Before examining the proposals in Lord Lester’s Defamation Bill it is perhaps worth summarising shortly the existing state of the Reynolds common law defence. To rely on the defence, the defendant must show, first, that that there is a real public interest in communicating and receiving the information (Jameel v Wall Street Journal  UKHL 44 , Baroness Hale). This is a matter of law for the judge (, Lord Hoffmann) and involves drawing a line between matters in which the public is legitimately interested and matters in which the public might have an interest, but which are nevertheless not matters of public interest ( Baroness Hale).
The second stage requires the court to answer whether the publication was made responsibly. This question is answered by having regard, inter alia, to the ten ‘non-exhaustive’ matters set out by Lord Nicholls in Reynolds bearing in mind that they are not hurdles to be cleared but factors to take into account and as necessary balanced against each other ( Lord Hoffmann;  Lord Bingham). While it has been said that an appellate court should be cautious before overturning the decision of the first instance judge on the appropriate balance, the Court of Appeal has recently made clear in Flood v Times Newspapers ( EWCA Civ 804) that the question whether Reynolds privilege applies does not involve the exercise of judicial discretion but instead a question of law ( and ). As such
‘When a question of law is to be decided there is only one correct answer, however difficult it may be to find. Thus, if the true meaning of a document is in issue, the fact that the construction preferred by the judge is plausible does not prevent an appellate court from deciding the matter for itself.’ 
Lord Nicholls ten non exclusive factors are as follows:
- ‘The seriousness of the allegation. The more serious the charge, the more the public is misinformed and the individual harmed, if the allegation is not true.
- The nature of the information, and the extent to which the subject-matter is a matter of public concern.
- The source of the information. Some informants have no direct knowledge of the events. Some have their own axes to grind, or are being paid for their stories.
- The steps taken to verify the information.
- The status of the information. The allegation may have already been the subject of an investigation which commands respect.
- The urgency of the matter. News is often a perishable commodity.
- Whether comment was sought from the plaintiff. He may have information others do not possess or have not disclosed. An approach to the plaintiff will not always be necessary.
- Whether the article contained the gist of the plaintiff’s side of the story.
- The tone of the article. A newspaper can raise queries or call for an investigation. It need not adopt allegations as statements of fact.
- The circumstances of the publication, including the timing.’ (Reynolds v Times Newspapers  2 AC 127, 205)
The burden of proving the existence of Reynolds privilege is on the defendant (Ibid, 203) who must show that there was a real public interest in publishing the matter complained, that the inclusion of the words complained of was justifiable, and that in the circumstances publication was made responsibly (See, for example, Lord Neuberger in Flood v Times Newspapers  EWCA Civ 804, at ). The existence or otherwise of qualified privilege is to be judged in all the circumstances at the time of the publication. Where the publication relates to a situation occurring in another country or where the sources for the defendant’s story are in a different country, it was accepted by Gray J in Al Misnad v Azzaman Ltd ( EWHC 1783 (QB)) that the test of responsible journalism may have to be adapted to take account of these factors.
While in most cases it will be a media defendant who wishes to rely on the defence, there is Privy Council authority to the effect that the privilege can be asserted by a non-media defendant (Seaga v Harper  UKPC 9).
Although there is Court of Appeal dicta to the contrary, (Kearns v General Council of the Bar  EWCA 331.) it is submitted that this is inconsistent with dicta from the House of Lords decision in Jameel (at  and ) and that the Privy Council decision should be followed. The question remains open whether at common law, the privilege can be asserted in respect of opinion. The view of the learned editors of Duncan & Neill (Duncan and Neil on Defamation (Butterworths, 3rd edition, 2009), at 17.26. is that it can but the better view, which was assumed to be the case by at least two of their lordships in Reynolds (at 201 and 193-5 per Lord Nicholls and 237-8 per Lord Hobhouse.) is that it cannot.
Lord Lester’s Responsible Publication defence
Under clause 1(1) of the Defamation Bill 2010, it is provided that any defendant has a defence where he shows that
‘(a) the words or matters complained of were published for the purposes of, or otherwise in connection with, the discussion of a matter of public interest; and (b) the defendant acted responsibly in making the publication.’
