INDEPENDENT NEWS AND MEDIA AND INDEPENDENT NEWSPAPERS IRELAND LIMITED v. IRELAND

BAILII is celebrating 24 years of free online access to the law! Would you consider making a contribution?
No donation is too small. If every visitor before 31 December gives just £1, it will have a significant impact on BAILII's ability to continue providing free access to the law.
Thank you very much for your support!

	[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	European Court of Human Rights
You are here: BAILII >> Databases >> European Court of Human Rights >> INDEPENDENT NEWS AND MEDIA AND INDEPENDENT NEWSPAPERS IRELAND LIMITED v. IRELAND - 55120/00 [2005] ECHR 402 (16 June 2005) URL: http://www.bailii.org/eu/cases/ECHR/2005/402.html Cite as: (2006) 42 EHRR 46, 42 EHRR 46, [2005] ECHR 402

CASE OF INDEPENDENT NEWS AND MEDIA AND INDEPENDENT NEWSPAPERS IRELAND LIMITED v. IRELAND

This judgment will become final in the circumstances set out in Article 44 § 2 of the Convention. It may be subject to editorial revision.

In the case of Independent News and Media and Independent Newspapers Ireland Limited v. Ireland,

The European Court of Human Rights (Former Third Section), sitting as a Chamber composed of:

1. The case originated in an application (no. 55120/00) against Ireland lodged with the Court under Article 34 of the Convention for the Protection of Human Rights and Fundamental Freedoms (“the Convention”) by two Irish registered companies, Independent News and Media plc and Independent Newspapers (Ireland) Limited (“the applicants”), on 20 December 1999.

2. The applicants were represented by Ms P. Mullooly, a solicitor practising in Dublin. The Irish Government (“the Government”) were represented by their Agents, Ms D. McQuade and, subsequently, Ms P. O'Brien, both of the Department of Foreign Affairs.

3. The applicants complained that the domestic safeguards against disproportionately high jury awards in libel cases were inadequate.

4. The application was allocated to the Fourth Section of the Court (Rule 52 § 1 of the Rules of Court). Within that Section, the Chamber that would consider the case (Article 27 § 1 of the Convention) was constituted as provided in Rule 26 § 1.

5. On 1 November 2001 the Court changed the composition of its Sections (Rule 25 § 1). This case was assigned to the newly composed Third Section (Rule 52 § 1).

7. The Government, but not the applicants, filed observations on the merits (Rule 59 § 1). Comments were also received from seven third parties all of whom had been given leave by the President to intervene (Article 36 § 2 of the Convention and Rule 44 § 2). The applicants replied to the Government's comments (Rule 44 § 5), and the parties to the third parties' comments, at the oral hearing.

8. A hearing took place in public in the Human Rights Building, Strasbourg, on 16 October 2003 (Rule 59 § 3).

The Court heard addresses by Messrs McCullough S.C., O'Donnell S.C. and Murray S.C..

9. The applicants are Irish registered companies. The second applicant publishes newspapers including the Sunday Independent and is a wholly owned subsidiary of the first applicant (formerly known as Independent Newspapers plc).

10. The case concerns an article published in the Sunday Independent, a newspaper with the biggest circulation of any Sunday newspaper and which sold in the region of 250,000 copies at the relevant time.

11. On 13 December 1992 an article was published in the newspaper written by a well-known journalist and entitled “Throwing good money at jobs is dishonest”. The article commented, inter alia, on a recently discovered letter (dated September 1986) to the Central Committee of the Communist Party of the Soviet Union. The letter had been signed by two persons one of whom was Mr de Rossa, a very well-known politician. The letter referred to “special activities” that had previously met shortfalls in the funding of the Worker's Party, a political party of which Mr de Rossa had been leader. At the time of publication, Mr de Rossa was leader of another political party (the Democratic Left), he was a member of parliament and he was engaged in post-election negotiations about his party's participation in government.

“Irish society is divided. As the political parties manoeuvre to try to form a Government a clear picture has emerged, revealing the nature of our differences.

On one side of the argument are those who would find the idea of Democratic Left in cabinet acceptable. These people are prepared to ignore Democratic Left leader Proinsias de Rossa's reference to the 'special activities' which served to fund the Workers Party in the very recent past.

The 'special activities' concerned were criminal. Among the crimes committed were armed robberies and forgery of currency.

The people engaged in this business occupied that twilight world where the line blurs between those who are common criminals and others of that ilk who would claim to be engaged in political activity.

This world is inhabited by myriad groups, some dealing in drugs, prostitution, protection rackets, crimes of which the weakest members of society are invariably the victims.

It is therefore, ironic, wickedly so, that a political party claiming to 'care' for the workers should accept funding from 'special activities' of a particularly nasty kind.

There is no doubt that elements of Proinsias de Rossa 's Workers Party were involved in 'special activities'. What remains unproven is whether de Rossa knew about the source of his party's funds. There is evidence, strengthened by revelations in the Irish Times this week, that de Rossa was aware of what was going on.

If one is to allow him the benefit of the doubt, and why not, one must nevertheless have some misgivings about those with whom he so recently associated.

Justice demands that we welcome Democratic Left's recent conversion to decency and indeed, acknowledge that their Dáil deputies are exemplary in the conduct of the work they engage in on behalf of their constituents.

Still, questions remain unanswered about the Workers Party's 'special activities' phase, not to mention their willingness to embrace the Soviet Communist party long after the world knew about the brutal oppression that this and other Communist regimes visited on workers, intellectuals and others who would think and speak freely.

Proinsias de Rossa's political friends in the Soviet Union were no better than gangsters. The Communists ran labour camps. They were anti-Semitic.

Men like Andrei Sakharov and Vaclav Havel were persecuted. Citizens who attempted to flee this terror were murdered. In Berlin, the bodies left to rot in no man's land between tyranny and liberty. Is it really necessary to remind ourselves of those 'special activities'?”

13. In 1993 Mr de Rossa initiated a libel action (High Court) against the first applicant. The first trial lasted eight days: the jury was discharged (following the publication of an article by the first applicant). The second trial lasted fifteen days: the jury failed to reach a verdict.

“... damages are meant to compensate a person for a wrong. ... The only remedy available to a person who says he has been wronged in a newspaper is damages. Damages are meant to put a person, in so far as money can do it, in the position that he or she would have been if the wrong had not taken place. That is the enterprise you are engaged in, in relation to damages.”

16. He then referred to Mr Justice O'Flaherty's judgment in an unnamed case (which was, in fact, Dawson and Dawson v. Irish Brokers Association, Supreme Court judgment of 27 February 1997, unreported):

“ ... in a recent case, Mr. Justice O'Flaherty of the Supreme Court said, that the approach in cases of this kind should be no different from any other type of proceedings. The jury should be told that their first duty is to try to do essential justice between the parties. They are entitled to award damages for loss of reputation as well as for the hurt, anxiety, trouble and bother to which the Plaintiff has been put.”

17. He went on to quote with approval Mr Justice Henchy's judgment in another unnamed case (which was Barrett v. Independent Newspapers Ltd [1986] I.R. 13) as follows:

“It is the duty of the Judge to direct the Jury that the damages must be confined to such sum of money as would fairly and reasonably compensate the Plaintiff for his injured feelings, and for any diminution in his standing among right thinking people as a result of the words complained of. The Jury have to be told they must make their assessment entirely on the facts found by them, and among the relevant considerations proper to be taken into account are the nature of the libel, the standing of the Plaintiff, the extent of the publication, the conduct of the Defendant at all stages of the case, and any other matter which bears on the extent of the damages.”

“Now Mr. Justice Henchy, in the case he was dealing with, said that the jury in that particular case wasn't given any real help as to how to assess compensatory damages, and he laid down a guide which could assist the Jury. He considered that in the case in question the jury could be asked to reduce the allegation complained of to actuality, and then to fit the allegation into its appropriate place in the scale of defamatory remarks to which the Plaintiff could be subjected.

Now that particular case affords you great assistance in placing the nature of the defamation in a scale, because that case Mr. Justice Henchy was referring to, revolved around an allegation by a politician that a journalist [sic.] tweaked his beard. Now it related to the time of one of the pushes against Mr. Haughey, and after an abortive push against him, everybody was coming out to a crowded area of Leinster House, bustling out, and something was written in the Evening Herald which involved an allegation [that] a politician tweaked the Evening Herald journalist's beard. Now the learned Trial Judge found that to be defamatory and directed there be an assessment of damages.

Going back to Mr. Justice Henchy's observation, if you examine the words and put them in a scale of things, compare the allegation with tweaking a journalist's beard, with an allegation that Mr. de Rossa was involved in or tolerated serious crime, and that he personally supported anti-Semitism and violent Communist oppression. It would not surprise me, Members of the Jury, if you went to the opposite end of the scale and even, apart from Mr. Justice Henchy's helpful observations, I think there can be no question in this case but that if you are awarding damages you are talking about substantial damages.

Now as Counsel told you, I am not allowed to suggest to you figures, and Counsel are not allowed suggest to you figures either. I have gone as far as I can to help in relation to that question. I don't think anybody takes issue with the proposition if you are awarding damages they are going to be substantial. Mr. de Rossa at the time was leader of a political party. The political party was seeking to go into government. Damages will be substantial. It is all I can say to you. It is a matter for you to assess what they ought to be, if you are assessing damages.”

19. The jury found that the impugned words implied that Mr de Rossa had been involved in or tolerated serious crime and that he had personally supported anti-semitism and violent communist oppression. The jury went on to assess damages at 300,000 Irish pounds (IR£).

20. The first applicant appealed the award. It accepted that the jury had been directed on damages in accordance with the law but noted that the trial judge had been therefore obliged to confine his directions to a statement of general principles and to eschew any specific guidance on the appropriate level of general damages. Neither counsel nor the trial judge could suggest any figures to the jury and this practice was inconsistent with the provisions of the Constitution and of the Convention. Specific guidelines should be given to the jury in such cases including a reference to the purchasing power of any award made and to the income which the award would produce, to what the trial judge and counsel considered to be the appropriate level of damages and to awards made in personal injuries and other libel cases. The first applicant further argued that the common law and the Constitution required the appellate court to subject jury awards in defamation actions to stricter scrutiny so that the test which had been outlined by Mr Justice Henchy in the above-cited Barrett case was no longer sufficient. A court of appeal should ask itself the following question (the “Rantzen test”): “could a reasonable jury have thought that this award was necessary to compensate the plaintiff and to re-establish his reputation?”. The first applicant relied on, inter alia, Ranzen v. M.G.N. Ltd [1993] 4 All ER 975, and John v. M.G.N. Ltd [1996] 2 All ER 35) and on the judgment of this Court in the case of Tolstoy Miloslavsky v. the United Kingdom (judgment of 13 July 1995, Series A no. 323).

