McGill Law Journal ~ Revue de droit de McGill
THE MEDIUM IS NOT THE MESSAGE: RECONCILING
REPUTATION AND FREE EXPRESSION IN CASES OF
INTERNET DEFAMATION
Robert Danay*
principles,
traditional
defamation
In this paper the author critiques the approach to
defamation over the Internet taken to date by the
law courts. In the emerging
Canadian common
jurisprudence, the courts have relied upon untenably
broad generalizations about Internet
technology,
repeatedly equating it with traditional broadcast media
and expressing grave concerns about the corresponding
threat to reputation posed by online defamation. This
has led the courts to hold that when defamatory words
are transmitted using the Internet, this will vitiate the
availability of any qualified privilege that would
otherwise have immunized the defendant from liability
under
and
substantially increase any resulting award of damages.
The author argues that this approach results in a
failure to strike the appropriate balance between free
expression and the protection of reputation. The
jurisprudence can also be seen as a product of a long-
standing and unfortunate analytical tendency
in
defamation
lawprimarily apparent through the
libel/slander distinctionwhereby common law courts
legal consequences to
attach extremely divergent
impugned statements based on
indefensibly broad
generalizations about the degree of danger to personal
reputation posed by the medium
in which the
statement was communicated. Drawing inspiration
from a comparison to defamation under the civil law of
Quebec, the author proposes a new approach that
eschews reliance upon unhelpful analogies and
generalizations about particular media including the
Internet, and involves the examination of impugned
statements on a case-by-case basis, paying careful
attention to the context in which these were actually
made.
Ils
lInternet.
immunis
reprises aux mdias
Dans cet essai, lauteur critique lapproche
adopte par les tribunaux de common law canadiens
sur la question de la diffamation sur Internet. Dans la
jurisprudence mergeante, les tribunaux se sont bass
sur des gnralisations indfendables quant la
lassimilent de
technologie de
nombreuses
lectroniques
traditionnels et expriment de graves proccupations
quant la menace correspondante que pose la
diffamation en ligne pour la rputation. Cette approche
a men les tribunaux statuer que lorsque des mots
diffamatoires sont transmis sur Internet, les privilges
qualifis qui auraient autrement
le
dfendeur contre toute responsabilit, suivant les
principes de la diffamation traditionnelle, sont vicis.
Le montant des dommages-intrts accords augmente
aussi de faon substantielle.
Lauteur soutient que cette approche ne permet pas
dtablir lquilibre appropri entre la libre expression et
la protection de la rputation. La jurisprudence peut aussi
tre vue comme le produit dune fcheuse tendance
analytique de longue date en matire de diffamation,
tendance qui ressort surtout dans la distinction entre
diffamation orale et crite. Suivant cette tendance, les
tribunaux de common law attachent des consquences
juridiques extrmement divergentes des dclarations en
litige, selon quelles soient orales ou crites. Ces
consquences sont bases sur des gnralisations larges et
indfendables quant au degr de menace pour la
rputation personnelle que pose le mdium par lequel la
dclaration a t communique. En sinspirant dune
comparaison avec la diffamation en droit civil qubcois,
lauteur propose une nouvelle approche qui vite les
analogies et les gnralisations peu utiles au sujet dun
mdia particulier, dont Internet. Lapproche propose
implique un examen au cas par cas des dclarations
contestes, prtant une attention particulire au contexte
dans lequel elles ont t nonces.
* BSc (Toronto), LLB (Osgoode Hall), BCL (Oxon), Member of the Ontario and British Co-
lumbia bars, litigator with the Canadian Department of Justice. The opinions expressed
in this paper are the personal opinions of the author and do not necessarily reflect those
of the Department of Justice or the Government of Canada.
Citation: (2010) 56:1 McGill LJ 1 ~ Rfrence : (2010) 56 : 1 RD McGill 1
Robert Danay 2010
2 (2010) 56:1 MCGILL LAW JOURNAL ~ REVUE DE DROIT DE MCGILL
Introduction
I.