The clause gives statutory endorsement to, and builds on, the common law defence in Reynolds’ and the Explanatory Notes state that its purpose is to ‘strengthen the protection afforded to the publication of matters of public interest whilst maintaining a robust standard of responsibility in publishing,’ (Explanatory Notes, at para 52.) it having been said previously that, ‘the common law defence of so-called ‘Reynolds privilege’, developed by the House of Lords, has not been as useful to publishers as had been hoped.’ (Explanatory Notes, at para 5(vi)).
On first reading, it might appear that, a few minor differences aside, this provision simply represents a codification of the common law privilege first articulated in Reynolds and later developed by the House of Lords in Jameel . In my view, however, the clause goes much further than the existing law and in so doing unjustifiably extends the protection currently available to defendants where they get their facts wrong.
The omission of any express reference to sources, the status of the information, whether the article contained the gist of the claimant’s side of the story and the tone of the article.
By clause 1(3), all the circumstances of the case are be taken into account in determining whether the defendant has acted responsibly and clause 1(4) provides a non-exclusive list of circumstances which may be relevant. At first sight this is very similar to the common law position, until one turns to the non-exclusive list and realises that four circumstances mentioned in Lord Nicholls’s speech do not make it into clause 1(4), that is to say; the source of the information, the status of the information, whether the article contained the gist of the claimant’s side of the story and the tone of the article. The omission of these circumstances was clearly deliberate and to say that they are caught by the general provision in clause 1(3) is inadequate. Why leave these out while retaining the others? The question therefore arises as to what the consequences of the omission of these circumstances might be?
Taking the circumstances of tone and whether the article contained the gist of the claimant’s first, it is strongly arguable that the intention of the drafters was to make clear that balance and restraint in reporting is not necessary to rely on Reynolds (though it is to rely on the reportage defence in clause 1(5)). What seems now to be required is that the ‘information gathering process’ should have been responsible (and that is consistent with the main terms of the Press Code) but that the way in which the story is told and the slant given does not have to be. The consequence of this is presumably that, should a case with similar facts to Grobbelaar v News Group Newspapers ( EWCA Civ 33.) Galloway v Telegraph Group ( EWCA Civ 17) Radu ( EWCA Civ 921) and even Reynolds recur, they would be decided differently as in all these cases an important consideration against the existence of the privilege was the fact that the claimants’ side of the story was not given and the writing in Grobbelaar and Galloway involved unacceptable embellishment of the facts and a ‘verbal kicking’ of the claimant. Should this be the practical result of the Bill, I would regard it as an unacceptable extension of protection to defendants. If ‘responsible’ means anything it should surely require that the claimant’s side of the story is put and the tone is measured. As Lord Nicholls noted in Reynolds itself,
‘It goes without saying that a journalist is entitled and bound to reach his own conclusions and to express them honestly and fearlessly. He is entitled to disbelieve and refute explanations given. But this cannot be a good reason for omitting, from a hard-hitting article making serious allegations against a named individual, all mention of that person’s own explanation … it is elementary fairness that, in the normal course, a serious charge should be accompanied by the gist of any explanation already given. An article which fails to do so faces an uphill task in claiming privilege if the allegation proves to be false and the unreported explanation proves to be true.’ ( 2 AC 127, at 206 per Lord Nicholls).
So far as the omission of the source of the information, this is again surprising. While serious difficulties stand in the way of any claimant if the defendant claims that its sources are confidential, if a claimant could establish that the defendant had relied on a single source known to be hostile to the claimant this would surely be powerful evidence that the defendant had not acted responsibly. For example, in James Gilbert Ltd v MGN ( EMLR 680) which involved allegations of child labour based on one source upon whom Eady Jsaid that a responsible journalist could place no reliance:
‘By any reasonable and objective test, Mr Ashok Verma cannot have been perceived by The Sunday Mirror journalist at that time as a wholly reliable source when standing alone–and I emphasise those words.’