21. The Chief Justice delivered the majority judgment of the court on 30 July 1999. He began by describing the role of juries in the assessment of damages in defamation actions. It had been conceded by the first applicant that the trial judge had followed the practice in cases of this nature, namely:

“...that of confining his directions to a statement of general principles, eschewing any specific guidance on the appropriate level of general damages”.

As pointed out by the Master of the Rolls in the above-cited John v. M.G.N. case:

“Judges, as they were bound to do, confined themselves to broad directions of general principle, coupled with injunctions to the jury to be reasonable. But they gave no guidance on what might be thought reasonable or unreasonable, and it is not altogether surprising that juries lacked an instinctive sense of where to pitch their awards. They were in the position of sheep loosed on an unfenced common, with no shepherd.”

22. This was explained by the fact that the assessment of damages in libel cases was “peculiarly the province of the jury” As stated by Chief Justice Finlay in the Barrett case (cited above) the assessment by a jury of damages for defamation had a “very unusual and emphatic sanctity” so that the appellate courts had been extremely slow to interfere with such assessments. As emphasised in the above-cited John v. M.G.N. case, the ultimate decision, subject to appeal, was that of the jury which was not bound by the submissions made to it.

23. The Chief Justice outlined the relevant domestic law. He considered that there was no conflict between the common-law and the Constitutional provisions, on the one hand, and Article 10 of the Convention, on the other. Article 10, as noted in the Tolstoy Miloslavsky judgment, required that “an award of damages for defamation must bear a reasonable relationship of proportionality to the injury to reputation suffered”. He continued:

“By virtue of the provisions of Article 40.6.1^o of the Constitution, the defendant is entitled, subject to the restrictions therein contained, to exercise the right to express freely its convictions and opinions.

The exercise of such right is subject however to the provisions of the Constitution as a whole and in particular the provisions of Article 40.3.1^o and 40.3.2^o which require the State by its laws to protect as best it may from unjust attack, and in the case of injustice done to vindicate the good name of every citizen.

Neither the common law nor the Constitution nor the Convention give to any person the right to defame another person.

The law must consequently reflect a due balancing of the constitutional right to freedom of expression and the constitutional protection of every citizen's good name (Hynes-O'Sullivan. v. O'Driscoll [1988] I.R. 436). This introduces the concept of proportionality which is recognised in our constitutional jurisprudence.”

He cited, as the law applicable in the State, the judgment of Mr Justice Henchy in the above-cited Barrett case (see also paragraphs 44-46 below) and considered that a passage therein (the duty of the trial judge to direct the jury to confine damages to a sum as would “fairly and reasonably compensate the plaintiff for his injured feelings and for any diminution in his standing among right-thinking people”) emphasised the following elements of Irish law:

“(a) ... it is the duty of the judge to direct the jury that the damages must be confined to such sum of money as will fairly and reasonably compensate the plaintiff for his injured feelings and for any diminution of his standing among right-thinking people as a result of the words complained of;

(b) ... it is a fundamental principle of the law of compensatory damages that the award must always be reasonable and fair and bear a due correspondence with the injury suffered; and

24. The obligations arising from the provisions of the Constitution and the Convention were met by the laws of Ireland, which “provides that the award must always be reasonable and fair and bear a due correspondence with the injury suffered and by the requirement that if the award is disproportionately high, it will be set aside.”

25. Accordingly, and as regards directions to be given to juries, neither the Constitution nor the Convention required a change as suggested by the first applicant. The added guidelines recommended by the Court of Appeal in the case of John v. M.G.N. were not based on the Convention but were a development of English common law. Indeed, he regarded the changes brought about by the case of John v. M.G.N. as not “modest” but “fundamental” in that they “radically altered” the general practice with regard to the instructions to be given to a jury as to the manner in which they should approach the assessment of damages in a defamation action. If the approach adopted in the Rantzen case and developed in the John v. M.G.N. case was to be adopted in Ireland, the jury would be buried in figures from the parties representatives and from the judge in respect of both libel and personal injuries' damages previously awarded, while at the same time being told that they were not bound by such figures. He was satisfied that the giving of such figures, even in guideline form, would constitute an unjustifiable invasion of the domain of the jury. Awards in personal injury cases were not comparable with libel awards and thus he preferred the view on this particular matter expressed in the Rantzen case as opposed to the John v. M.G.N case. Informing juries of libel awards approved by the Court of Appeal would not have been recommended in the John v. M.G.N. case but for the Courts and Legal Services Act 1990 (a law which concerned the power of the Court of Appeal) in the United Kingdom.

26. On the contrary, the jury must base its assessment entirely on the facts of the case as established by it (Mr Justice Henchy in the Barrett case) and a departure from that principle would lead to utter confusion. Each defamation action had its own unique features and a jury assessing damages had to have regard to each feature. Those features, which could vary from case to case, included the nature of the libel, the standing of the plaintiff, the extent of publication, the conduct of the defendant at all stages and any other relevant matters. Figures awarded in other cases based on different facts were not matters which the jury should be entitled to take into account. The Chief Justice was not therefore prepared to change the traditional guidelines given to juries in the assessment of damages in libel cases.

27. He clarified that this did not mean that the discretion of the jury in libel cases was limitless:

“... the damages awarded by a jury must be fair and reasonable having regard to all the relevant circumstances and must not be disproportionate to the injury suffered by the injured party and the necessity to vindicate such party in the eyes of the public. Awards made by a jury are subject to a right of appeal and on the hearing of such appeal, the awards made by a jury are scrutinised to ensure that the award complies with these principles.”

28. The Chief Justice then turned specifically to appellate reviews of such jury awards. He began quoting with approval Chief Justice Finlay in the Barrett case: while the jury assessment was not sacrosanct in the sense that it could never be disturbed on appeal, it had a very “unusual and emphatic sanctity” in that the jurisprudence had clearly established that the appellate courts had been “extremely slow” to interfere with such assessments. He also quoted with approval from the Court of Appeal judgment in the John v. M.G.N. case (at p. 55): “real weight must be given to the possibility that [the jury's] judgment is to be preferred to that of a judge”.

“Both judgments recognise that the assessment of damages is a matter for the jury and that an appellate court must recognise and give real weight to the possibility that their judgment is to be preferred to that of a judge.

Consequently, an appellate court should only set aside such an award made by a jury in a defamation action if the award is one which no reasonable jury would have made in the circumstances of the case and is so unreasonable as to be disproportionate to the injury sustained.”

30. He rejected the argument that larger awards should be subjected to a more searching scrutiny than had been customary in the past. He did not agree that the Rantzen test proposed by the first applicant (“could a reasonable jury have thought that this award was necessary to compensate the plaintiff and to re-establish his reputation”) was the test to be applied, noting that that test “differs substantially from the test which has hitherto applied”. If the Rantzen test were to be applied it would remove the “very unusual and emphatic sanctity” from jury awards and would take away the giving of “real weight” to the possibility that the jurors' judgment is to be preferred to that of the judge. He concluded:

“Consequently, while awards made by a jury must, on appeal be subject to scrutiny by the appellate court, that Court is only entitled to set aside an award if it is satisfied that in all the circumstances, the award is so disproportionate to the injury suffered and wrong done that no reasonable jury would have made such an award.”

31. Applying that test, the Chief Justice considered whether the damages awarded were excessive and disproportionate to any damage done to Mr de Rossa. He recalled that the factors to be taken into account were well established and he quoted with approval those outlined in the John v. M.G.N. judgment (pp. 47-48).

32. As to the gravity of the libel, he noted that the libel clearly affected Mr de Rossa's personal integrity and professional reputation. It was hard to imagine a more serious libel given the nature of the allegations, the profession of Mr De Rossa and the ongoing negotiations concerning his participation in Government.

33. As to the effect on him, the Chief Justice referred to his evidence before the High Court as to the hurt and humiliation caused to him and his determination to vindicate his personal and professional reputation. This evidence was obviously accepted by the jury and it was easy to imagine the hurt and distress allegations of this nature would cause.

34. The extent of the publication was wide: it was conceded by the parties that the “Sunday Independent” had a wide circulation throughout the State and was read each Sunday by over one million persons.

35. The Chief Justice then considered the conduct of the first applicant up to the date of the verdict, including whether or not an apology, retraction or withdrawal had been published. The lack of an apology was regarded as being of considerable importance, a matter highlighted by Mr de Rossa's evidence during the second and third trials. The passages cited by the Chief Justice demonstrated clearly, in his view, that all Mr de Rossa required was a withdrawal of the allegations in the absence of which he was obliged to endure three trials to secure vindication of his reputation during which he was subjected to “immensely prolonged and hostile cross-examination” by Counsel for the first applicant and his motives for bringing the action were challenged as were Mr de Rossa's bona fides and credibility.

“The Respondent is entitled to recover, as general compensatory damages such sum as will compensate him for the wrong which he has suffered and that sum must compensate him for the damage to his reputation, vindicate his good name and take account of the distress, hurt and humiliation which the defamatory publication has caused. Such sum should, however, be fair and reasonable and not disproportionate to the wrong suffered by the Respondent.

The jury found that the words complained of by the Respondent meant that the Respondent was involved in or tolerated serious crime and personally supported anti-Semitism and violent Communist oppression.

If these allegations were true, the Respondent was guilty of conduct, which was not only likely to bring him into disrepute with right-minded people but was such as to render him unsuitable for public office.

No more serious allegations could be made against a politician such as the Respondent herein.

Having regard to the serious nature of the said libel, its potential effect on the career of the Respondent, and the other considerations as outlined herein, it would appear to me that the jury would have been justified in going to the top of the bracket and awarding as damages the largest sum that could fairly be regarded as compensation.”

The jury assessed damages in the sum of £300,000. This is a substantial sum but the libel was serious and grave involving an imputation that the Respondent was involved in or tolerated serious crime and that he personally supported anti-Semitism and violent Communist oppression.

Bearing in mind that a fundamental principle of the law of compensatory damages is that the award must always be reasonable and fair and bear a due correspondence with the injury suffered and not be disproportionate thereto, I am not satisfied that the award made by the jury in this case went beyond what a reasonable jury applying the law to all the relevant considerations could reasonably have awarded and is not disproportionate to the injury suffered by the Respondent.”