The Canadian Common Law Approach to Cyber-Libel
A. Defamation over the Internet and the Qualified Privilege
Defence
B. Defamation over the Internet and the Calculation of
Damages
II.
Why This Matters: Constitutional and Normative Implications
III.
Contextualizing the Emerging Cyber-Libel Jurisprudence
IV.
A More Civil Approach: Defamation Law in Quebec
V.
A Prescription for Reform: A Move from Caricature to Context
A. Qualified Privilege: A Caseby-Case Assessment
B. Quantification of Damages
C. Summary
Conclusion: Grant v. Torstar Corp.A Potential Turning Point?
3
6
6
10
14
18
28
33
33
34
35
35
3
REPUTATION AND FREE EXPRESSION IN CASES OF INTERNET DEFAMATION
Introduction
More than a century ago defamation law was mockingly derided by
commentators as old and out of date, moss-covered with age,1 absurd in
theory, and very often mischievous in its practical operation,2 and in-
fected with the foolish conceits, absurd paradoxes, superstition, and artifi-
cial reasoning of a semi-barbarous age.3 Indeed this enfant terrible of the
[common] law,4 whose long and convoluted historical journey can be
traced back from the modern democratic and constitutional context to the
Roman delict of injuria via the Star Chamber in England,5 has for many
decades been the subject of intense academic and judicial scorn.6 Yet, de-
spite being a rather ungainly composite of two sister tortslibel and
slanderdefamation law has persisted.
Notwithstanding its evident longevity, one would be forgiven for won-
dering whether this cause of action, which once served the decidedly me-
dieval purpose of averting blood duels among easily inflamed British no-
blemen,7 might be on a collision course with the emerging communica-
tions technologies of the digital ageespecially the Internet. For, as we
will see, the medium of an allegedly defamatory communication has al-
ways beenand indeed continues to beone of the most important de-
terminants of how a plaintiff in a common law defamation action will fare
in meeting the requisite elements of the tort, fending off any defences that
the defendant might raise, and collecting a significant damage award at
the close of proceedings. It behooves us to ask: how will a tort, that over a
1 James C Courtney, Absurdities of the Law of Slander and Libel (1902) 36:4 Am Law
Rev 552 at 552.
2 Van Vechten Veeder, The History and Theory of the Law of Defamation (1903) 3:8
Colum L Rev 546 at 546.
3 Courtney, supra note 1 at 552.
4 As the tort was described in Donnelly: Richard C Donnelly, The Law of Defamation:
Proposals for Reform (1949) 33:6 Minn L Rev 609 at 609 [Donnelly, Proposals].
5 Veeder, supra note 2 at 550. The Star Chamber was composed of the highest dignitar-
ies of Church and State including the chancellor, treasurer, Lord Privy Seal, a bishop,
a temporal lord, and the two chief justices, or in their absence, two other judges as as-
sistants. The court did not abide by any particular forms, was not bound by any rules of
evidence, and heard only from its own appointed counsel (ibid at 562).
6 See e.g. Thorley v Lord Kerry (1812), 4 Taunt 355, 128 ER 367 (CP) [Thorley]; W Reade
Jr, The Law of Slander, Its Present State, and Possible Improvement (1866) 21:2 Law
Mag & Law Rev 215; Veeder, supra note 2 at 546; William L Prosser, Handbook of the
Law of Torts (St Paul, Minn: West Publishing, 1941) at 777, 807; Donnelly, Proposals”,
supra note 4; J Howard Toelle, The Law of DefamationSuggestions for Reform
(1948) 9 Mont L Rev 17.
7 Hill v Church of Scientology of Toronto, [1995] 2 SCR 1130 at para 117, 126 DLR (4th)
129 [Hill].
