[1]     Binnie J. — This appeal requires the Court to reexamine the defence of fair comment which helps hold the balance in the law of defamation between two fundamental values, namely the respect for individuals and protection of their reputation from unjustified harm on the one hand, and on the other hand, the freedom of expression and debate that is said to be the “very life blood of our freedom and free institutions”: Price v. Chicoutimi Pulp Co. (1915), 1915 CanLII 66 (SCC), 51 S.C.R. 179, at p. 194.  Under the present law, if a plaintiff shows the defendant published something harmful to his or her reputation, then both falsity and damage are presumed, and the onus shifts to the defendants to establish an applicable defence, including the defence of fair comment.  In Cherneskey v. Armadale Publishers Ltd., 1978 CanLII 20 (SCC), [1979] 1 S.C.R. 1067, Dickson J., in dissent, identified the elements of the “fair comment” defence as follows:

[2]     This is a private law case that is not governed directly by the Charter.  Yet it was common ground in the argument before us that the evolution of the common law is to be informed and guided by Charter values.  Particular emphasis was placed on the importance of ensuring that the law of fair comment is developed in a manner consistent with the values underlying freedom of expression.  However, the worth and dignity of each individual, including reputation, is an important value underlying the Charter and is to be weighed in the balance with freedom of expression, including freedom of the media.  The Court’s task is not to prefer one over the other by ordering a “hierarchy” of rights (Dagenais v. Canadian Broadcasting Corp., 1994 CanLII 39 (SCC), [1994] 3 S.C.R. 835), but to attempt a reconciliation.  An individual’s reputation is not to be treated as regrettable but unavoidable road kill on the highway of public controversy, but nor should an overly solicitous regard for personal reputation be permitted to “chill” freewheeling debate on matters of public interest.  As it was put by counsel for the intervener Media Coalition, “No one will really notice if some [media] are silenced; others speaking on safer and more mundane subjects will fill the gap” (Factum, at  para. 14).

[3]     The issue of balance is raised here in the context of a “shock jock” radio talk show hosted by the appellant Rafe Mair, a well-known and sometimes controversial commentator on matters of public interest in British Columbia.  The target of his “editorial” on October 25, 1999 was the respondent Kari Simpson, a widely known social activist.  The context was public debate over the introduction of materials dealing with homosexuality into public schools.  Mair and Simpson took opposing sides in the debate about whether the purpose of this initiative was to teach tolerance of homosexuality or to promote a homosexual lifestyle.  Simpson was a leading public figure in the debate, and the trial judge found that she had a public reputation as a leader of those opposed to any positive portrayal of a gay lifestyle.  The nub of Simpson’s complaint is the following portion of the Rafe Mair editorial broadcast on October 25, 1999:

[4]     The courts in British Columbia were divided on the legal outcome.  The trial judge dismissed the action on the basis that, while statements complained of in the editorial (in particular the imputation that Simpson “would condone violence toward gay people”) were defamatory, nevertheless, the defence of fair comment applied and provided a complete defence ((2004), 31 B.C.L.R. (4th) 285, 2004 BCSC 754, at para. 6).  The Court of Appeal reversed ((2006), 55 B.C.L.R. (4th) 30, 2006 BCCA 287).  In its view, the defence of fair comment was not available because there was no evidentiary foundation for the imputation that Simpson would condone violence; nor had Mair testified that he had an honest belief that Simpson would condone violence.  In my view, with respect, the Court of Appeal unduly favoured protection of Kari Simpson’s reputation in a rancourous public debate in which she had involved herself as a major protagonist.  The factual basis of the editorial was Simpson’s speech.  Mair stated in the editorial that he had listened “to the tape of the parents’ meeting the night before where Kari harangued the crowd”.  Simpson had been making speeches in a similar vein for some time.  Whatever view one may take of Mair’s commentary, the factual basis of the controversy was indicated in the editorial and widely known to his listeners.  In the absence of demonstrated malice on his part (which the trial judge concluded was not a dominant motive), his expression of opinion, however exaggerated, was protected by the law.  We live in a free country where people have as much right to express outrageous and ridiculous opinions as moderate ones.  I would therefore allow the appeal. 

[5]     Rafe Mair’s radio talk show is carried on station CKNW owned and operated by the appellant WIC Radio Ltd., which accepts legal responsibility for the broadcast.  Mair has a reputation for provoking controversy.  With controversy has come a measure of commercial success.  His listeners expect to hear extravagant opinions and, according to his counsel, discount them accordingly.

[6]     The trial judge found that Kari Simpson was a social activist with “a public reputation as a leader of those opposed to schools teaching acceptance of a gay lifestyle” (para. 10).  Simpson had earlier opposed three books placed in Surrey schools which portrayed family units with same-sex parents.  She helped write and promote a Declaration of Family Rights which asserted that children should not be exposed to any teaching which “portrays the lifestyle of gays . . . as one which is normal, acceptable or must be tolerated” (A.R., at p. 387).  The document included a form to be sent by parents to their children’s schools as follows:

[7]     The trial judge found that Kari Simpson’s reputation was earned as a result of her “very public actions and words” (para. 10).  Further, “[h]er reputation was fairly characterized by Mair at trial as the person who was associated by the media with the anti-gay side” (para. 11).  This characterization is supported by various of Kari Simpson’s speeches put in evidence at the trial, including the following extracts:

[8]     The trial judge also emphasized that Simpson championed her views through democratic means, not violence.  She encouraged sympathetic members of her audience to exercise influence through pressure on politicians at all levels.  She urged her listeners “to vote, to write, and to speak out about their values and views” (para. 15).  “As people as voters and tax payers you have enormous power.  Do not underestimate the value of a phone call” (A.R., at p. 535).  “Our mandate [Citizens Research Institute] is to ensure that the electorate is informed and participating in an informed way in the politics of the province” (A.R., at p. 652).  Typical of her speeches is the following exhortation:

[9]     Simpson claimed that certain words in the October 25 broadcast were defamatory in their ordinary and natural meaning.  Simpson alleged, as well, that the following defamatory innuendos were conveyed by words in the broadcast: (a) that she had advocated or was in favour of parents taking their children out of school because the children’s teacher was gay; (b) that she advocated keeping gay people out of Surrey’s public schools; (c) that she was hostile toward gay people to the point that she would condone violence toward gay people; (d) that she preaches hatred against gay people; (e) that she rants against gay people in a way that would influence someone to take the law into his own hands and do them harm; (f) that she would employ tactics against gay people similar to those used by Hitler and other bigots, such as former State Governor George Wallace, Governor Ross Barnett and Governor Orval Faubus; and (g) that she is a dangerous bigot apt to cause harm to gay people.   At trial, Mair testified that no such imputations of violence were intended by him nor in fact made.  He said:

[10] The trial judge noted Mair’s comparisons of Simpson in her public persona to Hitler, Wallace, Faubus, Barnett, the KKK and skinheads.  The meaning to be ascribed to these comparisons is that Simpson “would condone violence” (para. 6); this imputation was found to be defamatory.  Further, the following words, having regard to the comparisons, were also found to be defamatory:

[11] As to the defence of fair comment, the trial judge found that the facts stated in the editorial were true (para. 44).  Kari Simpson had spoken on Bill Good’s show, and at a rally the night before.  On both occasions she was “on message”.  Other facts of the controversy were well known at the time though unstated in the editorial (e.g. that Simpson was an active promoter of the Declaration of Family Rights (para. 52)).  The trial judge concluded:

[12] The trial judge found that the editorial was directed to a matter of public interest.  The issues underlying the broadcast such as tolerance, discrimination, and the place for discussion of homosexuality in public schools, were matters of widespread controversy.  The trial judge found (at paras. 64-66) that there was no basis upon which to challenge that Mair honestly believed what he said.  There was some evidence of intrinsic malice:

[13] The trial judgment was reversed.  In the view of Southin J.A., speaking for herself and Thackray J.A., the trial judge had “failed to apply the test of honest belief in the defamatory imputation” (para. 34).  The question, she said, was whether to succeed in the defence of fair comment, the defendant must honestly believe in the imputation (i.e. the  innuendo that Simpson “would condone violence toward gay people”, found by the trier of fact) or “need he only have an honest belief in what he himself subjectively intended by the words which he used?” (para. 37).  Subjectively, Mair had intended to convey simply that Simpson is an intolerant bigot.  If that was all he had said, he would be entitled to succeed because the facts indicate that this was his honest belief.  However, Southin J.A. pointed out, the trial judge concluded, in para. 30 of her reasons, that the objectively reasonable meaning of Mair’s words was that “the plaintiff would condone violence”.  She held this was an imputation of fact, not comment, and that there was “no evidentiary foundation for a finding that the appellant would condone violence” (para. 43).  Prowse J.A. delivered brief concurring reasons.  The case was therefore sent back to the trial judge for an assessment of damages.

[14] In the almost 30 years since Cherneskey, courts across the common law world have re-examined the balance between freedom of expression and the protection of private reputation.

[15] The function of the tort of defamation is to vindicate reputation, but many courts have concluded that the traditional elements of that tort may require modification  to provide broader accommodation to the value of freedom of expression.  There is concern that matters of public interest go unreported because publishers fear the ballooning cost and disruption of defending a defamation action.  Investigative reports get “spiked”, the Media Coalition contends, because, while true, they are based on facts that are difficult to establish according to rules of evidence.  When controversies erupt, statements of claim often follow as night follows day, not only in serious claims (as here) but in actions launched simply for the purpose of intimidation.  Of course “chilling” false and defamatory speech is not a bad thing in itself, but chilling debate on matters of legitimate public interest raises issues of inappropriate censorship and self-censorship.  Public controversy can be a rough trade, and the law needs to accommodate its requirements.

[16] Canadian courts have frequently pointed to the need to develop the common law in accordance with Charter values, including the law of defamation:

[17] The appellants claim that while the trial judge properly treated the imputation that Simpson would condone violence as a comment, the Court of Appeal mischaracterized it as a statement of fact, to which, of course, different defences apply.  The reasons of the Court of Appeal are not altogether clear on this point.  Southin J.A., after quoting at length from various court decisions, concluded without further discussion:

[18] The appellants argue that even if the Court of Appeal decided the appeal on the basis that the imputation “would condone violence” is a fact requiring a demonstration of truth, and even if the court was correct to do so, they are, nevertheless, entitled to succeed on the basis of a “responsible journalism” defence, based on developments in other common law jurisdictions.  The appellants cite cases in Australia, New Zealand and the United Kingdom on this point.

[19]  I do not propose to deal with this argument at length because, in my view, this is a case of comment, not imputation of fact.  However, having regard to the full argument placed before us, I will briefly take note of the contending positions, which will have to await resolution for another appeal.

[20] In Reynolds v. Times Newspapers Ltd., [1999] 4 All E.R. 609, the House of Lords decided that the common law doctrine of qualified privilege has enough “elasticity” (p. 625) to accommodate freedom of the media in a manner consistent with the freedom of expression guaranteed by Article 10 of the European Convention for the Protection of Human Rights and Fundamental Freedoms.  “[T]here are occasions when the public interest requires that publication to the world at large should be privileged” (p. 617), depending on such matters as the seriousness of the allegation, the nature of the information  published, the extent to which the subject matter is a matter of public concern, the source of the information and the steps taken to verify its truth (p. 626).  Where it applies, the “public interest defence of responsible journalism” protects the journalist even where he or she got material facts wrong.  Later English authority has noted that the effect of the Reynolds test is to shift the qualified privilege from the “occasion” to the specific publication:  Loutchansky v. Times Newspapers Ltd. (Nos. 2-5), [2002] 2 W.L.R. 640, [2001] EWCA Civ 1805.  See also Bonnick v. Morris, [2003] 1 A.C. 300, [2002] UKPC 31; Jameel v. Wall Street Journal Europe SPRL, [2006] 4 All E.R. 1279, [2006] UKHL 44; Cusson v. Quan (2007), 231 O.A.C. 277, 2007 ONCA 771 (leave to appeal granted April 3, 2008, [2008] 1 S.C.R. xii; hearing pending).

[21] Reliance was also placed by the appellants on the decisions of the High Court of Australia which has accepted a more narrow “government or political matters” defence.  In Lange v. Australian Broadcasting Corp. (1997), 145 A.L.R. 96, at p. 117, it was held that “the reputations of those defamed by widespread publications will be adequately protected by requiring the publisher to prove reasonableness of conduct . . . [or] if the person defamed proves that the publication was actuated by common law malice”.

[22] As in England, the Australian  adaptation of qualified privilege would put the onus on the media to prove reasonableness of conduct rather than on the plaintiff to prove lack of due diligence or negligence.  Moreover, it is argued that this approach would effectively sideline the issue of malice because a finding that the media acted reasonably would potentially foreclose a concurrent finding of malice in respect of the same publication.

[23] Recent New Zealand authority approaches qualified privilege from a different perspective.  In Lange v. Atkinson, [2000] 3 N.Z.L.R. 385, the New Zealand Court of Appeal disagreed with both the House of Lords in Reynolds and the Australian High Court in Lange, that a “responsible journalism” or “reasonableness” requirement should be imposed as a condition precedent to an occasion being classified as one of “qualified privilege”:

[24] While the legal position in both Australia and New Zealand was influenced by statutory provisions that have no direct counterpart in Canada, the Canadian law of qualified privilege will necessarily evolve in ways that are consistent with Charter values.  At issue will be both the scope of the qualified privilege (Reynolds is broader) and whether the burden of proof of responsible journalism should lie on the defendant (Reynolds) or irresponsible journalism on the plaintiff (Lange v. Atkinson).

[25] Despite the variations of the “responsible journalism” defence urged in argument, the fact is that Mair does not claim to have acted reasonably in relation to the publication of facts subsequently shown to be false.  He says he editorialized on facts that were shown to be true.  Commentators are allowed broad latitude under the existing law of fair comment.  Despite some ambiguity about the basis of the majority decision of the B.C. Court of Appeal, as more fully discussed below, fair comment is the essence of his defence.