37. The award approved by the Supreme Court, IR£300,000, was three times more than the highest libel award previously approved by that court. The award and Mr de Rossa's legal costs were discharged by the second applicant as were the first applicant's own legal costs.

38. As to the guidelines to be give to jurors and having reviewed relevant judgments from certain common-law jurisdictions and in the above-cited Tolstoy Miloslavsky case, Mrs Justice Denham was in favour of giving further guidelines to jurors including in respect of prior libel awards made or affirmed by the Supreme Court, prior awards in personal injuries' cases, the purchasing power of an award and the income it might produce together with the level of award deemed appropriate. There was nothing in principle to prevent comparative figures being so provided: it would not diminish the place of the jury if it was informed of issues relevant to the proportionality of the damages. Indeed, as in the John v. M.G.N. judgment, she considered that such information would enhance the role of the jury since it would be assisted by comparative and other relevant information.

39. As to the required test to be applied by the appellate court, she recalled and quoted with approval the judgments of Chief Justice Finlay and of Mr Justice Henchy in the Barrett case. She saw no reason why, if the Chief Justice in that case was making a comparative assessment of awards, this information should not be available to the jury. She agreed that the appellate court should strive to determine the reasonableness and proportionality of awards as outlined in the Barrett case, but the effectiveness of that appellate review depended on the prior availability to the jury at first instance of adequate guidelines on damage levels. Such an approach, she believed, would enable the system to be more consistent and comparative and would allow it to appear more rational.

40. As to whether the award in the present case was excessive, she noted that there were strong similarities between the present case and the case of McDonagh v. News Group Newspaper Limited (Chief Justice Finlay, Supreme Court judgment of 23 November 1993, unreported): both plaintiffs had a standing in the community and the relevant publications were seriously defamatory. However, the award in the McDonagh case was considered to be at the top of the permissible range. Even allowing for the additional aggravating matters in the present case, it was clear that the award was “beyond that range in the sense that it is so incorrect in principle that it should be set aside”. She considered that the award to Mr de Rossa should be reduced to IR£150,000 and concluded:

“In principle it is open to the Court to provide guidelines on the charge to be given by a judge to a jury in libel cases. Guidelines on levels of damages given by a judge would aid the administration of justice. Guidelines would give relevant information and aid comparability and consistency in decision-making. Such guidelines would relate only to the level of damages - not the kernel issue as to whether or not there had been defamation. Thus, such guidelines would not impinge of the area traditionally viewed in common law jurisdictions as a matter quintessentially for the jury. More specific guidelines on the level of damages would help juries and the administration of justice by bringing about more consistent and comparable awards of damages and awards which would be seen as such. Specific guidelines would also inform an appellate court in its determination as to whether an award is reasonable and proportionate. The award in this case was excessive and on the principles of reasonableness and proportionality I would reduce it to £150,000.”

41. Article 40(3) of the Irish Constitution provides, in so far as relevant, as follows:

“1. The State guarantees in its laws to respect, and, as far as practicable, by its laws to defend and vindicate the personal rights of the citizen.

2. The State shall, in particular, by its laws protect as best it may from unjust attack and, in the case of injustice done, vindicate the life, person, good name, and property rights of every citizen.”

“The State guarantees liberty for the exercise, subject to public order and morality: –

i. The right of the citizens to express freely their convictions and opinions. The education of public opinion being, however, a matter of such grave import to the common good the State shall endeavour that organs of public opinion, such as the radio, the press, the cinema, while preserving their liberty of expression, including criticism of Government policy, shall not be used to undermine public order or morality or the authority of the State.”

43. The jury assess damages following its finding of defamation. The Supreme Court can review and quash the award of a jury of the High Court. It does not substitute its own award but rather refers the matter back to the High Court for a further trial on damages before a different jury. The second jury will not be informed that an earlier award was quashed nor, consequently, of the decision or reasoning of the Supreme Court.

44. The case concerned a defamatory allegation that a politician had pulled a journalist's beard when leaving parliament. The jury award (IR£65,000) was set aside by the Supreme Court. The Chief Justice considered the following principles to apply to the award (at p. 19):

“Firstly, whilst the assessment by a jury of damages for defamation is not sacrosanct in the sense that it can never be disturbed upon appeal, it certainly has a very unusual and emphatic sanctity in that the decisions clearly establish that appellate courts have been extremely slow to interfere with such assessments, either on the basis of excess or inadequacy. Secondly, it is clear that whilst the damages in this case at least, where no question of punitive or exemplary damages arises, are fundamentally compensatory in form, that the plaintiff is entitled not only to be compensated for the damage to his reputation arising from the publication of the defamation, but also for the hurt, anxiety and distress to him arising by its publication and by the subsequent conduct of the defendant right up to the time of the assessment of the damages.”

45. He also maintained that certain factors which the jury were entitled to take into account (including the standing of the plaintiff, the nature of the allegation, the failure by the newspaper to publish the plaintiff's denial and its maintenance of the allegation until the verdict) would have justified the jury in going to the top of the bracket and awarding the largest sum that could fairly be awarded as compensation. He continued (at p. 20):

“Notwithstanding these views, and notwithstanding the fact that this is clearly a case in which a jury would be entitled to award really substantial damages ... the sum of £65,000 awarded by the jury is so far in excess of any reasonable compensation for the allegation which was made, that it should be set aside.”

“The second ground of appeal is that the award of £65,000 is so excessive as to be unsustainable. In a case such as this, ... it is the duty of the judge to direct the jury that the damages must be confined to such sum of money as will fairly and reasonably compensate the plaintiff for his injured feelings and for any diminution in his standing among right-thinking people as a result of the words complained of. The jury have to be told that they must make their assessment entirely on the facts found by them, and they must be given such directions on the law as will enable them to reach a proper assessment on the basis of those facts. Among the relevant considerations proper to be taken into account are the nature of the libel, the standing of the Plaintiff the extent of the publication, the conduct of the Defendant at all stages of the case and any other matter which bears on the extent of damages. ...

The fact remains, however, that the jury were not given any real help as to how to assess compensatory damages in this case. A helpful guide for a jury in a case such as this would have been to ask them to reduce to actuality the allegation complained of, namely, that in an excess of triumphalism at his leader's success the plaintiff attempted to tweak the beard of an unfriendly journalist. The jury might then have been asked to fit that allegation into its appropriate place in the scale of defamatory remarks to which the plaintiff might have been subjected. Had they approached the matter in this way, ... the allegation actually complained of would have come fairly low in the scale of damaging accusations. The sum awarded, however, is so high as to convince me that the jury erred in their approach. To put it another way, if £65,000 were to be held to be appropriate damages for an accusation of a minor unpremeditated assault in a moment of exaltation, the damages proper for an accusation of some heinous and premeditated criminal conduct would be astronomically high. Yet a fundamental principle of the law of compensatory damages is that the award must always be reasonable and fair and bear a due correspondence with the injury suffered. In my view, the sum awarded in this case went far beyond what a reasonable jury applying the law to all the relevant considerations could reasonably have awarded. It was so disproportionately high that in my view it should not be allowed to stand.”

2. McDonagh v. News Group Newspapers Limited (Supreme Court judgment of 23 November 1993, unreported)

47. The impugned words were found by the jury to mean that the plaintiff barrister was, inter alia, a sympathiser with terrorist causes and incapable of performing his duties objectively. The jury award IR£90,000: it was not set aside on appeal. The Chief Justice noted:

“... I am satisfied that there are not very many general classifications of defamatory accusation which at present in Ireland, in the minds of right-minded people, would be considered significantly more serious. To an extent the seriousness may be somewhat aggravated by the fact that it is an accusation which has been made against a person who has a role, by reason of his profession and by reason of his standing as a member of the bar, in the administration of Justice.”

“The combined accusations made against the Plaintiff are that he failed or was likely to fail completely to do that, and that instead as a piece of major professional misconduct he abused the function which had been entrusted to him by his client.”

“A statement which makes that accusation and in addition makes the accusation of sympathy with terrorist causes would be extraordinarily damaging to any person, irrespective of their calling or profession. I, as I have indicated, take the view that the assessment of damages made by this jury, though undoubtedly high and at the top end of the permissible range, is not beyond that range in the sense that it is so incorrect in principle that having regard to the general approach of an appellate court to damages assessed by a jury for defamation it should be set aside. I would, therefore, dismiss the appeal.”

3. Dawson and Dawson v. Irish Brokers Association (Supreme Court judgment of 27 February 1997, unreported)

50. The plaintiff brothers were insurance brokers and took a libel action against the Irish Brokers Association about a letter in which the latter informed various industry bodies including the relevant Minister that the plaintiffs' company's membership of the Association had been terminated for non-compliance with the requirements of insurance legislation. Having found the letter defamatory, the jury awarded IR£515,000.

“... I have reached the clear conclusion that the award is so excessive as to call for the intervention of this Court. It is wholly disproportionate to any injury suffered by the plaintiffs ...

The approach to the assessment of damages in a [defamation] action is in essence no different from any other type of proceeding. The jury should, in the first instance, be told that their first duty is to try to do essential justice between the parties. [In cases where damages could be compensatory only, the jury] were entitled to award damages for loss of reputation, as well as for the hurt, anxiety, trouble and bother to which the plaintiffs had been put. However, the defendants in defamation cases should never be regarded as the custodians of bottomless wells which are incapable of ever running dry. ... Further, unjustifiably large awards, as well as the costs attendant on long trials, deals a blow to the freedom of expression entitlement that is enshrined in the Constitution.”

52. Quoting with approval the judgment of Mr Justice Henchy in the above-cited Barrett case and noting the evidence of harm to the plaintiffs' reputation and of the defendant's conduct, Mr Justice O'Flaherty continued:

“Giving the case the most favourable construction in regard to the plaintiffs – in the sense of asking one's self what damages have the plaintiffs made out in regard to loss of reputation etc., and taking their case at the high water mark – nonetheless, the award viewed even from that perspective must be regarded as so excessive that it cannot stand.”

53. The Supreme Court ordered a re-trial. At the end of the fourth trial in the High Court, a jury awarded IR£135,000.

4. O'Brien v. M.G.N. Ltd (Supreme Court judgment of 25 October 2000, unreported)

54. Mr O'Brien was a well-known and successful businessman. The jury found defamatory M.G.N. Ltd's allegations that he had, inter alia, bribed politicians to secure radio licences and been involved in other corrupt practices. The jury awarded IR£250,000 in damages. M.G.N. Ltd requested the Supreme Court to re-consider its judgment in the de Rossa appeal arguing, inter alia, that the latter judgment was wrong in so far as it considered that the principles laid down in the Barrett case were consistent with Article 10 of the Convention and with the Constitution.