4 (2010) 56:1 MCGILL LAW JOURNAL ~ REVUE DE DROIT DE MCGILL
century ago was regarded as being three hundred years behind the age,8
rationally assimilate communications technologies that could have
scarcely been imagined even twenty-five years ago?
Though the jurisprudence is yet in its infancy, some alarming trends
have already begun to emerge in the common law cases involving allega-
tions of defamation using the Internet, sometimes referred to as cyber-
libel.9 Chief among these has been a tendency on the part of the courts to
rely upon generalizations about the Internet that paint the medium as be-
ing uniformly dangerous to individual reputation. In this regard, the re-
frain that has peppered the emerging jurisprudence has been that the
Internet is instantaneous, seamless, interactive, blunt, borderless and
far-reaching and potentially a medium of virtually limitless interna-
tional defamation.10 Fuelled by these generalizations, Canadian courts
have seemed to view their role in cyber-libel actions as that of a final bul-
wark against the defamatory excesses of those members of the general
public who might abuse the tremendous new power entrusted to them by
the Internet. This has led the courts to hold that when defamatory words
are transmitted using the Internet, the availability of any qualified privi-
lege that would otherwise have immunized the defendant from liability
under traditional defamation principles will be vitiated, and will substan-
tially increase any resulting award of damages.
These aspects of the emerging cyber-libel jurisprudence are of concern
because they disrupt the delicate balance between the two competing con-
stitutional values or goals that underpin the modern tort of defamation:
protecting reputation (including personal dignity) and securing freedom of
expression.11 By treating the vast and diverse world of Internet communi-
8 Courtney, supra note 1 at 564.
9 See Elizabeth F Judge, Cybertorts in Canada: Trends and Themes in Cyber-Libel and
Other Online Torts in The Honourable Justice Todd Archibald & The Honourable Jus-
tice Randall Echlin, eds, Annual Review of Civil Litigation 2005 (Toronto: Thomson
Carswell, 2006) 149; Barrick Gold Corp v Lopehandia (2004), 71 OR (3d) 416 at para
28, 239 DLR (4th) 577 (CA) [Barrick Gold].
10 Ibid at para 31; Matthew Collins, The Law of Defamation and the Internet (Oxford: Ox-
ford University Press) at para 24.02; Griffin v Sullivan, 2008 BCSC 827 at para 97
(available on WL Can) [Griffin]; Inform Cycle Ltd v Rebound Inc, 2008 ABQB 369 at
para 32, 438 AR 80; Vaquero Energy Ltd v Weir, 2004 ABQB 68 at para 18, 352 AR 191;
Sanjh Savera Weekly v Ajit Newspaper Advertising, 2006 Carswell Ont 3777 (WL Can)
at para 49 (Ont Sup Ct) affd 2008 ONCA 145 (available on CanLII); Spiros Pizza &
Spaghetti House Ltd v Riviera Pizza Inc, 2005 ABQB 80 at para 96, 377 AR 266 [Spi-
ros]. See also Manson v Moffett, 2008 CanLII 19789 (Ont Sup Ct) ([b]y any reasonable
definition, anything disseminated on the internet is intended for consumption by a wide
audience and easily meets any reasonable definition of publication. Indeed, the very
concept of a worldwide web invites no other reasonable interpretation at para 8).
11 Hill, supra note 7 at paras 100-21.
5
REPUTATION AND FREE EXPRESSION IN CASES OF INTERNET DEFAMATION
cations as an undifferentiated and uniformly menacing whole, the courts
improperly favour plaintiffs in most cyber-libel cases to the detriment of
vibrant online free expression, and devalue individual dignity and reputa-
tion in cases involving other, less-feared media.