It may be of course that this is a matter that would be taken into account by a court under clause 1(4)(d) in that it is relevant to the question of what steps were taken by the defendant to verify what was published. The omission of the reference to sources however seems to point, at the very least, to an intention to de-emphasise the factor.
Finally, there is the omission of the status of the information. Clearly the status of the information is potentially relevant, like the source of the information, to its reliability, or at least its apparent reliability: the more official and final the information is, the more a journalist is justified in relying on it. But there is also the matter raised in the recent Flood decision that where information is published at an early stage of an investigation there is, as Moore Bick LJ recognised:
‘a serious risk that once the allegations have been published the person against whom they are made will feel obliged to respond to them publicly, thereby depriving himself of the safeguard of the ordinary process and risking a measure of trial by press. I am not dealing here with the publication of the simple fact that a complaint has been made against a person, without any details being given, or with the publication of the fact that a person has been charged with a criminal offence. Such information is likely to be a matter of public interest. … However, it is unnecessary and inappropriate for such reports to set out the details of the allegations made against the person charged; the description of the charge itself is sufficient to inform the public of what it has an interest in knowing. The alternative is trial by press without proper safeguards, which is clearly not in the public interest.” ( EWCA Civ 804, at ).
Again, it is hard to resist the interpretation that the status of the information was left out deliberately and that the intention was to allow the publication of matters at a much earlier stage than would be permitted under the existing common law. There are, as Moses LJ pointed out in Flood, serious dangers with this if one takes reputation seriously:
‘That a person is accused is generally of far greater interest than his or her subsequent triumphant acquittal. Once an accusation is dismissed, the blaring headline of accusation on page 1 becomes a tepid reference in the graveyard of page 2.’ (Ibid, at )
Moreover, if Lord Lester’s Bill was to pass, it is strongly arguable that, were cases with facts identical to Flood , Miller v Associated Newspapers ( EWHC 2799), Galloway v Telegraph Group ( EWCA Civ 17) and Henry v BBC ( EWHC 2787) to recur, there would be a strong case for arguing that they should be decided differently as in all of these cases a factor militating against privilege was that the status of the information being reported on was insufficiently final.
Not only are the omissions of these factors interesting and revealing in themselves but it is worth noting that the last two common law Supreme Courts that have themselves recognised a defence of responsible publication, Namibia (Trustco International v Shkongo (2010) SA 8/2009) and Canada (Grant v Torstar Corporation  SCC 61) have both emphasised, inter alia, the importance of the status and reliability of sources and also whether the plaintiff’s version of the story was sought and accurately reported. Referring to the issue whether the plaintiff’s side of the story was sought and accurately reported, the Supreme Court of Canada noted that it has been said that ‘this is perhaps the core Reynolds factor’ (at ) and in most cases ‘it is inherently unfair to publish defamatory allegations of fact without giving the target an opportunity to respond.’ (at ). So far as sources were concerned, both courts emphasised the danger of relying on anonymous sources. In Trustco International v Shkongo O’Regan AJA noted the obligation to exercise care to avoid inaccuracy is
‘particularly acute where the original source of the story wishes to remain anonymous given the risk that an anonymous source may be serving a particular agenda not apparent to the journalist. To minimize this risk, the rules of good practice require journalists to investigate the motives of anonymous sources and where possible to base sources upon open forms of information gathering. As the Canadian Supreme Court commented in Grant “it is not difficult to see how publishing slurs from unidentified sources could … be irresponsible”’ 
Not just a media defence
Apart from the omission of these three circumstances from the list of relevant circumstances, clause 1 differs in a number of other ways from its common-law cousin. First, in making the defence available to any defendant, the drafters have made clear that Reynolds privilege is not confined to the media thereby confirming that the decision of the Privy Council in Seaga v Harper ( UKPC 26) to the effect that Reynolds privilege applies to any person who publishes material of public interest also represents English law. This is a sensible clarification but raises a question whether different standards of responsibility will apply depending on whether the defendant is, or is not, part of the media. Though this is not dealt with in the Bill, the status of the publisher is surely a relevant circumstance that a court should take into account under clause 1(4).