55. The Chief Justice delivered the majority judgment of the court (joined by Mr Justice Murphy and Mr Justice O'Higgins), refusing to reconsider its de Rossa judgment but setting aside the jury award. Its previous judgment would not be reconsidered as it was not so “clearly wrong” that there were “compelling reasons” why it should be overruled. The O'Brien appeal had to be dealt with therefore on the basis of the principles outlined by the Supreme Court in the de Rossa and Barrett cases.

56. The general principle which the Chief Justice considered he must apply to his review of the award was that outlined by Mr Justice Henchy in the Barrett case, namely:

“Yet a fundamental principle of the law of compensatory damages is that the award must always be reasonable and fair and bear a due correspondence with the injury suffered. In my view, the sum awarded in this case went far beyond what a reasonable jury applying the law to all the relevant considerations could reasonably have awarded. It was so disproportionately high that in my view it should not be allowed to stand.”

57. In determining proportionality, he considered that there was nothing which precluded the Supreme Court from determining an appeal on jury libel awards in the light of other such awards which had also been approved by that court provided a degree of caution was exercised.

58. The Chief Justice considered the allegations against Mr O'Brien to be “undoubtedly seriously defamatory statements which justified the award of substantial damages”. Although he considered the damages' award to be in the “highest bracket of damages appropriate to any libel case” and that it was comparable to the non-pecuniary award “in the most serious cases of paraplegic or quadriplegic injuries”, he considered the libel as serious but not coming within the category of the grossest and the most serious libels to have come before the courts. He went on to compare that case to the de Rossa and McDonagh cases, although he acknowledged that:

“... ultimately ... this case has to be decided having regard to its own particular facts and circumstances. I am conscious of the care which must be exercised by an appellate court before it interferes with the assessment of damages by a jury in a case of defamation, but, having weighed up all the factors to which I have referred, I am satisfied that the award in this case was disproportionately high and should be set aside.”

59. Mr Justice Geoghegan in his partly dissenting opinion agreed with the Supreme Court's judgment in the de Rossa case but did not consider that the jury award had to be set aside.

60. He noted that various formulations of words had been used by appellate courts in Ireland and England as to when an appellate court in a libel action could interfere with a jury award. Although the language was sharper and stronger in some cases than in others, he was not sure that there was ever any intended difference and he was inclined to think that the form of words adopted by Mr Justice Henchy in the Barrett case (and already cited by the Chief Justice in that case – see above) was the most helpful. Having noted Chief Justice Finlay's comment also in the Barrett case about the assessment of the jury having “a very unusual and emphatic sanctity”, he indicated that he doubted whether Mr Justice Henchy and Chief Justice Finlay intended to say anything different:

“The true principle would seem to be that in all cases of compensatory damages whether in libel or in personal injuries or otherwise an appeal court will not interfere because its own judges thought the award too high. The court will only interfere if the award is so high that it is above any figure which a reasonable jury might have thought fit to award. But although that principle is the same in all cases of compensatory damages, the application of the principle will necessarily be different in the case of libel from the case of personal injuries. In the case of personal injuries an appeal court can determine with some confidence what would be the range of awards which a reasonable jury ... might make. ... In the case of a libel appeal however the appeal Court although it has to engage in the same exercise, it can only do so with diffidence rather than confidence. ... Unlike personal injury cases every libel action is completely different from every other libel action and therefore the guidelines available to an appeal court in settling the reasonable parameters of an award are much more limited.”

“having regard to the diffidence with which an appeal court should approach the possible setting aside of a jury award in a libel action, I could not have formed the view that the jury award was beyond reason.”

62. He went on to explain why comparisons with other libel awards approved by the Supreme Court were dangerous but that, even if he had to so compare, his view that the award should not be set aside was not affected by the facts or award in the de Rossa or McDonagh cases.

63. Mrs Justice Denham also dissented: she considered that there were compelling reasons to reconsider the Supreme Court's majority judgment in de Rossa. However, given the view of the majority that it would not depart from the de Rossa judgment, she applied it, compared that case and the McDonagh awards approved by the Supreme Court and found:

“Even allowing for the circumstances of the case, it is an award which in my view is beyond the range in that it is so incorrect in principle, it is so disproportionate, that it should be set aside.”

5. Hill v. the Cork Examiner Publications Limited (Supreme Court judgment of 14 November 2001, unreported).

64. Mr Hill was in prison having pleaded guilty to a charge of assault occasioning actual bodily harm (to a police officer). The defendant published an article which was entitled “Isolation of Cork Jail's C Wing” and which explained that C Wing prisoners were child molesters and sexual offenders and it included a photograph of Mr Hill's prison cell. He issued proceedings in December 1995 arguing that the juxtaposition of the article and the photograph meant and were understood to mean that he was a child molester or a sexual offender. During the trial, the foreman of the jury asked for guidelines. While the trial judge explained that he could not do so, he gave certain parameters (including the circumstances in which the photograph came to be taken, that large damages were not merited and that he was not entitled to damages as if he had a blameless character). The jury agreed that the article was defamatory and awarded Mr Hill IR£60,000.

65. The newspaper appealed arguing that the award was disproportionate and taking issue with the absence of guidelines to the jury. The Supreme Court did not set aside the award, Mr Justice Murphy noting:

“... it is difficult, if not impossible, to find any nexus between the pain, embarrassment or disfigurement suffered by a plaintiff and the sum of money which would be appropriate to compensate him for any such consequences of a wrong doing. Judges in charging juries as to their responsibilities in determining damages or in performing the same task themselves can say or do little more than recall that damages are designed to compensate for the consequences of a wrong doing and not to punish the wrong doer. It will always be said - perhaps unhelpfully – that the sum awarded should be reasonable to the plaintiff and also reasonable to the defendant. In relation to the extent to which a trial judge could and should give guidance as to an appropriate measure of damages was considered by [the Supreme Court in the De Rossa case] and again in O'Brien .v. M.G.N.... . Whilst other jurisdictions have accepted the concept of such guidelines that concept has been rejected in this jurisdiction. Apart from any other consideration there would appear to be insuperable difficulties for any judge to assemble the appropriate body of information on which to base such guidelines.”

“There is no doubt that the sum of £60,000 awarded by the jury was a substantial sum. It may well be at the higher, or even the highest, of the figures in the range which would be appropriate to compensate a Plaintiff for the wrong doing which he has suffered. However I am not satisfied that the figure awarded is so disproportionate to the injury sustained by the Plaintiff (Respondent) that it can or should be set aside by this Court.”

67. By judgment of 23 July 1996 (Heaney and McGuinness v. Ireland) the Supreme Court rejected the applicants' appeal finding section 52 of the Offences Against the State Act 1939 not inconsistent with the Constitution. It considered that the right to silence was a corollary to freedom of expression (guaranteed by Article 40 of the Constitution) and that the relevant assessment was to consider the proportionality of the restriction on the right to silence against the public order exception to Article 40. It noted that the 1939 Act was aimed at actions and conduct calculated to undermine public order and the authority of the State and that the proclamation made under Article 35 of the 1939 Act (that “the ordinary courts are inadequate to secure the effective administration of justice and the preservation of public peace and order”) remained in force.

68. As to whether section 52 restricted the right to silence more than was necessary in light of the disorder against which the State was attempting to protect the public, the court noted that an innocent person had nothing to fear from giving an account of his or her movements even though such a person may wish, nevertheless, to take a stand on grounds of principle and to assert his or her constitutional rights. However, it considered that the entitlement of citizens to take such a stand must yield to the right of the State to protect itself. The entitlement of those with something relevant to disclose concerning the commission of a crime to remain silent must be regarded as of an even lesser order. That court concluded that the restriction in section 52 was proportionate to the State's entitlement to protect itself.

69. The case of Murphy v. the Independent Radio and Television Commission ([1999] 1 IR 12) concerned the ban on the broadcasting of religious advertising pursuant to Section 10(3) of the Radio and Television Act 1988 (“the 1988 Act”). The Supreme Court considered that the impugned provision of the 1988 Act was a restriction of the appellant's right freely to communicate and of his right to freedom of expression (Articles 40(3) and 40(6)(1) of the Constitution, respectively) which rights could be limited in the interests of the common good. The real question was whether the limitation imposed upon those constitutional rights was proportionate to the purpose parliament wished to achieve. Quoting with approval previous case-law, it described the principle of proportionality:

“In considering whether a restriction on the exercise of rights is permitted by the Constitution the courts in this country and elsewhere have found it helpful to apply the test of proportionality, a test which contains the notions of minimal restraints on the exercise of protected rights and the exigencies of the common good in a democratic society. This is a test frequently adopted by the European Court of Human Rights and by the Supreme Court of Canada in the following terms. The objective of the impugned provision must be of sufficient importance to warrant over-riding a constitutionally protected right. It must relate to concerns pressing and substantial in a free and democratic society. The means chosen must pass a proportionality test. They must (a) be rationally connected to the objective and not be arbitrary, unfair or based on irrational considerations; (b) impair the right as little as possible; and (c) be such that the effects on the rights are proportional to the objective.”

70. The Supreme Court found that section 10(3) of the 1988 Act complied with this test and concluded that:

“It therefore appears to the court that the ban on religious advertising contained in section 10(3) of the 1988 Act is rationally connected to the objective of the legislation and is not arbitrary or unfair or based on irrational considerations. It does appear to impair the various constitutional rights referred to as little as possible and it does appear that its effects on those rights are proportional to the objective of the legislation.”

71. The LRC consultation paper of March 1991 considered a number of possible reforms of the law of defamation in Ireland and provisionally recommended, inter alia, that parties to defamation actions in the High Court should continue to have the right to have the issues of fact determined by a jury with the damages in such actions being assessed by the Judge following the jury's determination whether nominal, compensatory or punitive damages should be awarded.