[26] The pleaded innuendo that Simpson was so “hostile toward gay people to the point that she would condone violence toward gay people” (trial reasons, at para. 19 (emphasis added; emphasis in original deleted)) is framed as an inference (“would condone violence”) from a factual premise, (i.e. was so “hostile toward gay people”).  In Ross v. New Brunswick Teachers’ Assn. (2001), 201 D.L.R. (4th) 75, 2001 NBCA 62, at para. 56, the New Brunswick Court of Appeal correctly took the view that “comment” includes a “deduction, inference, conclusion, criticism, judgment, remark or observation which is generally incapable of proof”.  Brown’s The Law of Defamation in Canada (2nd ed. (loose-leaf)) cites ample authority for the proposition that words that may appear to be statements of fact may, in pith and substance, be properly construed as comment.  This is particularly so in an editorial context where loose, figurative or hyperbolic language is used (Brown, vol. 4, at p. 27-317) in the context of political debate, commentary, media campaigns and public discourse.  See also, R. D. McConchie and D. A. Potts, Canadian Libel and Slander Actions (2004), at p. 340.

[27] The respondent on this appeal did not challenge the view that Mair’s imputation, that Simpson “would condone violence toward gay people”, was a comment  not an imputation of fact (Factum, at para. 40).  I agree that the “sting” of the libel was a comment and it would have been understood as such by Mair’s listeners.  “What is comment and what is fact must be determined from the perspective of a ‘reasonable viewer or reader’” (Ross, per Daigle C.J.N.B., at para. 62).  Mair was a radio personality with opinions on everything, not a reporter of the facts.  The applicable defence was fair comment.  On that point, I agree with the trial judge.

[28] For ease of reference, I repeat and endorse the formulation of the test for the fair comment defence set out by Dickson J., dissenting, in Cherneskey as follows:

[29] The Canadian Civil Liberties Association (“CCLA”) argues that to bring the common law of defamation into compliance with the Charter requires “a presumption in favour of expressive activity”, so that a plaintiff ought to bear the initial onus of establishing that the defendant’s defamatory comment (i) is not a comment at all but a fact which must be proven or otherwise justified, or (ii) is not a comment on a question of public interest.

[30] I do not think a shift of onus on these points is required.  In the first place, the Charter is about “expressive activity” but it also protects the dignity and worth of individuals, whose reputation may be their most valued asset.  Plaintiffs must prove defamation to get their case on its feet.  At that point, it is the media that seeks to excuse defamatory remarks on the basis that they are “comment” on a “matter of public interest”.  Ordinary principles of litigation put the burden of proof on the party making the assertion  (Ontario Equitable Life and Accident Insurance Co. v. Baker, 1926 CanLII 69 (SCC), [1926] S.C.R. 297).  In any event, the onus on these two issues is relatively easy to discharge.  The public interest is a broad concept.  The cases establish that the notion of “comment” is generously interpreted.  Putting the onus on the defendant in these respects is not too high a price to pay for a defamer to avoid legal liability for an allegation already found to have wronged the plaintiff’s reputation. 

[31] It is true that “[t]he comment must explicitly or implicitly indicate, at least in general terms, what are the facts on which the comment is being made”; Brown, vol. 2, p. 15-36, and Gatley on Libel and Slander (10th ed. 2004), at para. 12.12.  What is important is that the facts be sufficiently stated or otherwise be known to the listeners that listeners are able to make up their own minds on the merits of Mair’s editorial comment.  If the factual foundation is unstated or unknown, or turns out to be false, the fair comment defence is not available (Chicoutimi Pulp, at p. 194).

[32] On this point, as mentioned, Southin J.A. concluded that there could be no “consideration of fair comment because there is no evidentiary foundation for a finding that the appellant would condone violence” (para. 43).  Perhaps Southin J.A. meant that Mair’s comment was too remote from the facts to which his listeners understood the comment related, or perhaps she concluded that Mair’s factual premise that Kari Simpson was so deeply hostile to the gay community’s initiative was unproven.  Southin J.A. did not elaborate. 

[33] In Vander Zalm v. Times Publishers (1979), 1979 CanLII 2780 (BC SC), 96 D.L.R. (3d) 172 (B.C.S.C.), the trial court gave judgment for the plaintiff, but the Court of Appeal, in reversing ((1980), 1980 CanLII 389 (BC CA), 109 D.L.R. (3d) 531, at p. 536), observed that the trial judge held that the defence failed because the facts pleaded as the basis for the alleged comment “could not . . . fairly lead to the imputation arising from the cartoon” (emphasis added).  The Court of Appeal held that on making this finding the judge applied the wrong test; the proper test being whether the comment made by the cartoon met the test of fair comment, which the evidence indicated that it did.  The trial judge spoke of facts “fairly” giving rise to the comment, thereby introducing an unwelcome requirement of reasonableness and proportionality. Southin J.A. in the present case more prudently spoke of “no evidentiary foundation” (para. 43 (emphasis added)).

[34] I agree with Southin J.A. that a properly disclosed or sufficiently indicated (or so notorious as to be already understood by the audience) factual foundation is an important objective limit to the fair comment defence, but the general facts giving rise to the dispute between Mair and Simpson were well known to Mair’s listening audience, and were referred to in part in the editorial itself.  Simpson’s involvement in the Declaration of Family Values was familiar to Mair’s audience.  Her repeated invitations to her followers to pick up the phone and call talk shows and politicians assured her views a measure of notoriety (Barltrop v. Canadian Broadcasting Corp. (1978), 1978 CanLII 2145 (NS CA), 25 N.S.R. (2d) 637 (C.A.)).  The respondent has offered no persuasive reason to justify the Court of Appeal’s interference with the trial judge’s conclusion that

[35] The respondent Simpson relies on this Court’s judgment in Cherneskey, for the proposition, as stated by Ritchie J., at p. 1081, that

[36] Concern about the obvious anomalies in such a requirement has prompted the intervener CCLA to urge that “the honest belief requirement be eliminated” altogether (Factum, at para. 7), despite the description in Cherneskey, at p. 1082, of honesty of belief as the “cardinal test” of the defence of fair comment (Ritchie J. quoting Lord Denning in Slim v. Daily Telegraph Ltd., [1968] 1 All E.R. 497 (C.A.), at p. 503).  I do not think abolition of the requirement of honest belief, however formulated, would be “incremental”.  In R. v. Salituro, 1991 CanLII 17 (SCC), [1991] 3 S.C.R. 654, the Court said, at p. 670:

[37] The common law judges long ago decided that the gravamen of the defence of fair comment would not be the reasonableness or proportionality of the comment in relation to the facts (which would, of course, create stronger protection for the person defamed, but would depend in its application on the mental yardstick employed by a particular court) but whether the comment reflected honest belief.  The intervener Media Coalition would substitute for honest belief the requirement that the relationship of the comment to the facts merely “be one of relevancy” (transcript, at p. 54).  However, there is a world of difference between an attack made without honest belief and an attack whose relevance to the underlying facts may be disputed.  By way of illustration, reference may be made to a famous exchange in the 1954 U.S. Senate Army-McCarthy Hearings when Joseph Welch, counsel to the U.S. army, denounced the scandal mongering Senator for a smear of one of Welch’s associates:

[38] The relevancy of the comment to the facts has long been part (but only part) of the fair comment test as is shown in paragraph (b) in Dickson J.’s list:  see also McQuire v. Western Morning News Co., [1903] 2 K.B. 100 (C.A.), at p. 113.  As Gaynor J. put it in Howarth v. Barlow, 99 N.Y.S. 457 (App. Div. 1906), at p. 459:

[39] Of course it is true that the comment must have “a basis” in the facts, but a requirement that the comment be “supported by the facts”, read strictly, might be thought to set the bar so high as to create the potential for judicial censorship of public opinion. Even  the assessment of “relevance” has in the past misled courts into asking whether the facts “warranted” the comment, or whether the comment “fairly” arose out of the facts (Vander Zalm), or other such judgmental evaluations.  Insistence on a court’s view of reasonableness and proportionality was thought to represent too great a curb on free expression, but it was not too much to ask a defamer to profess an honest belief in his or her defamatory comment.  If the speaker, however misguided, spoke with integrity, the law would give effect to freedom of expression on matters of public interest.