In order to fully appreciate the nature of the emerging cyber-libel ju-
risprudence, it is important to understand that it is actually the product
of a long-standing, and unfortunate, analytical methodology that has
woven itself into the very fabric of defamation law. This methodology,
which can be traced back to the centuries-old distinction between libel
and slander, flows from a judicial tendency among common law judges to
attach disproportionate legal significance to the medium through which
allegedly defamatory statements were made. Courts have historically re-
lied upon crude generalizations about particular media and have tended
to categorize them as being either extremely dangerous or presumptively
unthreatening to reputation. In this calculation, the side of the fence that
a particular medium fell was, and in many jurisdictions continues to be,
crucial for potential plaintiffs since the legal consequences for being cate-
gorized one way or the other would often determine the outcome of their
cases. Though the direct significance of the widely reviled libel-slander di-
chotomy has been dampened somewhat in Canadaat least in those
provinces that have eliminated it by statute12the judicial tendency to
rely on generalizations characterizing new modes of communication as be-
ing tremendously pernicious persists in the emerging cyber-libel jurispru-
dence and has expressed itself through rulings dealing with the qualified
privilege defence and the calculation of damages.
The link between judicial hostility toward the use of the Internet in
emerging cyber-libel case law and the historical common law tendency to
place too much emphasis on menacing caricatures of new communications
media, is illustrated by a comparison with the civil law approach to the
law of defamation in the province of Quebecfor cases both involving the
Internet as well as in general. The law in this jurisdiction developed be-
yond the long shadow of the libel-slander dichotomy and pays scant atten-
12 The distinction has been eliminated in all Canadian provinces except British Columbia,
Ontario, and Saskatchewan. It never applied under the civil law of Quebec. See Defa-
mation Act, RSA 2000, c D-7, s 1(b); The Defamation Act, RSM 1987, c D20, s 1, CCSM c
D20, s 1; Defamation Act, RSNS 1989, c 122, s 2(b); Defamation Act, RSNB 1973, c D-5,
s 1; Defamation Act, RSPEI 1988, c D-5, s 1(b); Defamation Act, RSNL 1990, c D-3,
s 2(b); Defamation Act, RSY 2002, c 52, s 1; Defamation Act, RSNWT 1988, c D-1, s 1. In
Australia, the libel-slander distinction has been abolished in New South Wales, Queen-
sland, and Tasmania. In the United States, it has been abolished in Illinois, New Mex-
ico, Virginia, and Washington, and it never applied in Louisiana. See Raymond E
Brown, Defamation Law: A Primer, 1st ed (Toronto: Thomson Carswell, 2003) at 12
[Brown, Primer].
6 (2010) 56:1 MCGILL LAW JOURNAL ~ REVUE DE DROIT DE MCGILL
tion to the particular medium in which defamatory imputations happen to
have been conveyed. Rather, in each case the civil law courts look to the
broader context in order to determine both liability and the quantum of
damages. The essence of this approach, which can be easily transplanted
into the common law context, represents the key to maintaining the ap-
propriate balance between the two competing constitutional values in cy-
ber-libel actions and in all defamation cases.
In Part II of this paper, I will attempt to elucidate the distrustful judi-
cial stance toward Internet communications that has animated the Cana-
dian common law cyber-libel jurisprudence to date. In Part III of the pa-
per, I will argue that this approach is unjustifiably disruptive to the
proper balance between free expression and protection of reputation. In
Part IV, I will seek to demonstrate how the modern common law approach
to cyber-libel is actually the product of an impoverished analytical meth-
odology that has governed defamation law for centuries. In Part V of the
paper, I will canvass the law of defamation in Quebec so as to draw the
unwise aspects of the common law into sharper relief and to offer a poten-
tial avenue for reform. In Part VI, I will detail the contours of a more nu-
anced and less categorical approach to defamation adjudication in the
digital age that takes inspiration from the civil law approach, and more
appropriately reconciles freedom of expression and the vindication of per-
sonal reputation. Finally, in Part VII I will briefly discuss the Supreme
Court of Canadas recent decision in Grant v. Torstar Corp., which may
mark the beginnings of a laudable turning point in the Canadian common
law approach to cyber-libel along the lines advocated in Part VI.13
I. The Canadian Common Law Approach to Cyber-Libel
The common law Canadian courts that have to date been faced with
allegations of defamation over the Internet, have considered such com-
munications to be indiscriminate and akin in many respects to defama-
tory statements made on broadcast radio or television. As a result, the
present state of the law appears to be that the qualified privilege defence
will rarely, if ever, be available in cases of cyber-libel, and resulting dam-
age awards will be greatly increased where defamation is proven.