72. The LAG was established by the Minister for Justice, Equality and Law Reform with a view to examining reforms of the libel laws to bring them into line with other States. As regards the respective roles of the judge and jury, its report of March 2003 provided as follows:

“The initial starting point for the Group's consideration of this matter was the specific recommendation of the Law Reform Commission that the parties to defamation actions should continue to have the right to have issues of fact determined by a jury but that the damages in such actions should be assessed by a judge. ... The Group was also alert to the valuable role which juries have to play in defamation actions given the importance, in such actions, of getting the perspective of the ordinary persons as to whether the matter complained of should, or should not, be considered defamatory. At the same time, the Group recognised that there is considerable dissatisfaction with the law as it currently stands whereby juries are deprived of guidance when it comes to deciding upon the level of damages which should be awarded to a successful plaintiff in a defamation action.

The Group was very much of the view that the division of function as between judge (assessment of damages) and jury (assessment of liability) would not operate well in practice. Indeed, the view was taken that such a division would place judges in a difficult position since they would not be privy to the seriousness with which the jury viewed the defamatory matter. Accordingly, the Group concluded that juries should continue to have a role in assessing damages in the High Court. However, this role should not be unfettered. Rather, it was agreed that the parties to the proceedings should be able to make submissions to the court and address the jury concerning damages. Furthermore, a statutory provision should be introduced which would require the judge in High Court proceedings to give directions to the jury on this matter. Such a provision should be general in nature but would, in an appropriate case, allow a judge to refer to the purchasing power of the likely award, the income which it might produce, the scale of awards in previous defamation cases and the appropriate level of damages in all the circumstances of the case. These provisions should be in addition to the basic provision which would specify a broad range of factors to which regard should be had when making an award of [non-pecuniary] damages. It was felt that provisions of this kind would be consistent with recent developments within the United Kingdom and other common law jurisdictions .... and would accord well with the freedom of expression entitlement enshrined in both the Constitution and the European Convention on Human Rights.

The final element considered by the Group under this heading concerned the desirability of having a statutory provision which would make it clear that, in a defamation appeal from the High Court, the Supreme Court could substitute its own assessment of damages for the damages awarded in the High Court. The Group is of the view that there is considerable merit in a provision of this kind given the additional costs which litigants would have to bear should a new trial be ordered and where the only issues for the appellate court to determine is the appropriateness of the damages award.

• The function of assessing damages in defamation proceedings heard before a jury should remain with the jury;

• Parties to proceedings should be able to make submissions to the court and address the jury concerning damages; Judges would be required to give directions to a jury on the matter of damages;

• In making an award of damages, regard would have to be had to a non-exhaustive list of matters including, for example, the nature and gravity of any allegation in the defamatory matter, the extent to which the defamatory matter was circulated and the fact that the defendant made or offered an adequate, sufficient and timely apology, correction or retraction, as the case might be. ...

• There should be an avoidance of doubt provision to the effect that, in a defamation appeal from the High Court, the Supreme Court could substitute its own assessment of damages for the damages awarded in the High Court.”

73. The Court of Appeal observed that the grant of an almost limitless discretion to a jury failed to provide a satisfactory measurement for deciding what was “necessary in a democratic society” or “justified by a pressing social need” for the purposes of Article 10 of the Convention. It continued:

“... the common law if properly understood requires the courts to subject large awards of damages to a more searching scrutiny than had been customary in the past. It follows that what had been regarded as the barrier against intervention should be lowered. The question becomes: could a reasonable jury have thought that this award was necessary to compensate the plaintiff and to re-establish his reputation?”

74. As to what guidance the judge could give to the jury, the Court of Appeal was not persuaded that the time had come to make references to awards by juries in previous libel cases. Nor was there any satisfactory way in which awards made in actions involving serious personal injuries could be taken into account. It was to be hoped that in the course of time a series of decisions of the Court of Appeal, taken under section 8 of the Courts and Legal Services Act 1990, would establish some standards as to what would be “proper” awards. In the meantime the jury should be invited to consider the purchasing power of any award which they may make and to ensure that any award they make is proportionate to the damage which the plaintiff has suffered and is a sum which it is necessary to award him to provide adequate compensation and to re-establish his reputation.

75. The Court of Appeal concluded in that case that, although a very substantial award was clearly justified in the case, judged by any objective standards of reasonable compensation or necessity or proportionality, an award of 250,000 pounds sterling (GBP) was excessive and it substituted GBP 110,000.

76. The Court of Appeal held that in assessing compensatory damages in a defamation case a jury could in future properly be referred by way of comparison to the conventional compensation scales in personal injury cases and to previous libel awards made or approved by the Court of Appeal. As the Master of the Rolls pointed out:

77. While the ultimate decision (subject to appeal) was that of the jury which was not bound by submissions made to them, there was no reason why the judge or counsel should not indicate to the jury the level of award which they considered appropriate:

“The plaintiff will not wish the jury to think that his main object is to make money rather than clear his name. The defendant will not wish to add insult to injury by underrating the seriousness of the libel. So we think the figures suggested by responsible counsel are likely to reflect the upper and lower bounds of a realistic bracket. The jury must, of course, make up their own mind and must be directed to do so. They will not be bound by the submission of counsel or the indication of the judge. If the jury make an award outside the upper or lower bounds of any bracket indicated and such award is the subject of appeal, real weight must be given to the possibility that their judgment is to be preferred to that of the judge.

The modest but important changes of practice described above would not in our view undermine the enduring constitutional position of the libel jury. Historically, the significance of the libel jury has lain not in their role of assessing damages, but in their role of deciding whether the publication complained of is a libel or not. The changes which we favour will, in our opinion, buttress the constitutional role of the libel jury by rendering their proceedings more rational and so more acceptable to public opinion. ...

The [Convention] is not a free standing source of law in the United Kingdom. But there is, as already pointed out, no conflict or discrepancy between Art. 10 and the common law. We regard Art. 10 as reinforcing and buttressing the conclusions we have reached and set out above. We reach those conclusions independently of the [Convention], however, and would reach them even if the convention did not exist.”

78. As to the factors of which one should take account in assessing the damages to be awarded, the Court of Appeal found:

“The successful plaintiff in a defamation action is entitled to recover, as general compensatory damages, such sum as will compensate him for the wrong he has suffered. That sum must compensate him for the damage to his reputation, vindicate his good name and take account of the distress, hurt and humiliation which the defamatory publication has caused. In assessing the appropriate damages for injury to reputation, the most important factor is the gravity of the libel ... The extent of publication is also very relevant ... It is well established that compensatory damages may and should compensate for additional injury caused to the plaintiff's feelings by the defendant's conduct of the action as when he persists in an unfounded assertion that the publication was true, or refuses to apologise, or cross-examines the plaintiff in a wounding or insulting way.”

80. The NNI is the representative body for Irish national newspapers including a number of newspapers owned by the applicants. It considered, inter alia, that the decision of the Supreme Court in the present case did not accord with the above-cited Tolstoy Miloslavsky judgment. The NNI endorsed the recommendations of the LRC and of the LAG (paragraphs 71-72 above) about the parties and the trial judge addressing the jury directly on the level of damages. More generally, it maintained that many other aspects of defamation law were in urgent need of reform so that that the freedom of speech of journalists in Ireland was unreasonably inhibited.

81. This company is part of a larger media group known as Associated Newspapers Limited based in the United Kingdom and it publishes an Irish national Sunday newspaper. It submitted, inter alia, that various aspects of Irish defamation law acted as a chilling effect on the press' freedom of expression including the Supreme Court's inability to substitute its own award together with the associated inability to inform the jury on a re-trial of the Supreme Court's views and the connected costs impact of an appeal.

82. The Irish Times Limited is the owner and publisher of the “Irish Times” newspaper one of Ireland's leading daily newspapers which is also distributed in the United Kingdom and in Europe. It has defended many defamation actions, was particularly concerned about the restrictions on instructing a jury on damages and it endorsed the work and recommendations of the LAG.

D. Thomas Crosbie Holdings Limited (“TCH”) and Examiner Publications (Cork) Limited (“EPC”)

83. The subsidiaries of these holding companies publish, print and distribute national and regional newspapers in Ireland and the United Kingdom. EPC was itself subjected to effectively the same treatment as the present applicants (the above-cited Hill case). The failure to implement the proposals of the LAC and LAG was prejudicial to the Irish media.

84. MGN Ltd publishes many Irish daily and weekly newspapers. As a former defendant in libel proceedings in Ireland (O'Brien v. M.G.N. Ltd case, see paragraphs 54-63 above), it regretted that the Supreme Court did not substitute its own award for that of the jury: sending a case back for re-trial was costly and, because the second jury was not informed of the appeal court's view, the risk of disproportionality remained.

85. These companies publish numerous weekly and daily papers in Ireland and in the United Kingdom. They underlined their support for this Court's judgment in the above-cited case of Tolstoy Miloslavsky and for the Court of Appeal in the above-cited Ranzen and John v. M.G.N. cases.

86. The NUJ is the largest union of journalists in the world and its Irish branch represents (97% (about 3000) of Irish journalists). It considered that Irish libel laws prevented journalists from carrying out their duties and denied access to fair and efficient proceedings to protect one's reputation. As to the lack of guidance to juries, it considered that the size and arbitrary nature of jury awards were powerful chilling factors on the press.

87. The applicants complained that, given the exceptional damages' award and the absence of adequate safeguards against disproportionate awards, their rights under Article 10 were violated. They considered their case indistinguishable from the above-cited Tolstoy Miloslavsky case.

“1. Everyone has the right to freedom of expression. This right shall include freedom to hold opinions and to receive and impart information and ideas without interference by public authority and regardless of frontiers. ...

2. The exercise of these freedoms, since it carries with it duties and responsibilities, may be subject to such formalities, conditions, restrictions or penalties as are prescribed by law and are necessary in a democratic society, in the interests of ... for the protection of the reputation or rights of others ...”

89. The Government objected to the applicants' overall approach. A balance had to be struck between protecting expression and reputations so that, once there was a finding of defamation, the weight of Convention support shifted to the protection of reputation. This latter right, guaranteed by Article 8, had been infringed to a devastating extent in the present case. The only remaining Article 10 issue was to ensure that the damages' award was proportionate to the harm done to that reputation, bearing in mind any chilling effect on further similar publications. The applicants' approach, on the other hand, reduced the Convention issues and the Tolstoy Miloslavsky judgment to simplistic mathematical formulae as if the only right at issue was freedom of expression without regard for the underlying values and contextual complexities of the matter including the power of the media, the devastating effects of defamatory allegations on reputations, the consequent destruction of the “human potential” which Article 10 supports and the respective roles of the domestic and European courts.