[40] “Honest belief”, of course, requires the existence of a nexus or relationship between the comment and the underlying facts.  Dickson J. himself stated the test in Cherneskey as “could any man honestly express that opinion on the proved facts” (p. 1100 (emphasis added)).  His various characterizations of “any man” show the intended broadness of the test, i.e. “however prejudiced he may be, however exaggerated or obstinate his views” (p. 1103, citing Merivale v. Carson (1887), 20 Q.B.D. 275 (C.A.), at p. 281).  Dickson J. also agreed with the comment in an earlier case that the operative concept was “honest” rather than “‘fair’ lest some suggestion of reasonableness instead of honesty should be read in” (p. 1104).

[41] There is a further practical objection to the proposal of the CCLA and the Media Coalition to eliminate altogether the honest belief requirement.  By way of explanation to a jury of what is meant by the test of whether the comment is based on relevant or true facts, the court would have to warn the jury not to embark on a reasonableness inquiry.  An effective way of explaining to the jury how the necessary connection between the comment and the facts is to be established would be to tell them to ask themselves the question: Could any person honestly express that opinion on the proved facts?  We would therefore be back at the point of departure.

[42] Curiously, Cherneskey also involved a strange debate over the role and function of “honest belief”.  In that case, a couple of law students had written a provocative letter to the editor of the Saskatoon Star-Phoenix which was found to have libelled a city alderman as a racist.   The newspaper published the letter but its publisher testified at trial that he did not agree with its contents.  The letter writers were not called to testify.  In the absence of any evidence from anybody associated with its publication that the letter represented his, her or its “honest belief”, it was held in this Court by a 6-3 majority that the defence of fair comment was not available because the newspaper offered no proof that the defamatory opinions were “honestly held” by the actual writers.  Dickson J., dissenting, pointed out with justice that

[43] The Dickson J. test may be thought to marginalize the “honest belief” requirement, as it is possible to imagine most silly or ridiculous opinions finding a home somewhere in the minds of silly or ridiculous people.  However, his dissent was really driven by an appreciation that the originator and the publisher of a defamatory comment play different roles.  Nobody expects the newspaper publisher personally to have an honest belief in all of the contradictory opinions expressed on a “letters to the editor” page, much of it inspired by disagreement with something the newspaper itself has said in an editorial.  Dickson J. was responding to the need to protect free expression on matters of public interest in a democratic society.

[44] Nevertheless, it remains true that an effective way to establish that somebody could “honestly express that opinion on the proved facts” is to call the defamer (if available) to establish that he or she did indeed express an honest belief.  As the philosopher Bertrand Russell once observed with suitable gravity, the existence of a thing is absolute proof of its possibility.

[45] Other complexities of the “honest belief” requirement emerge in the present appeal.  Mair’s editorial about Kari Simpson clearly defamed her.  Attributing to Simpson bigotry of a type associated with Hitler and a couple of notoriously racist Governors in the Southern United States at the height of the desegregation crisis would, I think, tend to lower her in the opinion of right-thinking people (some might call it a “smear”), and the appellants were right to concede the point in this Court.  That being the case, it was entirely proper to have Mair go into the witness box to affirm his honest belief in what he had said about her.  Yet that was not the end of the issue.

[46]  The trial judge found a difference between what Mair subjectively intended to say (Simpson is a bigot) and objectively what he is taken to have said (Simpson would condone violence).  The gap between the intended meaning and what the court determined to be the effect Mair’s words conveyed to reasonable members of the audience has important implications.  On the test of subjective honest belief applied by the Court of Appeal, Mair would be robbed of his defence, even though on the public record someone could honestly express the view imputed to him that Simpson “would condone violence toward gay people”, and thus the objective honest belief test would be met.

[47] It may be noted that such circumstances are not uncommon.  In much modern media, personalities such as Rafe Mair are as much entertainers as journalists.  The media regularly match up assailants who attack each other on a set topic.  The audience understands that the combatants, like lawyers or a devil’s advocate, are arguing a brief.  What is important in such a debate on matters of public interest is that all sides of an issue are forcefully presented, although the limitation that the opinions must be ones that could be “honestly express[ed] . . . on the proved facts” provides some boundary to the extent to which private reputations can be trashed in public discourse.

[48] Of course the law must accommodate commentators such as the satirist or the cartoonist who seizes on a point of view, which may be quite peripheral to the public debate, and blows it into an outlandish caricature for public edification or merriment.  Their function is not so much to advance public debate as it is to exercise a democratic right to poke fun at those who huff and puff in the public arena.  This is well understood by the public to be their function.   The key point is that the nature of the forum or the mode of expression is such that the audience can reasonably be expected to understand that, on the basis of the facts as stated or sufficiently indicated to them, or so generally notorious as to be understood by them, the comment is made tongue-in-cheek so as to lead them to discount its “sting” accordingly.  In Cherneskey, Dickson J. emphasized that “the objective limits of fairness [i.e. fair comment] are very wide” (p. 1109).  The accuracy of this observation is born out in Vander Zalm where the defendant, a political cartoonist, had drawn a cartoon of the plaintiff, a Cabinet Minister then holding the office of Minister of Human Resources in British Columbia, which appeared on the editorial page of the Victoria Times.  The cartoon, which was uncaptioned, depicted the plaintiff William Vander Zalm with an evil grin on his face, seated at a table, and engaged in plucking the wings from a fly.  Other flies, without wings, were shown moving around on the table.  On the plaintiff’s lapel were inscribed the words “HUMAN RESOURCES”.  The Court of Appeal found that both the cartoonist and the Victoria Times publisher had satisfied the “honest belief” test.

[49] The test represents a balance between free expression on matters of public interest and the appropriate protection of reputation against damage that exceeds what is required to fulfill free expression requirements.  The objective test is now widely used in common law jurisdictions as the “honest belief” component of fair comment, including the United Kingdom: Telnikoff v. Matusevitch, [1991] 3 W.L.R. 952 (H.L.), quoting with approval Dickson J.’s dissent, at p. 959.  In Australia, the High Court recently affirmed a similar approach; see the observation of Gleeson C.J.:

[50] Admittedly, the “objective” test is not a high threshold for the defendants to meet, but nor is it in the public interest to deny the defence to a piece of devil’s advocacy that the writer may have doubts about (but is quite capable of honest belief) which contributes to the debate on a matter of public interest.