A. Defamation over the Internet and the Qualified Privilege Defence
The defence of qualified privilege seeks to carve out a zone of commu-
nication within which, for reasons of general social utility, the protection
13 2009 SCC 61, [2009] 3 SCR 640 [Grant].
7
REPUTATION AND FREE EXPRESSION IN CASES OF INTERNET DEFAMATION
of reputation must yield to open and free discussion.14 The rationale for
this defence is that, due to the public benefit in encouraging certain kinds
of communications to be made, no matter how harsh, hasty, untrue, or
libellous the publication[,] … the amount of public inconvenience from the
restriction of freedom of speech or writing [if an action for defamation was
available] would far outbalance that arising from the infliction of private
injury.15 The legal effect of making out a claim for qualified privilege is
that the inference that the words were published with malice is rebutted,
which arises whenever a plaintiff proves that the defendant published de-
famatory words about them to a third party.16
It is important to note that qualified privilege does not attach to the
impugned communication itself. Rather, the privilege attaches to the oc-
casion at issue.17 An occasion will be regarded as being privileged if it in-
volves a communication made by a person in the discharge of a public or
private duty and is communicated to an audience that has some corre-
sponding interest or duty to receive the statement.18 Such duties or inter-
ests (the categories of which are not closed) may be personal, social, busi-
ness-related, financial, moral, or legal.19 Classic examples of privileged oc-
casions include the provision of employment references, business and
credit reports, and complaints to police, regulatory bodies, or other public
authorities.20
The privilege associated with a particular occasion will be lost if the
statement is not commensurate with the occasion.21 This will arise if the
content of the information or the manner in which that information was
communicated was not reasonably appropriate for the occasion.22 For ex-
14 Cusson v Quan, 2007 ONCA 771 at para 37, 87 OR (3d) 241[Cusson], revd on other
grounds 2009 SCC 62, 314 DLR (4th) 55. See also Dinyer-Fraser v Laurentian Bank
2005 BCSC 225 at para 196, 40 BCLR (4th) 39.
15 Huntley v Ward (1859), 6 CB (NS) 514 at 517, 175 ER 848 [Huntley], cited in Cusson,
supra note 14 at para 39.
16 Hill, supra note 7 at para 144; Horrocks v Lowe, [1975] AC 135 at 149, [1974] 2 WLR
282 (HL (Eng)).
17 Campbell v Jones, 2002 NSCA 128 at paras 31, 209 NSR (2d) 81 [Campbell] [emphasis
added]; Hill, supra note 7 at para 143; Adam v Ward, [1917] AC 309 at 334 (HL (Eng))
[Adam].
18 Hill, supra note 7 at para 143; Campbell, supra note 17 at paras 30-34.
19 See RTC Engineering Consultants Ltd v Ontario (Solicitor General) (2002), 58 OR (3d)
726 at para 16, 156 OAC 96 (CA); Hill, supra note 7 at para 143; Adam, supra note 17
at 334.