The Government considered “indirect and remote” any possibility of a chilling effect on political commentary by the press by the present or other damages awards. No such causal link had been demonstrated in the present case and, in any event, awards in libel cases were inherently and unavoidably uncertain. The Government further criticised the applicants and third parties' comments on numerous aspects of Irish libel law not relevant to the present case.

90. The applicants maintained that the simple fact was that their case was not distinguishable from that of Tolstoy Miloslavsky. The Government was wrong to suggest that the applicants might have lost their Article 10 rights following a finding of defamation: the Court was just as concerned about the chilling effect on the press (particularly on its political expression) of excessive penalties, an effect the applicants considered significant in the present case.

91. The Government considered that the present damages' award could not violate Article 10 of the Convention.

92. Their primary submission was that it must be shown that the award was disproportionate to the harm to reputation before the Court had to examine the domestic safeguards against disproportionate awards. This required an assessment of injury against certain criteria (the gravity of the libel, Mr de Rossa's position as an elected politician, the timing of the libel, the circulation of the libel, the conduct of the first applicant and the personal impact on Mr de Rossa). Since the award was clearly proportionate to the devastating harm inflicted, it was not necessary to examine the safeguards.

93. In any event, the applicants' comparative approach was flawed. It did not compare like with like (it did not make sense to compare awards in the same jurisdiction and they should have referred to previous jury awards and not those approved by the Supreme Court). It was arbitrary (a large award delivered between the judgments of the Supreme Court and of this Court would have undermined the comparative approach). It was perilous since it was asking this Court to fix the cap on damages' awards in Irish libel cases and to do so lower than in other jurisdictions. It also amounted to second-guessing a domestic appeal court's finding of proportionality despite the margin of appreciation accorded to the national authorities in making such assessments and the subsidiary nature of the Convention system.

94. Even applying the applicants' defective test, the present award was not exceptionally high. It was one sixth of the award in the Tolstoy Miloslavsky case and the highest prior libel award in Ireland was not IR£90,000 (the above-cited McDonagh case) - two previous jury awards were higher (IR£275,000 in Denny v. Sunday News, High Court, Irish Times of 14 November 1992, unreported, and IR£515,000 in the above-cited Dawson case), although the Government did accept that no defence or appeal had been filed in the Denny case and that the Supreme Court in the Dawson case had set aside the second award.

95. For these reasons, the Government considered that it was not necessary to examine the domestic legal safeguards against disproportionate awards.

96. The applicants clarified that they were arguing that the award was of such significance that one could not conclude as to its proportionality without first examining the domestic safeguards against disproportionate awards. The Government's primary argument, that safeguards were examined after the award had been found to be disproportionate, misinterpreted the Court's approach in Tolstoy Miloslavsky. If the Court in that case found the award to be disproportionate at the outset, there would have been no need to go further and examine the domestic safeguards. In fact the jury award in Tolstoy Miloslavsky was sufficiently significant as to trigger a review of the adequacy of the safeguards against disproportionate awards. The more exceptional the award, the more scrutiny of safeguards required.

97. As to how the significance of the award was to be assessed, the applicants maintained that a two-fold test was required: did the facts support a “relatively large award” and even if so (as was accepted in the present case), was the award exceptional. In this latter respect, it was over three times any award previously upheld by the Supreme Court (the McDonagh case), the awards in the Denny and Dawson being irrelevant for the precise reasons outlined above by the Government. The Supreme Court accepted that the present award went to the “top of the bracket” and, in a later case, that a much lower award was in the “highest bracket of damages appropriate in a libel case” comparable to the general damages awarded in “the most serious cases of paraplegic or quadriplegic injuries” (O'Brien v. M.G.N, paragraphs 54-63 above).

98. The Government argued that the domestic safeguards against disproportionate awards were adequate.

Most importantly, they underlined that the Irish Constitution expressly protected freedom of expression and one's reputation. Central to striking a balance between these two rights was a fundamental notion of constitutional law, namely that of proportionality. It was a notion which was equivalent to the Convention concept: the applicants disagreement with this amounted to saying that the Supreme Court was mistaken or that it did not mean what it said. It was a notion which was an important aspect of Irish libel law and a significant safeguard at first (jury) and second instance (the Supreme Court) in libel cases. It was consequently a key factor distinguishing the present case from the Tolstoy Miloslavsky case. The Government also emphasised that its choice of how to provide adequate safeguards fell within the State's margin of appreciation.

99. The applicants reiterated that, compared to the Tolstoy Miloslavsky case, the present jury had even less guidance and the Supreme Court did not exercise a more stringent review. Accordingly, if the law in that case violated Article 10 of the Convention, so did the domestic law at issue in the present case. They accepted that a State enjoyed a margin of appreciation as regards how it complied with Article 10 of the Convention: however, it was much reduced given the press and political speech context. In addition, while that margin meant that a State could choose how to develop the safeguards and, notably, could develop them differently to the Court of Appeal in the above-described Rantzen and John v. M.G.N. cases, this did not change the fact that, as domestic law stood at the relevant time, it was in violation of Article 10 of the Convention.

100. The Government underlined the cherished nature of the principle that lay persons were considered the most effective arbiters when deciding, not only what was defamatory, but the appropriate level of compensation. The applicants were effectively asking the Court to assume that jurors were unable to value reputation in accordance with certain factors outlined to them in order to arrive at a rational and proportionate decision without further guidance. Not only was that an inappropriate assumption, but the calculation made by a jury attracted an even wider margin of appreciation than that completed by, for example, a judge. In this latter respect, they explained why framing and applying defamation laws in a modern democracy was a complex exercise requiring a delicate calibration of a variety of interests. The domestic authorities were therefore clearly better placed to judge how the most appropriate balance could be struck in a given situation and, further, an authority comprising a group of informed, reasonable and conscientious citizens (a jury) would be best placed to reach that balance given their direct and continuous contact with the realities of life within their countries.

Different methods of guiding jurors in other jurisdictions were not necessarily the only means of achieving a proportionate jury award. The Supreme Court was entitled to consider that allowing comparative figures to be supplied to jurors would lead to them being buried in figures (from the parties and the trial judge, which they would assimilate with difficulty and would lead to confusion) and would therefore be an unjustifiable invasion in their province. Personal injury awards were not (as the Supreme Court also found) useful given the unique nature of the libel action and awards in other libel cases would not assist as each libel case fell to be considered on its own facts.

101. In any event, the present jurors were given greater guidance than those in the Tolstoy Miloslavsky case. They were advised in accordance with the constitutional principle of proportionality: if the word “proportionate” was not used, they were told repeatedly (and to a far greater extent than in the Tolstoy Miloslavsky case) to tailor the award to the harm to reputation. The comments of the present trial judge were more moderate than in the Tolstoy Miloslavsky case. Even though the guidance did not outline awards in prior cases, the jury was advised of a hierarchy in the gravity of libels and given an example of a relatively minor libel of someone of similar standing to the present plaintiff. The trial judge also explained to the jury relevant factors to take into account in assessing damages.

102. The applicants argued that the guidance to the jury was extremely limited. The jury was advised: to be reasonable and fair; of the purpose of awarding damages; and to compare the defamation with other possible defamations. No figures were opened to it so no awards in other libel or personal injuries' cases could be mentioned. The Chief Justice even accepted that the present trial judge was restricted by law to giving the jury “guidance of so general a nature as to be meaningless”. Moreover, a system which deprived the first instance determining body of the core relevant information (the comparative figures) could never provide adequate and effective safeguards against disproportionate awards. Furthermore, that such comparative figures were used on appeal did not resolve the problem since the threshold at which the appellate court would set aside a first instance award was extremely high.

103. There were two important differences between the directions given to the present jury and to the jury in Tolstoy Miloslavsky. The first was not relevant – it was simply of no assistance to a jury to tell it that it should assess a particular defamation in the context of other defamations without giving the jury any information about the awards in the other cases. Indeed, the Chief Justice pointed out that a jury must base its assessment entirely on the facts found by them, that departure from this principle would lead to utter confusion and that figures awarded in other cases based on different facts were not matters which the jury was, or should be, entitled to take into account. The second difference demonstrated that the guidance to the Tolstoy Miloslavsky jury was, in fact, stronger than in the present case. The trial judge in the former case was able to mention the ability of money to purchase particular items (a house) whereas no such guidance was, or could have been, given under Irish law by the present trial judge.

104. While it was not for the applicants to propose solutions (the State could examine the possibilities in accordance with its margin of appreciation if the Court accepted that the absence of guidelines was a breach of Article 10), they considered (for the reasons outlined by Mrs Justice Denham – paragraphs 38-40 above) that a jury could be usefully given comparative figures from other libel and personal injury cases.

105. The Government maintained that the proportionality test applied on appeal was stricter than the “irrationality” test applied on appeal in Tolstoy Miloslavsky and that the applicants were simply incorrect in arguing any differently. Indeed the English Courts had, since incorporation of the Convention, recognised the limitations of the “irrationality” test when compared to the Convention proportionality test. The enhanced control resulting from the application of such a test was evidenced by the depth of the Supreme Court's review in the present case. The fact that the present jury award was upheld did not, of course, mean that the appellate test of proportionality was inadequate and the Government considered the overturning of later substantial awards as demonstrative of the fact that the appeal review was an effective safeguard (the above-noted cases of Dawson and Dawson v. the Irish Brokers Association and O'Brien v. M.G.N.).

106. The applicants accepted that the Supreme Court tested the “proportionality” of the award but pointed out that its measure of “proportionality” was far below that of the Convention.

In particular, there was no difference between the proportionality test of the Supreme Court and the pre-Rantzen “irrationality” test. The difference between the pre-Rantzen test (considered insufficient in Tolstoy Miloslavsky) and the post-Rantzen one (later approved in the Tolstoy Miloslavsky judgment) was the development towards a test of “necessity”, a concept which mirrored the Convention notion of proportionality but not the Irish Supreme Court's notion. However, the Supreme Court expressly rejected the Rantzen “necessity” test, it stated that it could only set aside an award if it was satisfied that in all the circumstances the award was so disproportionate to the injury suffered and wrong done that no reasonable jury would have made such an award and that was precisely the formulation considered insufficient by the Court in Tolstoy Miloslavsky. This was not a distinction without a difference: an award could be considered reasonable but not “necessary” to compensate.