[51] Of course, even the latitude allowed by the “objective” honest belief test may be exceeded.  “Comment must be relevant to the facts to which it is addressed.  It cannot be used as a cloak for mere invective”; Reynolds, at p. 615. 

[52] As usual, the debate is about onus:

[53] Some commentators have suggested that proof of honest belief negates the possibility of a finding of malice.  This is not necessarily true.  If a defendant relies on objective honest belief the defence can still be defeated by proof that subjective malice was the dominant motive of the particular comment.

[54] In a lengthy and careful judgment, the trial judge dealt with the issues in an appropriate sequence:

[55] At common law, the judge is to make a legal determination “whether there is a case or an issue to go to the jury” by deciding if the words are “capable of being a statement of a fact or facts”.  It is then “for the jury to decide as to what is fact and what is comment” (Court of Appeal reasons, at para. 42, citing Jones v. Skelton, [1963] 1 W.L.R. 1362 (P.C.), at pp. 1379-80).  As pointed out by the intervener British Columbia Civil Liberties Association, “the judge’s role in that test is a response to concerns about freedom of expression” (Factum, at para. 34).

[56] The “full context” is important.  While argument in this Court largely focussed on the innuendo that Simpson “would condone violence toward gay people”, the broader analysis of the trial judge left no doubt about her view of the “[u]nwholesome virulence” (para. 78) of the editorial taken as a whole.  The appellants argue that in assessing meaning, the Court is to consider what reasonable and right-thinking listeners would understand.  The Court is to avoid putting the worst possible meaning on the words:  Color Your World Corp. v. Canadian Broadcasting Corp. (1998), 1998 CanLII 1983 (ON CA), 38 O.R. (3d) 97 (C.A.), at pp. 106-7, and Scott v. Fulton (2000), 73 B.C.L.R. (3d) 392, 2000 BCCA 124,  at paras. 13-15.  However, both courts below found that Mair’s editorial about Simpson was defamatory.  This is a mixed question of law and fact.  There is no reason to interfere with that conclusion.  It is plainly correct.

[57] The public debate about the inclusion in schools of educational material on homosexuality clearly engages the public interest.  As the Ontario Court of Appeal recognized over a century ago in words that apply equally to the case on appeal, “[w]hoever seeks notoriety, or invites public attention, is said to challenge public criticism; and [s]he cannot resort to the law courts, if that criticism be less favorable than [s]he anticipated” (Macdonell v. Robinson (1885), 12 O.A.R. 270, at p. 272).

[58] The trial judge, after reviewing the editorial as a whole, concluded:

[59] The law requires the comment be based on a sufficient substratum of facts to anchor the defamatory comment:  Vander Zalm, Court of Appeal reasons, at p. 536; and Ross, at paras. 73, 78 and 83.  This is another mechanism to prevent tenuous facts serving as a springboard for defamatory comment, which, in my view, would be the danger of the “relevance” test proposed by the CCLA.  Simpson does not dispute the contents or tone of her speeches in the court record.  In my view, as in the view of the trial judge, the factual substratum exists.

[60] Mair testified as to his subjective honest belief in what he intended to say, but acknowledged that he did not honestly believe that Simpson would condone violence.  Notwithstanding the absence of a subjective honest belief that Simpson would condone violence, Mair and WIC Radio, like the newspaper publisher in Cherneskey, were entitled to rely on the objective test, i.e. could any person honestly have expressed the innuendo that Simpson would condone violence toward gay people on the proven facts?  As mentioned earlier, Simpson’s public speeches were full of references to “war . . . [where] the spoils turn out to be our children”, “militant homosexuals”, “[w]ar, you shoot, they shoot” and so on.  Simpson’s use of violent images could support an honest belief on the part of at least some of her listeners that she “would condone violence toward gay people”, even though Mair denied that he intended to impute any such meaning.

[61] The respondent, Simpson, emphasizes the distinction taken by the B.C. Court of Appeal between what Mair thought he was saying (that Simpson was a non-violent person whose loose words inadvertently opened the door to less peaceable individuals to resort to violence) and the specific defamatory imputation that was found to have arisen from the words Mair used (i.e. that Simpson “would condone violence toward gay people”).   It makes little sense to deny the defence of fair comment to a speaker whose opinion has been misunderstood, even if carelessness in the use of words is the source of the misunderstanding.  The Court of Appeal framed the issue as follows:  “Is it the law that to succeed in the defence of fair comment, the defendant must honestly believe in the imputation,  . . . or need he only have an honest belief in what he himself subjectively intended by the words which he used?” (para. 37).  In its view, the former requirement applied.  I do not think the “either/or” question of the Court of Appeal exhausts the possibilities.  On the contrary, I think the proper question is whether the defamatory imputation that Kari Simpson “would condone violence toward gay people” is an opinion that could be held by an honest person in the circumstances.

[62] The trial judge concluded that Mair honestly believed what he thought he had said:

[63] The defence is defeated if the commentary was actuated by malice in the sense of improper motive, proof of which lay on the plaintiff.  Simpson does not appeal against the trial judge’s conclusion that Mair’s fair comment defence was not vitiated by malice.

[64] Applying the elements of the fair comment defence set out above, I conclude that the trial judge was correct to allow the defence.

[65] I would allow the appeal, set aside the judgment of the Court of Appeal and restore the trial judgment dismissing the action, with costs throughout.

[66] LeBel J. — Although I agree with Binnie J. that the appellants are not liable in defamation, I arrive at this conclusion in a different way.  In particular, I am not convinced that Mr. Mair’s comments were prima facie defamatory.  Further, I disagree that the fair comment defence should include an element of honest belief.  Rather, to establish fair comment, the defendant should simply have to show that the impugned words constituted comment, that they had a basis in fact and that they concerned a matter of public interest. I accept the facts as they are set out by the majority, and I agree that this case is one of comment rather than of fact. I will therefore proceed directly to an analysis of the defamatory nature of the comment and the requirement of honest belief.  I will conclude with a brief review of the existing test for malice, which I endorse and apply.

[67] The issue of whether Mair’s comments were in fact prima facie defamatory was not raised on appeal to this Court, and I would therefore not interfere with the trial judge’s finding in this regard.  However, I disagree with Binnie J.’s observations that the trial judge’s conclusion on this point was “plainly correct” (para. 56) and that the editorial “clearly defamed” Ms. Simpson (para. 45).  Some discussion on this issue is warranted.  Although defamation is not easily defined, one generally accepted test is the one from Salmond on the Law of Torts (17th ed. 1977), at pp. 139-40, which is based on the test proposed by Lord Atkin in Sim v. Stretch (1936), 52 T.L.R. 669 (H.L.), at p. 671, and was approved by the B.C. Court of Appeal in Vander Zalm v. Times Publishers (1980), 1980 CanLII 389 (BC CA), 109 D.L.R. (3d) 531, at p. 535:

[68] This test is often construed as setting a low threshold for establishing prima facie defamation.  Gatley on Libel and Slander (10th ed. 2004) (“Gatley”), notes that “it may well be the case that the common law takes a rather generous line on what lowers a person in the estimation of others” (p. 18, footnote 32).  Dickson J. made a similar point in Cherneskey v. Armadale Publishers Ltd., 1978 CanLII 20 (SCC), [1979] 1 S.C.R. 1067, in referring to the “low level of the threshold which a statement must pass in order to be defamatory” (p. 1095).