20 Cusson, supra note 14 at para 39.
21 Hill, supra note 7 at para 146.
22 Ibid at paras 146-47.
8 (2010) 56:1 MCGILL LAW JOURNAL ~ REVUE DE DROIT DE MCGILL
ample, if the words complained of were published to the public generally
or, as it is sometimes expressed, to the world,23 the qualified privilege
will generally have been exceeded since it reached a large number of re-
cipients who did not have the requisite interest in receiving the communi-
cation.24 In this regard, statements communicated using certain media
have been branded by courts and commentators as necessarily represent-
ing communications to the world. Citing a number of lower-court au-
thorities from the 1970s,25 Professor Brown maintains that newspapers,
radio, and television publish information indiscriminately and must
therefore be regarded as being publications to the world in respect of
which the qualified privilege defence will be unavailable.26
The first case in Canada to directly deal with the availability of the
qualified privilege defence in a cyber-libel action was Christian Labour
Association of Canada v. Retail Wholesale Union, a decision by the British
Columbia Supreme Court.27 In this case, Justice Rice was faced with an
application by way of summary trial to dismiss a defamation action. The
underlying facts were not in dispute and the two parties were both unions
with a history of competing for members. Statements alleging that the
plaintiff was a rat union in the habit of signing substandard agreements
were posted on the defendants website and remained online for about
thirteen months. The defendants membership was about 2,300 and the
total number of visits to the impugned page was estimated to be about
23 Jones v Bennett, [1969] SCR 277, 2 DLR (3d) 291.
24 It should be noted that in some cases the courts have found that the qualified privilege
associated with a particular occasion was not lost in a communication to the world
where the matter at issue was one of general public interest and the party who pub-
lished it owed a duty to communicate it to the general public. See e.g. Camporese v Par-
ton (1983), 150 DLR (3d) 208 at 226-27, 47 BCLR 78 (SC (AD)); Parlett v Robinson
(1986), 30 DLR (4th) 247, 5 BCLR (2d) 26 (CA); Grenier v Southam Inc, 1997 CanLII
4460 (Ont CA); Leenen v Canadian Broadcasting Corp (2000), 48 OR (3d) 656 at para
695, 50 CCLT (2d) 213 (Sup Ct), affd (2001), 54 OR (3d) 612, 6 CCLT (3d) 97 (CA), leave
to appeal to SCC refused (2002), 289 NR 200 (note), 164 OAC 200 (note); Young v To-
ronto Star Newspapers Ltd (2003), 66 OR (3d) 170, 18 CCLT (3d) 244 (Sup Ct), affd
(2005), 77 OR (3d) 680, 259 DLR (4th) 127 (CA). As discussed in greater detail in Part
VII below, the Supreme Court of Canada in Grant, found that the to the world limita-
tion on the qualified privilege defence unduly impacted upon the right to freedom of ex-
pression. It thus created a new defence of responsible communication that would al-
low communications on matters of public interest to be inoculated from liability where
certain steps to ascertain the truth of the facts asserted are taken (supra note 13).
25 Whitaker v Huntington (1980), 15 CCLT 19 (BCSC); Loan v MacLean (1975), 58 DLR
(3d) 228 (BCSC (AD)).
26 Raymond E Brown, The Law of Defamation in Canada, vol 1, 2d ed (Scarborough, Ont:
Carswell, 1994) at s 13.7(2)(i) [Brown, Defamation].
27 Christian Labour Association of Canada v Retail Wholsesale Union, 2003 BCSC 2000,
130 ACWS (3d) 643 [Christian Labour Association].
9
REPUTATION AND FREE EXPRESSION IN CASES OF INTERNET DEFAMATION
530. The parties were not able to provide conclusive evidence as to the
identity of those that had accessed the posting, so it was unknown
whether it had been viewed by individuals other than members of the de-
fendant union to whom the message was directed. The defendant had not
put in place any access controls such as user ID numbers or passwords
that would have restricted the ability of non-union members to view the
page.28
The parties agreed that the words complained of were defamatory.