107. In any event, the applicants considered that an appellate review (even applying the correct “necessity” test”) could not, of itself, constitute a sufficient safeguard against disproportionate awards. In the first place, the cherished “sanctity” of jury awards militated against and discouraged disturbing such awards on appeal. Secondly, if such reverence was to be accorded to a jury award, then that jury process must itself respect Article 10 of the Convention. Thirdly, it would be destructive of a defendant's Article 10 rights to be obliged to risk the high costs of an appeal in order to defend those rights: this was particularly so when the net result of a successful appeal is simply a reference back to the High Court where the whole flawed process would start again without, moreover, any information about the original jury award or of the appeal court's views.

108. The fact that other awards had been set aside did not prove, in the applicants' opinion, that the control exercised by the Supreme Court was adequate in the present case, not least because the Government were able to refer to only two such cases (the above-cited cases of Dawson and Dawson v. the Irish Brokers Association and O'Brien v. M.G.N.).

109. The parties did not dispute that the award of damages was an interference with the applicants' freedom of expression, that it pursued the legitimate aim of protecting Mr de Rossa's reputation or that the interference was “prescribed by law”.

The Court does not see any reason to disagree. It considers that the award constituted an interference with the second applicant's Article 10 rights (it published the relevant newspaper article and paid the damages' award) and with those of the first applicant (the parent company was the named defendant in the domestic proceedings). It further considers that that interference was “prescribed by law” (the above-cited Tolstoy Miloslavsky case, §§ 38-44) and pursued the legitimate aim of protecting the “reputation and the rights of others”.

110. The parties also agreed, and indeed it was made clear in the Tolstoy Miloslavsky judgment (at § 49), that an award of damages following a finding of libel must be “necessary in a democratic society” so that it must bear a reasonable relationship of proportionality to the injury to reputation suffered. The jurisprudence does not provide for a shifting protection of the rights involved once libel is established (as suggested by the Government at paragraph 90 above): rather the Court assesses whether the compensatory response to a libel was proportionate one by finding where the appropriate balance lies between the conflicting Convention rights involved (Von Hannover v. Germany, no. 59320/00, § 58, ECHR 2004-...).

111. However, the parties diverged on the question of whether the present award was proportionate. The applicants considered the award to be of such significance that the Court could not conclude as to its proportionality without examining the adequacy and effectiveness of the domestic safeguards against disproportionate awards and maintained that their application was indistinguishable from that of Tolstoy Miloslavsky. The Government were of the view that the issues raised were more complex than a mechanical application of that judgment and that, in any event, the present case was clearly distinguishable from the Tolstoy Miloslavsky case.

112. The Court considers that the Tolstoy Miloslavsky judgment must be its point of departure in examining this case. That judgment reads, in so far as relevant, as follows:

“48. The Court recalls at the outset that its review is confined to the award as it was assessed by the jury, in the circumstances of judicial control existing at the time, and does not extend to the jury's finding of libel. It follows that its assessment of the facts is even more circumscribed than would have been the case had the complaint also concerned the latter.

In this connection, it should also be observed that perceptions as to what would be an appropriate response by society to speech which does not or is not claimed to enjoy the protection of Article 10 of the Convention may differ greatly from one Contracting State to another. The competent national authorities are better placed than the European Court to assess the matter and should therefore enjoy a wide margin of appreciation in this respect.

49. On the other hand, the fact that the applicant declined to accept Lord Aldington's offer to settle for a lesser sum ... does not diminish the United Kingdom's responsibility under the Convention in respect of the contested damages award.

However, the Court takes note of the fact that the applicant himself and his counsel accepted that if the jury were to find libel, it would have to make a very substantial award of damages ... . While this is an important element to be borne in mind it does not mean that the jury was free to make any award it saw fit since, under the Convention, an award of damages for defamation must bear a reasonable relationship of proportionality to the injury to reputation suffered.

The jury had been directed not to punish the applicant but only to award an amount that would compensate the non-pecuniary damage to Lord Aldington ... The sum awarded was three times the size of the highest libel award previously made in England ... and no comparable award has been made since. An award of the present size must be particularly open to question where the substantive national law applicable at the time fails itself to provide a requirement of proportionality.

50. In this regard it should be noted that, at the material time, the national law allowed a great latitude to the jury. The Court of Appeal could not set aside an award simply on the grounds that it was excessive but only if the award was so unreasonable that it could not have been made by sensible people and must have been arrived at capriciously, unconscionably or irrationally ...

In a more recent case, Rantzen v. [M.G.N.], the Court of Appeal itself observed that to grant an almost limitless discretion to a jury failed to provide a satisfactory measurement for deciding what was “necessary in a democratic society” for the purposes of Article 10 of the Convention. It noted that the common law – if properly understood - required the courts to subject large awards of damages to a more searching scrutiny than had been customary.

As to what guidance the judge could give to the jury, the Court of Appeal stated that it was to be hoped that in the course of time a series of decisions of the Court of Appeal, taken under section 8 of the Courts and Legal Services Act 1990, would establish some standards as to what would be “proper” awards. In the meantime the jury should be invited to consider the purchasing power of any award which they might make and to ensure that any award they made was proportionate to the damage which the plaintiff had suffered and was a sum which it was necessary to award him to provide adequate compensation and to re-establish his reputation ...

The Court cannot but endorse the above observations by the Court of Appeal to the effect that the scope of judicial control, at the trial and on appeal, at the time of the applicant's case did not offer adequate and effective safeguards against a disproportionately large award.

51. Accordingly, having regard to the size of the award in the applicant's case in conjunction with the lack of adequate and effective safeguards at the relevant time against a disproportionately large award, the Court finds that there has been a violation of the applicant's rights under Article 10 of the Convention. ...

55. In sum, the Court concludes that the award was "prescribed by law" but was not "necessary in a democratic society" as there was not, having regard to its size in conjunction with the state of national law at the relevant time, the assurance of a reasonable relationship of proportionality to the legitimate aim pursued. Accordingly, on the latter point, there has been a violation of Article 10.”

113. Accordingly, the essential question to be answered in the present case is whether, having regard to the size of the present award, there were adequate and effective domestic safeguards, at first instance and on appeal, against disproportionate awards which assured a reasonable relationship of proportionality between the award and the injury to reputation.

114. This examination, with due regard to relevant Irish domestic law and practice, will necessarily determine the well-foundedness of the Government's general argument (see paragraph 89 above) that, inter alia, the applicants' reliance on the Tolstoy Miloslavsky judgment was incorrect. In addition, it is not necessary to rule on whether the present damages' award had, as a matter of fact, a chilling effect on the press: as matter of principle, unpredictably large damages' awards in libel cases are considered capable of having such an effect and therefore require the most careful scrutiny (Bladet Tromsø and Stensaas v. Norway [GC], no. 21980/93, § 64, ECHR 1999-III). Accordingly, and even if, as the Government argued, the assessment of damages in libel cases is inherently complex and uncertain, any such uncertainty must to be kept to a minimum.

115. The Court considers that a general finding that an award of damages is “unusual” is sufficient to prompt its review of the adequacy and effectiveness of the domestic safeguards against disproportionate awards.

The depth of that review does not depend, as the applicants suggested (paragraph 96 above), on how unusual the award is: once a review is triggered as above, the Court will apply the Convention provisions and jurisprudence equivalently in each case. Neither does it accept the Government's argument that an award must be found disproportionate before a safeguards' review becomes relevant: the Tolstoy Miloslavsky case clearly shows that the former conclusion depends on the latter review.

116. The Court has assessed the present damages' award in the same manner as it did in its Tolsoy Miloslavsky judgment: while the defamation was undoubtedly serious, the present award was three times more than the highest libel award ever previously approved by the Supreme Court and the Government have not pointed to a “comparable award” made since then (the Government's submissions at paragraphs 93-94 above and Tolstoy Miloslavsky judgment, § 49). The seriousness of the libel has therefore only some relevance to this general assessment: Count Tolstoy Miloslavsky accepted that any award against him would be substantial and, while that was considered to be “an important element to be borne in mind”, it did not prevent the Court in that case from reviewing the domestic safeguards.

The Court does not consider useful the Government's direct mathematical comparison of the awards in the present and in the Tolstoy Miloslavsky cases in the light of the case- and country-specific matters which influence jury awards in different jurisdictions and in different cases (Tolstoy Miloslavsky, at § 48). In addition, since the finding of unusualness is general and merely acts as a trigger to further examination as described above, it could not reasonably be interpreted (as the Government suggested, see paragraph 93) as the fixing by this Court of a cap on damages awards in Irish libel cases or, still less, as this Court second-guessing a domestic finding of proportionality.

117. However, it is true that, prior to the Supreme Court's judgment in the present case, Irish juries had already made relatively similar awards in libel cases (IR275,000 in Denny v. Sunday News and IR£250,000 O'Brien v. M.G.N. Ltd, both cited above) and, indeed, a significantly higher award (IR515,000 in Dawson and Dawson, cited above). While the Denny case was not defended or appealed and the awards in the Dawson and O'Brien cases were subsequently set aside by the Supreme Court, this domestic case-law, nonetheless, indicates that the award against the first applicant was not as unusual as that at issue in Tolstoy Miloslavsky: in that case the award was three times the size of the highest libel award ever previously made in England and it had not yet been matched when this Court examined that case.

118. While this constitutes a relevant point of distinction between the present and the Tolstoy Miloslavsky cases, the Court considers that the present jury award remained sufficiently unusual as to require this Court's review of the adequacy and effectiveness of the domestic safeguards against disproportionate awards.

119. The parties disputed whether or not the required safeguards imposed on trial judges (and the parties to proceedings) the necessity to provide further and more specific guidance to juries on the level of the damages to be awarded.

120. The Court recalls that the main purpose of the Convention is to lay down certain international standards to be observed by States but that this does not mean that absolute uniformity is required. A State remains free to choose the measures which it considers best adapted to address domestically the Convention matter at issue (the Belgian Linguistic case (preliminary objection), judgment of 9 February 1967, Series A no. 5, p. 19 and Sunday Times v. the United Kingdom (no. 1), judgment of 26 April 1979, Series A no. 30, § 59). Accordingly, while in Tolstoy Miloslavsky the Court endorsed the developments in English law towards giving such further guidance in the Ranzen and St John cases (both cited above), this does mean that it considered the Rantzen route as the only means of safeguarding respondents against disproportionate awards. The important question is whether, having regard to the entire proceedings, the protection against disproportionate awards sufficed.