[69] The case law generally bears these opinions out.  However, courts should not be too quick to find defamatory meaning — particularly where expressions of opinion are concerned.  The test is not whether the words impute negative qualities to the plaintiff, but whether, in the factual circumstances of the case, the public would think less of the plaintiff as a result of the comment.  Relevant factors to be considered in assessing whether a statement is defamatory include: whether the impugned speech is a statement of opinion rather than of fact; how much is publicly known about the plaintiff; the nature of the audience; and the context of the comment.  I will demonstrate, based on the first two of these factors in particular, that Mair’s comments would likely not have led “right‑thinking” members of the public to think less of Simpson.

[70] It should go without saying that people evaluate statements of opinion differently than statements of fact.  In discussing what constitutes a statement of fact as opposed to comment, Lord Herschell noted that 

[71] Although distinguishing facts from comment may sometimes be difficult, a comment is by its subjective nature generally less capable of damaging someone’s reputation than an objective statement of fact, because the public is much more likely to be influenced in its belief by a statement of fact than by a comment.  I therefore agree with the following observation by R. E. Brown:

[72] There is no doubt that a comment may be defamatory.  It must simply be borne in mind that just because someone expresses an opinion does not mean that it will be believed and therefore affect its subject’s reputation.

[73] This is all the more true in an age when the public is exposed to an astounding quantity and variety of commentaries on issues of public interest, ranging from political debate in the House of Commons, to newspaper editorials, to comedians’ satire, to a high school student’s blog.  It would quite simply be wrong to assume that the public always takes statements of opinion at face value.  Rather, members of the public must be presumed to evaluate comments in accordance with their own knowledge and opinions about the speaker and the subject of the comments.

[74] Members of the public will generally have a more solid basis on which to evaluate a comment about a public figure than one about someone who is unknown.  Thus, although public figures are certainly more open to criticism than those who avoid the public eye, this does not mean that their reputations are necessarily more vulnerable.  In fact, public figures may have greater opportunity to influence their own reputations for the better.

[75] People who voluntarily take part in debates on matters of public interest must expect a reaction from the public.  Indeed, public response will often be one of the goals of self‑expression.  In the context of such debates (and at the risk of mixing metaphors), public figures are expected to have a thick skin and not to be too quick to cry foul when the discussion becomes heated.  This is not to say that harm to one’s reputation is the necessary price of being a public figure.  Rather, it means that what may harm a private individual’s reputation may not damage that of a figure about whom more is known and who may have had ample opportunity to express his or her own contrary views.

[76] Turning to the facts of this case, I agree that the impugned statement constituted comment rather than fact.  As a result, Mair’s audience would necessarily treat it differently than a statement of fact.  In addition, both Mair and Simpson were public figures involved in an ongoing public debate on the issue of the introduction of materials dealing with gay issues in the classroom.  That debate would have informed public opinion.  Even those not familiar with the issue would have understood the comment in the context of this debate because Mair made reference to it in the impugned editorial.  Further, Mair’s “sizeable following” (trial reasons (2004), 31 B.C.L.R. (4th) 285, 2004 BCSC 754, at para. 5) would have understood his comments in light of his well‑known style, which involves strong opinions sometimes conveyed with colourful and provocative language.

[77] Thus, although associating the respondent’s bigotry with Hitler would clearly be defamatory if taken at face value, the test for defamation is a contextual one and relates to what people would think in the circumstances of publication of the comments (Gatley, at pp. 108-10).  In the context of this particular debate, I do not believe that Mair’s audience would have taken his comments at face value.  To paraphrase Professor Brown, Simpson’s reputation was not adversely affected, either because Mair’s opinions were consistent with the facts and approximated the subjective opinions of his listeners, or because Mair’s opinion was not consistent with the facts and the subjective opinions of his listeners and only reflected badly on Mair himself.

[78] Triers of fact should be mindful of ensuring that the plaintiff’s reputation is actually threatened by the impugned statements before turning to the available defences.  I do not mean to imply that damage to reputation must be proved, since actual harm to reputation is not required to establish defamation.  However, before a prima facie case can be made out, there must be a realistic threat that the statement, in its full context, would reduce a reasonable person’s opinion of the plaintiff.  In my view, Mair’s comments posed no realistic threat to Kari Simpson’s reputation, and I would therefore not have found them to be prima facie defamatory.

[79] If, however, the traditional test were so narrow as to catch the comments made in Mair’s editorial, then in my opinion, it would be unduly restrictive and would need to be expanded to better reflect the values of modern Canadian society.  The law of defamation — whose purpose is to protect reputation — exists as a limitation on freedom of expression, which is protected by s. 2(b) of the Canadian Charter of Rights and Freedoms.  Reputation is an important element of human dignity and must be protected.  However, even if a fair comment defence is available, it cannot be consistent with the Charter value of freedom of expression to treat spirited statements of opinion in a debate on matters of public interest as being prima facie defamatory.

[80] As I mentioned above, since the issue was not raised before us, I would not interfere with the lower courts’ finding that the editorial was defamatory.  Further, since I agree that the defence of fair comment applies, it is not necessary to the proper disposition of this appeal to resolve this issue.  However, I consider it important to state my discomfort with the majority’s assessment that the editorial “clearly defamed” Simpson (para. 45).

[81] The majority would retain a requirement of honest belief in the fair comment defence primarily on the basis: (a) that to eliminate it would constitute more than an incremental change to the law (para. 36); and (b) that it provides additional and appropriate protection to reputation beyond that afforded by the malice and “based on fact” inquiries (see para. 49).  I disagree with both these propositions.

[82] The first proposition is based on the fact that honest belief is the “cardinal test” of fair comment in Canada (para. 36).  This may be so, but a closer examination of fair comment reveals that it is subjective rather than objective honest belief that is considered the cardinal test of fair comment.  Further, in Commonwealth countries such as the U.K., Australia and New Zealand, the requirement of a subjective honest belief exists only as an aspect of the malice inquiry in respect of which the burden of proof is on the plaintiff, not as an actual element of the defence.  As for objective honest belief, although Commonwealth countries such as the U.K. and Australia retain an objective element in the defence, they have recognized that in the context of fair comment, there is no longer any justification for judging the reasonableness of the content of opinions.  An objective honest belief requirement conflicts with the principle that even unreasonable comments should be protected in a democratic society.  Thus, the common law in the U.K., Australia and Canada has gradually moved away from assessing comment in terms of reasonableness. 

[83] I also disagree with the second proposition, which assumes that an honest belief requirement provides appropriate protection to reputation beyond that provided by other elements of the fair comment defence (including malice, which is, strictly speaking, not an element of the defence).  I will begin by reviewing the U.K. decisions that established subjective honest belief as the “cardinal test” of fair comment.  I will then discuss the history and the current role of the objective honest belief requirement.  I will conclude by demonstrating that, in any event, an objective honest belief requirement provides little or no additional protection to reputation.