‘Subject matter as a whole’ or ‘words or matters complained of’
While the clarification that the defence now applies to any defendant is welcome, the same cannot be said about the wording of clause 1(1)(a) which makes the defence applicable where ‘the words or matters complained of were published for the purposes of, or otherwise in connection with, the discussion of a matter of public interest.’ Under the existing law, it is the publication taken as a whole that is considered in determining whether the matter published was in the public interest (see eg Jameel,  and ) To suggest it is the words or matter complained of that must be of public interest is to focus on the wrong thing. The publication of false and defamatory words can never be a matter of public interest. If the Bill survives, this provision should be changed to make clear that it is the publication as a whole that must be of public interest.
Application of responsible publication to opinion
A second unjustified change in the existing law is made by clause 1(2) which provides that the defence in clause 1(1) applies irrespective of whether the publication contains statements of fact or opinions. The change is justified in the Explanatory Notes on the grounds of the difficulty of distinguishing between fact and comment. It is said that bringing opinion within the defence will mean that ‘it cannot be argued, relying on a technicality, that part of the publication falls outside the defence simply because it not factual in nature.’ (Explanatory Notes, at para 54). Setting aside for a moment the fact that there appears to me to be no reason why Reynolds privilege could not be claimed in respect of defamatory allegations of fact contained in an article merely because there happened also be to be some defamatory opinions in the article, the extension of the defence to cover expressions of opinion gives me serious concern. It goes against the views of the House of Lords in Reynolds where both Lords Nicholls and Hobhouse made clear that Reynolds privilege, when it applies, gives a defence notwithstanding that the facts published were untrue while the expression of opinion is protected by the fair comment ( 2 AC 127, at 201 and 193-5 per Lord Nicholls and 237-8 per Lord Hobhouse).
A division between fact and comment makes sense because where a defamatory opinion is published, readers will recognise it as such and either agree, discount it or reach their own different opinion. That is not to say that opinion should never be actionable just that the basis on which exemption from liability for defamatory opinion should be given should be different from liability for defamatory fact. The capacity of defamatory opinion to mislead is different from that of an inaccurate fact. Where false facts are concerned, readers are unlikely to have any means for determining their truth or otherwise. Thus in respect of false statements of fact protection should only be afforded to the defendant if, for reasons of public policy, the occasion or nature of the publication demands it. Where defamatory opinion is concerned, provided the facts on which the opinion is based are true, the possibility of reasonable readers being misled does not arise. The reason for sanctioning publication of a defamatory opinion must therefore be different. Under English law the publication of defamatory opinion, where based on true facts, is only actionable if it is not an opinion that an honest minded person could hold or the defendant did not actually believe what he published. The problem with the Bill is that in bringing opinion within the defence, the possibility opens up of seriously misleading and unjustified opinions being published that would not be defensible under fair comment becoming defensible under Reynolds. Thus, if the defendant ‘responsibly’ invents facts that turn out to be untrue and then comments on those facts, he would have a defence under Reynolds but not fair comment. The very risk that cannot happen with the fair comment defence (reasonable readers being misled) is thereby allowed to happen. This I do not think is desirable.
The true effect of Lord Lester’s responsible publication clause
My final point on the Reynolds replacement involves a synthesis of the above. It appears that the intention of the authors of the Bill may be to see the law treat as responsible – and hence legitimate – the publication of material that includes vituperative comment, based on erroneous facts, that does not include any explanation offered by the subject. This amounts, in effect, to a comprehensive defence for public interest publication that would be excluded only very rarely on grounds of irresponsibility. I cannot accept that such publications are worthy of protection. Indeed, given that by definition any such commentary would be focused on matters of significant public concern, a law of this type risks errant publications that might do immense damage to the public good. If all this is what is intended, then I think that the authors of the Bill should have been straightforward in presenting this intention instead of relying on an obscurantist sleight of hand. The relation to Reynolds is minimal; the revision is not merely technical. I note also that if the tone of the piece is irrelevant to the detemination of responsibility and the honest opinion defence is to apply to ‘responsibly published’ as well as true underpinning facts (see clause 3(4)(b)), then this would be the position notwithstanding the exclusion of comment from the responsible publication defence.
