away by a defendant calling attention to the fact that other people had already been saying the same thing as he had said, and pleading that for this reason alone the plaintiff had the less reputation to lose, the libelled man would never get his full vindication.”
It follows that, in the present case, Goody cannot provide a proper basis for leading evidence of the ICT conviction in mitigation of damage.
The state of knowledge of readers of the Report
The majority of the Court of Appeal considered (para 57) that the people who read the relevant footnote in the Report “must have been individuals who had an unusually developed interest in ‘Challenging Hateful Extremism’ … and these are persons who are therefore overwhelmingly likely to have been aware of Mr Mueen-Uddin’s conviction by the [ICT] in Bangladesh and the contents of the Channel 4 Dispatches programme”.
There are two difficulties with this reasoning. First, it again relies on the publication of the same allegation against the claimant on earlier occasions, contrary to the rule in Dingle. Secondly, the conclusions arrived at by the majority of the Court of Appeal are conclusions on matters of fact. The court is not in a position to reach those conclusions at the present stage in the proceedings.
The particulars of claim state that the Report received widespread national media coverage and was shared by the Home Office Twitter account and the Twitter account of the Commission for Countering Extremism, reaching a possible 915,000 and 2,752 followers respectively. It is said that those tweets were themselves engaged with and shared by numerous other individuals, one of whom, the human rights campaigner Peter Tatchell, retweeted the Report to his 90,000 followers. It is also said that the claimant was contacted by numerous colleagues and associates who had read the allegations about him in the Report. In his witness statement, the claimant makes clear his concern that the fact that the allegations were published in a Home Office report, and on the government’s website, gave them greater authority than they had previously had: “I was deeply concerned that a great many people would believe that the allegations made against me were true, because the Home Office had said it was so”. The position adopted by the claimant in his pleadings and his witness statement cannot be ignored or dismissed out of hand.
It is also material to note that the earlier media coverage of the ICT conviction reported that the claimant denied the charges and maintained that the trial was politically motivated and unfair. It also reported that the fairness of the proceedings had been criticised by human rights organisations and other commentators (para 20 above). By contrast, the Report contained an unqualified assertion of the claimant’s guilt.
Against that background, it is impossible to assume that everyone who read the allegation made against the claimant was already aware of it; or that, even in the case of those who already knew of the allegation, its publication in an apparently authoritative official report had no impact on the claimant’s reputation. The extent of publication, and its effect on the claimant’s reputation, are matters to be determined at trial.
The vindicatory effect of a successful action
The majority of the Court of Appeal considered that these proceedings could not possibly result in the vindication of the claimant’s reputation, “because those who do not accept the result will point to the absence of relevant witnesses and evidence” (para 58). This point invites a number of responses. One is to ask, why does it matter if some people will not accept the result? It is possible in any litigation that there may be a group of people who will not accept the result, but that is not a principled reason for denying a party the right to bring his claim to trial.
A second response is to ask why the majority of the Court of Appeal focused on the reaction of those who would not accept the decision of the court, rather than those who would. The court normally proceeds on the basis that the outcome of its proceedings will be accepted—certainly, one might have thought, by the “right-thinking members of society generally” (Sim v Stretch [1936] 2 All ER 1237, 1240) with whose estimation of a person the law of defamation is concerned.
A third response is to note (again) that the Court of Appeal had no information before it (as Phillips LJ recorded at para 82(iii)) as to the availability of witnesses or other evidence: see para 69 above. A fourth response is to observe that there is, in any event, nothing unusual about cases in which the passage of time results in a shortage of witnesses and documentary evidence, but in which the decision of the court is nevertheless accepted: see paras 70–71 above.
The claimant has a legitimate interest in vindicating his reputation in this country, where he resides and of which he is a citizen, against an extremely serious allegation made by the government of this country. It is impossible to conclude, at this stage of the proceedings, that if he were to succeed in establishing that he had been defamed by the Secretary of State, a finding to that effect, and an appropriate award of damages, would be incapable of vindicating his reputation. On the contrary, it is reasonable to expect that success in these proceedings would be a major vindication.
Finally, in relation to the Court of Appeal’s balancing of the costs of the litigation against the value of the claim, the majority considered that the costs “are likely to be very substantial” (para 58). It is not obvious why that should be so if, as the majority also found, any trial will proceed “without much live evidence” (ibid), to the extent that “it will be impossible to obtain a fair hearing of the Secretary of State’s defence” (ibid) and the legitimacy of the court’s decision will be undermined by “the absence of relevant witnesses and evidence” (ibid). In reality, the court is in no position at this early stage of the proceedings to speculate as to the issues which might be in dispute, the evidence which might be led, or the likely costs of the proceedings. There is no pleaded defence, or even an outline indication of what evidence the Secretary of State might seek to rely on. As a result, the court cannot assess with any care, or any accuracy, the likely costs of the proceedings, or how the procedure might be managed so as to enable the claim to be adjudicated in a proportionate way.
A multi-factorial approach?
On the facts of a particular case, it is possible that more than one form of abuse of process may be relevant. It is also possible that some of the categories of abuse of process may overlap. The approach adopted by the majority of the Court of Appeal proceeded, however, on a different basis. In relation to Hunter abuse, they concluded that even if the claimant’s challenge to his conviction did not constitute such abuse, it could nevertheless contribute, along with other considerations, to the conclusion that the claim was an abuse of process. In relation to Jameel abuse, although their reasoning is less clear, they appear to have treated their finding that the value of the claim was greatly exceeded by the cost of the litigation as a factor which contributed to an overall conclusion that the claim was an abuse of process. The supposed unfairness resulting from the Secretary of State’s difficulty in establishing a defence of truth appears to have been treated as another factor in the overall assessment.
This approach was rightly described by Phillips LJ, at para 84, as unprincipled. The Hunter principle and the Jameel principle protect different aspects of the public interest, and have different rationales. Where neither principle is satisfied, the considerations which were relevant to each principle cannot simply be lumped together. If they are considered to be relevant to some other principle, that principle has to be identified and defined.
For the reasons explained earlier, it is not an abuse of process for the claimant to bring a claim which involves a collateral challenge to his conviction by the ICT: on the contrary, it is a proper use of civil procedure. That being so, the fact that the claim involves a collateral challenge to a prior conviction cannot count against the claimant in some wider assessment of whether the proceedings are an abuse of process. Similarly, if the claimant has suffered more than minimal damage to his reputation and so meets the test laid down in Jameel and Lachaux, then the fact (as the majority of the Court of Appeal considered it to be) that the value of the claim is less than the cost of the litigation cannot bear on an overall assessment that the claim is an abuse of process, in the absence, at least, of some other principle of abuse of process, to which that fact is relevant. No such principle has been identified in this case. Furthermore, the reasoning which led the majority of the Court of Appeal to conclude that the claim was of little value was erroneous, as explained at paras 90–115 above. Finally, for the reasons explained at paras 65–71 above, the Secretary of State’s difficulties in tracing witnesses and other evidence do not render the claim an abuse of process. Those difficulties simply reflect the fact that the Secretary of State has chosen to make an allegation against the claimant concerning events which occurred more than 50 years ago, and bears the burden of proof in establishing a defence of truth. It is perfectly proper for the claimant to bring these proceedings notwithstanding the Secretary of State’s difficulties in establishing a defence of truth. That being so, the Secretary of State’s difficulties cannot count against the claimant in some wider assessment of whether the proceedings are an abuse of process.
Conclusion
For all these reasons, I would allow Mr Mueen-Uddin’s appeal. The order striking out his claim should be set aside, and he should be permitted to pursue his claim at trial.