[84] Although it is not entirely clear from Slim who has the burden of proving or disproving subjective honest belief, it is settled law in the U.K. that this burden falls to the plaintiff once the elements of fair comment are proved:

[85] Thus, the “cardinal test” of fair comment — at least in the U.K. — is subjective rather than objective honest belief, and it is the plaintiff who bears the burden of proof in the context of the malice inquiry. That being said, my colleague correctly notes that the requirement of an objective honest belief persists in the defence of fair comment in some common law countries, and he concludes that to excise it from Canadian law would constitute more than an incremental change.  I disagree.  Although this requirement does continue to exist in some common law countries, its influence and utility have been waning such that, in my opinion, it no longer offers anything of value in the exercise of balancing the right to comment fairly on matters of public opinion against the right to reputation.  Doing away with it may constitute a significant change to the form of the test for fair comment, but the effect of doing so would not be to modify the scope of the defence in any significant way.  The advantage of eliminating the objective honest belief requirement, on the other hand, is that this would constitute formal recognition of what some common law courts have been saying for decades: that it is no longer justifiable, for purposes of the fair comment defence, to judge a person’s opinions on an objective basis other than to require that they have some basis in fact.

[86] In the U.K., fair comment is a “two stage” issue.  At the first stage, the defendant establishes that the words are objectively capable of constituting comment.  At the second stage, the plaintiff may attempt to prove malice, which will defeat the protection of the defence (Gatley, at p. 311).  The objective stage incorporates both objective honest belief and relevance to the underlying facts.  At one time, fair comment was limited to what was objectively “fair, reasonable and temperate” (Soane v. Knight (1827), M. & M. 74, 173 E.R. 1086, at p. 1086, cited in P. Mitchell, The Making of the Modern Law of Defamation (2005), at p. 174).  Eventually, however, this view was held to be overly restrictive of speech on matters of public interest.  In Merivale v. Carson (1887), 20 Q.B.D. 275 (C.A.), Lord Esher held that the correct test was whether any “fair man, however prejudiced” could hold the impugned opinion (p. 280).  Later, even the word “fair”, which gave the defence of fair comment its name, was replaced by the word “honest” to eliminate any residual requirement of reasonableness (see Mitchell, at p. 181.  See also Dickson J. dissenting in Cherneskey, at p. 1104.) The objective reasonableness of the opinion was held to be of no relevance in limiting fair comment.  As early as 1863, the English courts held that the defendant’s opinion need not be reasonable.  Rather, in addition to subjective honest belief, a jury must simply find that the defendant’s “belief was not without foundation” (Campbell v. Spottiswoode (1863), 3 B. & S. 769, 122 E.R. 288, at p. 290).

[87] This history is reflected in the words of Lord Nicholls in Reynolds v. Times Newspapers Ltd., [1999] 4 All E.R. 609, at p. 615:

[88] Some Canadian courts have cited this passage from Reynolds with approval.  For example, in Ross v. New Brunswick Teachers’ Assn. (2001), 201 D.L.R. (4th) 75, 2001 NBCA 62, the New Brunswick Court of Appeal quoted it and then went on to conclude:

[89] These passages confirm that little is left of the objective “fairness” requirement other than that a comment must be “relevant to the facts to which it is addressed”.  However, despite the waning role of the objective honest belief requirement, it seems that British law, like Canadian and Australian law, still requires a trier of fact to ask: “Would any [honest] man, however prejudiced he might be, or however exaggerated or obstinate his views, have written this criticism?” (Gatley, at p. 307).  Alternately, in the words of Dickson J. in Cherneskey, at p. 1100, as endorsed by the majority in this case, “could any man honestly express that opinion on the proved facts?”  What, then, does it mean that a person could honestly express that opinion on the proved facts?

[90] If the speaker’s prejudices or inclination toward exaggeration and obstinacy are irrelevant, it would similarly be irrelevant to consider the objective reasonableness of the comment aside from the requirement that it have a basis in fact.  This is certainly consistent with the shift away from the idea that fairness amounts to reasonableness.  For example, in Cherneskey, Dickson J. substituted the word “honest” for “fair” in the objective test “lest some suggestion of reasonableness instead of honesty should be read in” (p. 1104).  In this context, then, “honest belief” must refer either to whether someone could express the comment with an honest motive or to whether someone could believe the comment if he or she were being honest with him or herself.  In the former case, since a dishonest motive for publication can defeat a fair comment defence in the malice inquiry, it would seem redundant to inquire into the defendant’s motives at the honest belief stage of the defence.  In the latter case, the “honest belief” label would also be redundant, since what one does not honestly believe, one does not believe at all.  This relates to subjective honest belief, albeit to what someone could subjectively believe rather than to what the defendant subjectively believed.  I take this to be the sense in which the majority, and the courts generally, use the expression “honest belief”.

[91] The requirement that someone be capable of believing the comment in light of the facts does not do away with the problem of assessing the objective reasonableness of the comment.  It is not clear how — other than by requiring a simple basis in fact — the limits to what someone could subjectively believe can be determined without resorting to objective reasonableness.  Admittedly, the threshold for establishing a factual basis for a comment remains low.  However, the threshold for establishing that someone could believe the comment on the basis of the relevant facts is also low.  The tests involve similar, if not identical, questions.  Where a comment is objectively incapable of belief, this will presumably be because it does not have a basis in fact.  If there is any difference between what is capable of belief and what is based in fact, it must relate to what is reasonable for a person to believe, given certain facts.  As I mentioned above, common law courts, including those in Canada, have long rejected an approach that involves judging the objective reasonableness of a comment. 

[92] The honest belief inquiry adopted by the majority, which is clearly not intended to assess the objective reasonableness of the comment, must therefore be an inquiry into whether the comment has a basis in fact.  Binnie J. acknowledges as much by noting that a jury could be instructed to determine whether comment has a basis in fact by asking whether a person could honestly express the opinion on the proved facts (para. 41).  I agree that this is a useful jury instruction regarding the “basis in fact” element.  But what, then, is the additional benefit of a distinct element of honest belief?  The defence of fair comment would include two elements — basis in fact and honest belief — that address exactly the same issue.

[93] Since the elements address the same issue, honest belief provides no additional protection for reputation.  For example, an honest belief test would not protect against a bad faith attack made without honest belief, such as Senator McCarthy’s smear campaign of the 1950s.  Such attacks would pass the majority’s honest belief test so long as they constituted comment that someone could believe.  In the context of McCarthyism, someone could very well have believed defamatory allegations relating to communism.  If no one could believe the allegations, it must be because they have no basis in fact, and the defamer could not rely on the fair comment defence.  Further, even without an honest belief element, a plaintiff who is the victim of a bad faith attack may prevail by demonstrating malice.

[94] Thus, the only justifiable remnant of the former requirement that a fair‑minded person be capable of holding the opinion is that the comment must be based on known facts.  This is implied in the above-quoted observations of Lord Nicholls from Reynolds.  In Canada, fair comment includes a based on true facts element that is independent of concerns about whether an honest person could hold the opinion.  Thus, I see no reason to retain an objective honest belief element.  Eliminating that element is an incremental change.  This Court has the power, and indeed the responsibility, to make such changes when the common law falls out of step with its underlying principles and with modern values, and when a test has proven to be unworkable or to serve no useful purpose.