Finally, clause 1(5) provides a reportage defence that may go beyond the existing law (Despite the warning given by Sedley LJ in Roberts v Gable  EWCA Civ 721  that the defence needs to be ‘treated restrictively’) in that it does not require that the impartial and accurate report be about an ongoing dispute between two parties: all that the Bill requires is that the report be on a pre-existing matter that ‘it is in the public interest for the existence of that matter, and anything reported in connection with it, to be the subject of a report.’ The existing common law defence is an unsatisfactory one (interestingly it has not been uniformly adopted in the United States) sitting uneasily as it does with the repetition rule. It has never seemed to me to be sensible to privilege the publication of material which the writer believes to be untrue simply because the matter being written about is of public interest and the writer reports what has been said accurately and impartially. Rather than codification, I would have preferred to see statutory abolition.
What should the defence look like?
To avoid the risk of being criticised as an armchair academic, I set out below what I believe should be the elements of a responsible publication defence. There is no need for a statutory restatement of the Reynolds defence. In this respect, I am persuaded by what Lord Hoffmann had to say during the second reading debate of Lord Lester’s Bill:
‘But I am slightly puzzled by what it does do—which is to take the public interest defence, as laid down by your Lordships’ House in Reynolds and Jameel, and restate it in its own language. I am always nervous, speaking as a former judge, about legislative attempts to restate rules of common law. They lead to expensive litigation over whether or not Parliament intended to change things. As the Jameel case appeared to be generally welcomed by the press and has been followed by the Canadians, I should have thought that there was a case for leaving well alone.’
More generally, I would make the following points
- The defence should not be restricted to media defendants but should potentially be available to any defendant.
- The defence should be restricted to facts and should not be extended to cover opinion.
- In determining whether the publication is on a matter of public interest the court should have regard to the publication as a whole and not simply the ‘words or matters complained of.’
- The non-exclusive factors identified by Lord Nicholls in Reynolds represent a perfectly sensible statement of the factors that should be taken into account when determining whether the publication was made responsibly. Omitting matters such as the status and source of the information, the story’s tone and whether the gist of the claimant’s story has been given would be an error. These factors are important elements that contribute to the balance between the protection of reputation and freedom of expression that must be struck by Reynolds privilege.
- Though the matter is not directly addressed in the Bill, the courts should not, despite what Lord Nicholls said in Reynolds, give particular weight to freedom of expression when deciding whether the publication was made responsibly (Reynolds, 205). Where the claimant’s article 8 rights are engaged then in considering the application of Lord Nicholls’ ten factors neither the article 10 right nor the article 8 right has any presumptive priority. Instead the approach set out in Re S (A Child) (Identification: restriction on Publication) (2005] 1 AC 593) and In re BBC ( 3 WLR 142, per Lord Brown at  and Lord Hope at -), suitably adapted for the law of defamation, should be followed. Thus, the court should begin from the position that neither the claimant’s article 8 nor the defendant’s article 10 rights have precedence over the other. Instead where these values are in conflict, an intense focus is necessary on the comparative importance of the specific rights being claimed in the individual case. The justification for interfering with or restricting each right must be taken into account and finally the proportionality test must be applied to each. To the extent that Lord Nicholls in Reynolds had indicated that where publication of the matter complained of was of public interest any ‘lingering doubts [as to the appropriate balance] should be resolved in favour of publication’ this cannot stand (see Flood v Times Newspapers Ltd  EWHC 2375  and see the Court of Appeal at  EWCA Civ 804, )
- The reportage variant of the defence should be abolished
Alastair Mullis is Professor of Law and Dean at the Norwich Law School, UEA. The author would like to thank Mr Benjamin Pell for his invaluable discussions of the Defamation Bill and Reynolds privilege generally.