121. The present trial judge directed the jury that damages had to be confined to such sum of money as would “fairly and reasonably” compensate the injured party for his injured feelings (including for any hurt, anxiety, trouble and bother to which the injured party had been put) and for any diminution in his standing among right-thinking people. The assessment was to be made entirely on the facts established by the jury and relevant considerations were to include the nature of the libel, the standing of the injured party, the extent of the publication and the conduct of the newspapers at all stages of the case. The trial judge also gave the jury an example of a case (without naming it) of a relatively minor defamation to allow the jury to fix the present defamatory article in the overall scale of seriousness of defamatory remarks. He went on to indicate that he would not therefore have been surprised if the present jury would go to the opposite (higher) end of the scale: indeed, the trial judge clearly indicated that any damages awarded would be “substantial”.

122. While the traditional limitations on providing more specific guidance to juries on the level of the award were similar in the present and Tolstoy Miloslavsky cases, the Court has approached with some caution a comparison of the relative merits of the actual jury directions given in those cases since such directions are inevitably framed to respond to specific issues arising at the different trials.

It is also true that Irish law (paragraphs 23-24 above) required damages to be fair and reasonable in the circumstances and not to be disproportionate to the injury to reputation suffered. However, even if that notion of proportionality enhanced the principles of compensatory damages at issue in the Tolstoy Miloslavsky case (paragraphs 128-129 below), the present trial judge did not expressly remit this notion to the jury as he could have. It is not therefore possible to rely, as the Government did, on this element of Irish law to distinguish the present jury guidance to that given in the Tolstoy Miloslavsky case.

123. There are, however, points of distinction between the directions given to the juries in the Tolstoy Miloslavsky and the present cases. In the former case, the jury was asked to consider the purchasing power of money and reference was made to the price of a house. Nonetheless, that direction remained a somewhat imprecise and obvious one rather than constituting any form of indication as to the level of damages it could award in that case. Of some note was the reference by the trial judge on two occasions in his charge to the jury to the use by the defendant himself of the word “enormous” to describe the possible level of damages. However, he emphasised that that was a matter for the jurors. In contrast, in the present case, the trial judge gave the jury two concrete indications, not provided in the Tolstoy Miloslavsky case, as to the level of any damages to be awarded. He provided the example of a relatively minor defamatory comment to allow the jury in the present case to assess the relative seriousness of the defamatory article published by the second applicant. He then followed up that example with a clear direction to the jury that, if it was to award damages, they would have to be substantial.

124. The Court considers therefore that the trial judge's directions in the present case can be considered to have given somewhat more specific guidance to the jury than those examined in the Tolstoy Miloslavsky case.

125. In its Tolstoy Miloslavsky judgment, the Court found inadequate a review which examined whether the award was “so unreasonable that it could not have been made by sensible people and must have been arrived at capriciously, unconscionably or irrationally”. That judgment also endorsed the Rantzen appellate test (whether “a reasonable jury would have thought that this award was necessary to compensate the plaintiff and to re-establish his reputation” – see paragraph 73 above).

126. The general principles of compensatory damages in Irish libel cases are noted at paragraph 122 above. The meaning of proportionality was also developed in some detail by the Chief Justice in the present case: he pointed out that finding a due balance between conflicting constitutional rights (in the present case those guaranteed by Articles 40(3)(2) – reputation - and 40(6)(1) – expression - of the Constitution) relied on the notion of proportionality in Irish law, which concept mirrored that of the Convention (see also certain Irish constitutional cases applying the notion of proportionality at paragraphs 67-70 above).

However, the Chief Justice also explained in some detail why the depth of appellate review of awards, for compliance with those principles of compensatory damage, was limited. Having underlined the “unusual and emphatic sanctity” of jury awards so that Irish appellate courts had been “extremely slow” to interfere with such awards, he expressly disagreed with the above-outlined Rantzen appellate test because he considered that its application would remove the sanctity of jury awards and would mean that an appellate court would no longer give “real weight” to the possibility that the jurors' judgment was to be preferred to that of the judge. Accordingly, the Chief Justice described the level of appellate control of jury libel awards as follows (see paragraphs 28-30 above):

“..., while awards made by a jury must, on appeal be subject to scrutiny by the appellate court, that Court is only entitled to set aside an award if it is satisfied that in all the circumstances, the award is so disproportionate to the injury suffered and wrong done that no reasonable jury would have made such an award.”

128. They argued, in the first place, that this test was, in substance, no stricter than the inadequate appellate review in the Tolstoy Miloslavsky case. The Court considers this incorrect and is of the view that the appellate review is one of the main points of distinction between the two cases.

It is true that the Chief Justice stated that the depth of appellate review in Irish law could not be as intrusive as that developed in the Rantzen case (paragraph 126 immediately above) cited with approval in the Tolstoy Miloslavsky judgment (at § 50). It nevertheless remains that the nature of the Supreme Court's review was more robust than that at issue in the Tolstoy Miloslavsky judgment because of the requirement in Irish domestic law that jury awards in libel cases be proportionate within the meaning described at paragraphs 122-126 above. It was the absence of this proportionality requirement in English law which meant that the libel award in the Tolstoy Miloslavsky case was considered to be “particularly open to question” (the last sentence of § 49 of that judgment).

129. That this requirement of proportionality distinguishes the appellate review at issue in the present and Tolstoy Miloslavsky cases is evident from the actual review conducted by the Supreme Court in the present case.

The Supreme Court (see paragraphs 31-36 above) took into account a number of relevant factors, including the gravity of the libel, the effect on Mr de Rossa (a leader of a political party) and on his negotiations to form a government at the time of publication, the extent of the publication, the conduct of the first applicant newspaper and the consequent necessity for Mr de Rossa to endure three long and difficult trials. Having assessed these factors, it concluded that the jury would have been justified in going to the top of the bracket and awarding as damages the largest sum that could fairly be regarded as compensation. While IR£300,000 was a substantial sum, it noted that the libel was serious and grave, involving an imputation that Mr de Rossa was involved in or tolerated serious crime and personally supported anti-Semitism and violent Communist oppression. “Bearing in mind that a fundamental principle of the law of compensatory damages is that the award must always be reasonable and fair and bear a due correspondence with the injury suffered and not be disproportionate thereto”, the Supreme Court was not satisfied that the present jury award went beyond what a reasonable jury applying the law to all the relevant considerations could reasonably have awarded and considered it “not disproportionate to the injury suffered by the Respondent.”

130. While the Court has noted the comments of Mr Justice Geoghegan (partly dissenting) in the above-cited O'Brien v. MGN Ltd case concerning the “diffidence” with which an appeal review of a jury libel award is conducted (see paragraphs 59-62 above), that judge endorsed the principles of review outlined by the Supreme Court in the present case which principles led, in fact, to the jury award being quashed in the O'Brien case.

131. Secondly, the applicants considered this appellate review incapable of remedying the “defects” (insufficient guidelines) at first instance for the following reasons. They maintained that the review could not operate properly once all relevant information had not been opened to the first instance decision-maker; they underlined the heavy costs' implications of relying on an appeal to obtain an informed view on damages; and they pointed out (along with certain third parties and the LAG) that, even if that appeal was successful, the Supreme Court could not substitute its own award but rather sent cases back for re-trial on damages before a new jury which was not informed of the appellate intervention.

However, the fact that the present jury was not given such figures clearly did not prevent the Supreme Court from carrying out its own assessment of the proportionality of the award. Reimbursement of legal costs on appeal can be claimed by the successful party and, as a general rule, costs follow the event (see, for example, Dawson and Dawson v. Ireland (dec.), no. 21826/02, pp. 2 and 12, 8 July 2004). Whether or not this re-trial process could be considered unnecessarily cumbersome as argued (see, for example, the above-cited Dawson case), the present applicants would only have been relevantly affected by this if there had been a finding by the Supreme Court in their favour.

132. Accordingly, having regard to the particular circumstances of the present case, notably the measure of appellate control, and the margin of appreciation accorded to a State in this context, the Court does not find that it has been demonstrated that there were ineffective or inadequate safeguards against a disproportionate award of the jury in the present case.

Holds by 6 votes to 1 that there has been no violation of Article 10 of the Convention.

Done in English, and notified in writing on 16 June 2005, pursuant to Rule 77 §§ 2 and 3 of the Rules of Court.

In accordance with Article 45 § 2 of the Convention and Rule 74 § 2 of the Rules of Court, the following dissenting opinion of Mr Cabral Barreto is annexed to this judgment.

1. Freedom of expression constitutes one of the essential foundations of a democratic society and one of the basic conditions for its progress and for each individual's self-fulfilment. Subject to paragraph 2 of Article 10, it is applicable not only to “information” or “ideas” that are favourably received or regarded as inoffensive or as a matter of indifference, but also to those that offend, shock or disturb.

These principles are of particular importance with regard to the press. While it must not overstep the bounds set, inter alia, for “the protection of the reputation of others”, its task is nevertheless to impart information and ideas on political issues and on other matters of general interest.

As to the limits of acceptable criticism, they are wider with regard to a politician acting in his public capacity than in relation to a private individual.

A politician inevitably and knowingly lays himself open to close scrutiny of his every word and deed by both journalists and the public at large, and he must display a greater degree of tolerance, especially when he himself makes public statements that are susceptible of criticism.

Determining whether the interference in question was “necessary in a democratic society” requires the Court to establish whether it corresponded to a “pressing social need”, whether it was proportionate to the legitimate aim pursued and whether the reasons given by the national authorities to justify it are relevant and sufficient (see Lopes Gomes da Silva v. Portugal, no. 37698/97, § 30, ECHR 2000-X).

2. Above all, I would emphasise that the present case clearly involved a political debate on matters of general interest, an area in which restrictions on the freedom of expression should be interpreted narrowly.

That said, it seems to me that the fundamental issue in this case is whether the award of damages was proportionate to the legitimate aim of protecting Mr de Rossa's reputation or rights.

The majority, however, attach too much importance to the safeguards afforded by Irish law for reviewing domestic decisions (see paragraph 114 of the judgment).

I am not disputing the value of these safeguards, but that does not seem to me to be a sufficient reason for finding that there has been no violation of Article 10.

The important aspect to my mind was, rather, not only whether the safeguards functioned properly but also whether, despite the margin of appreciation enjoyed by the domestic authorities, the final decision was consistent with the principles set forth in our case-law.

Weighing up all the circumstances of the case, I came to the conclusion that the amount of damages which the first applicant was ordered to pay, notwithstanding the review of the award by the Supreme Court, was so high that the reasonable relationship of proportionality between the interference and the legitimate aim pursued was not observed.

I therefore consider that there was a violation of Article 10 of the Convention.