[95] Even though, in principle, I do not consider it appropriate to exclude from fair comment that which no honest person could believe, it is nevertheless true that in practice, such speech will rarely, if ever, be protected as fair comment even without an honest belief requirement.  This is because in defamation law, reputation is already protected in three ways from comments that are not “fair” in the sense of being objectively capable of belief on the relevant facts.

[96] First, to the extent that, as discussed above, honest belief and basis in fact amount to the same inquiry, whatever has no basis in fact is not capable of belief, and vice versa.  However, even if there were some difference between the two concepts, two other aspects of fair comment protect reputation against objectively unbelievable comment.

[97] For a comment to be prima facie defamatory, there must be a possibility that its audience will believe it; otherwise, it cannot harm the plaintiff’s reputation.  The audience is presumed to consist of “ordinary, reasonable, fair-minded” people (Charleston v. News Group Newspapers Ltd., [1995] 2 W.L.R. 450 (H.L.), at p. 454).  The test is not whether any person, no matter how unreasonable or unfair, could think less of the plaintiff because of the comment.  Therefore, the more disconnected the comment is from its underlying facts, the less likely it is to be defamatory.  It follows that comments that are not capable of objective honest belief, given the relevant facts, will rarely be defamatory.  For example, in a modern developed society, an imputation that someone practises witchcraft would not be defamatory, because it would not be believed and therefore would not harm the plaintiff’s reputation (Loukas v. Young, [1968] 3 N.S.W.R. 549 (S.C.)).  Thus, the test for prima facie defamation, if correctly applied, will exclude from protection many situations involving comments that are not capable of objective honest belief.

[98] Further, it is difficult to imagine any situation in which a comment with a basis in true facts that would be incapable of belief could be made with a motive that is not predominantly malicious.  By definition, the publisher of the comment would not believe the opinions he or she expressed.  Why would someone publish a comment that he or she does not believe and that no one else could honestly believe, if not out of malice?  The dominant motive could not be for the sake of argument, since no one could believe the comment.  Nor could it be a desire to report on issues of public interest, because here again, no one could believe the comment. 

[99] I am therefore of the opinion that the only additional protection for reputation afforded by a requirement of objective honest belief is an inappropriate one, in that it places a reasonableness restriction on the opinions a person may legitimately express.  The common law courts in this country and in the U.K. have long been uncomfortable with the idea of limiting fair comment to what is reasonable, even in the broadest sense.  The time has come to formally acknowledge that such a reasonableness requirement has outlived its purpose and that, in any event, in its present broad form, it provides little or no protection for reputation.  In my opinion, therefore, the defence of fair comment should simply require the defendant to prove (a) that the statement constituted comment, (b) that it had a basis in true facts and (c) that it concerned a matter of public interest.  On the facts of this case, there is no dispute that each of these requirements is met.

[100]      If the defendant is successful in establishing the elements of the fair comment defence, the inquiry may turn to malice, which the plaintiff must prove if he or she alleges it.  I see no reason to alter the existing burden of proof.  To require the defendant to prove a lack of malice would amount to presuming malice.  A society that seeks to promote healthy debate should require evidence of a malicious motive before restricting the expression of opinions based on true facts that concern matters of public interest.  It would protect spirited — but not mean-spirited — speech. Proof of malice may be intrinsic or extrinsic: that is, it may be drawn from the language of the assertion itself or from the circumstances surrounding the publication of the comment.  It may involve inferences and evidentiary presumptions.

[101]      I also see no reason to alter the nature of the malice inquiry.  In Cherneskey, Dickson J. described malice as follows:

[102]      I adopt this definition, although I wish to emphasize that while proof that the comment is not the honest expression of the publisher’s real opinion may be evidence of malice, it is not determinative.  Indeed, there may be non‑malicious and valid reasons for publishing views one does not personally hold. 

[103]      I would not adopt the British malice test (reproduced here for convenience), which is based on honesty of belief rather than the motive for publication.

[104]      In the U.K., the motive for publication appears to be relevant only with respect to secondary publishers: malice will not be found where someone publishes the opinions of others without malicious intent (Gatley, at pp. 310-11).  However, there are good reasons for maintaining a malice test that withholds protection from a comment expressed with a predominantly malicious motive.  As Greer L.J. noted in Watt v. Longsdon, [1930] 1 K.B. 130 (C.A.), at pp. 154-55, quoted with approval by the B.C. Court of Appeal in Christie v. Westcom Radio Group Ltd. (1990), 1990 CanLII 776 (BC CA), 75 D.L.R. (4th) 546, at p. 554, leave to appeal refused, [1991] 1 S.C.R. vii:

[105]      Similarly, an Australian court has noted that “personal animosity may perfectly well consort with sincerity to produce a comment which is harmful and unfair” (Renouf v. Federal Capital Press of Australia Pty. Ltd. (1977), 17 A.C.T.R. 35 (S.C.), at p. 54).  Australia maintains essentially the same test of malice as exists in Canada: whether malice was the dominant motive for publication (see M. Gillooly, The Law of Defamation in Australia and New Zealand (1998), at pp. 132-33).

[106]      The requirement that malice be the dominant motive for expressing an opinion in order to defeat fair comment helps maintain a proper balance between protecting freedom of expression and reputation.  Arguments between ideologically-opposed participants in a public debate often breed bitterness, but such debate remains valuable and worthy of protection in a democratic society.  However, while it is not appropriate to judge the objective fairness of an opinion, the protection of reputation may justify judging the motive for expressing it.  After all, the purpose of the fair comment defence is to protect and encourage free debate on issues of public importance.  Opinions published with the primary intention of injuring another person (for example), rather than furthering public debate, are sufficiently far removed from the type of speech the defence was intended to protect that they may justifiably be excluded from the scope of its protection.

[107]      The trial judge concluded that malice was not the dominant motivation for publication and, for the above reasons, I would allow the appeal and dismiss the action with costs throughout.

[108]      Rothstein J. — I have had the benefit of reading the reasons of my colleagues Binnie J. and LeBel J.  I agree with Binnie J. that the statements in question were defamatory but that the defence of fair comment applies.

[109]      However, I disagree that to satisfy the fair comment defence, there is a requirement to prove objective honest belief.  On this issue, I am in agreement with the reasons and analysis of LeBel J.  The defence of fair comment should only require the defendant to prove (a) that the statement constituted comment, (b) that it had a basis in true facts and (c) that it concerned a matter of public interest; and these requirements were met in this case.  Although the issue of malice is not before the Court on this appeal, I also agree with LeBel J.’s discussion in respect of the element of malice.

[110]      If objective honest belief means the honest belief of anyone, no matter how “prejudiced . . . exaggerated or obstinate” in his or her views, I cannot think of an example in which the test of objective honest belief could not be met once it is demonstrated that the comment has a basis in true facts.  In my respectful view, the test of objective honest belief adds only an unnecessary complexity to the analysis of fair comment.

[111]      I agree with Binnie J. that the trial judge’s conclusion that the fair comment defence was not vitiated by malice was not appealed. 

[112]      Like both Binnie and LeBel JJ., I would allow the appeal with costs throughout and dismiss the action.