They also agreed that the defendants members had an interest in receiv-
ing the union news such that the communication was made on an occa-
sion of qualified privilege. Finally, the parties agreed that the general
public did not have any legitimate interest in accessing the defamatory
message.29
The only question to be determined by Justice Rice on the summary
trial application was whether the availability of the qualified privilege de-
fence had been lost by virtue of the fact that the impugned publication
had been posted on the defendants website. Justice Rice concluded that it
had indeed been lost and, in arriving at this conclusion, the court focused
on the absence of access controls to the defendants website. On that basis,
Justice Rice felt bound to conclude that probably a significant number of
those who accessed and presumably read the message were not within the
group of interested persons entitled to receive the information.30 Since
the defendant was unable to adduce compelling evidence explaining why a
posting on a website without restrictions was reasonably necessary un-
der the circumstances, the court found that the qualified privilege associ-
ated with the occasion had been lost.31
The court in Christian Labour Association seemed to accept the basic
proposition espoused by Professor Brown that certain media, such as ra-
dio and television, transmit publications in an indiscriminate way such
that the qualified privilege defence will generally be lost when employed.
Indeed, immediately preceding the courts analysis of the impugned web
page, Justice Rice cited Professor Browns views on the availability of the
qualified privilege defence in cases of indiscriminate publications32 and
lamented that there is no authority that either counsel or I could find on
28 Ibid at para 6.
29 Ibid.
30 Ibid at para 24.
31 Ibid at paras 25-30.
32 Brown, Defamation, supra note 26 at s 13.7(2)(i), cited in Christian Labour Association,
supra note 27 at para 20.
10 (2010) 56:1 MCGILL LAW JOURNAL ~ REVUE DE DROIT DE MCGILL
this issue where the medium was a website.33 Ultimately, Justice Rice
implicitly concluded that a web posting that is not protected by strict ac-
cess controls does indeed represent an indiscriminate publication akin to
a broadcast on radio or television, and ruled accordingly.
The courts reasoning in Christian Labour Association was more re-
cently endorsed by the Alberta Court of Appeal in Angle v. LaPierre.34
This case involved a posting on a website devoted to critiquing the state of
public education in Alberta. The posting was found by the trial judge to
have been defamatory of both the defendant teachers and their union. In
curtly dismissing the availability of the qualified privilege defence the
Court of Appeal cited Christian Labour Association and did not even con-
sider, as Justice Rice had, whether there was any evidence regarding the
number or identity of those who would have likely visited the site. Rather,
the court accepted an analytical framework pursuant to which statements
to the world that are made using indiscriminate media such as televi-
sion or radio are automatically disqualified from the protection of the
qualified privilege defence. While it did not say so explicitly, the court
appeared to suggest that any publication on the Web will be found to be
excessive,35 such that the qualified privilege defence will be unavailable
unless it can be shown that various access controls limiting the scope of
publication were either employed or were practically unreasonable to
have been employed under the circumstances.
By equating all web publicationsno matter how obscure in nature
with indiscriminate broadcasts such as traditional television or radio,
the courts in Christian Labour Association and Angle effectively eviscer-
ated the qualified privilege defence in respect of almost any posting made
on the Web. This is so since the vast majority of online publications are
not protected by overt access controls of any kind and are thus open to be-
ing labelled by the courts as being indiscriminate.
B. Defamation over the Internet and the Calculation of Damages
With regard to the calculation of general damages in a defamation ac-
tion, these are presumed from the very publication of the defamatory
statement itself.36 However, in determining the quantum of general dam-
ages, courts do not treat all defamatory publications alike. Some correla-
33 Christian Labour Association, supra note 27 at para 21.
34 2008 ABCA 120 at paras 16-20, 425 AR 378 [Angle].
35 Ibid at para 16.
36 Hill, supra note 7 at para 164, citing Ley v Hamilton (1935), 153 LT 384 at 386 (HL
(Eng)).
11
REPUTATION AND FREE EXPRESSION IN CASES OF INTERNET DEFAMATION
tion between the severity of the damage to the reputation of the plaintiff
and the resulting quantum must be sought. As such, in calculating dam-
ages, the trier of fact is required to consider, inter alia, the conduct of the
plaintiff, the plaintiffs position and standing, the nature of the defama-
tory statement, the mode and extent of publication, and the absence of
any retraction or apology.37
The leading Canadian case on the calculation of damages in cases of
cyber-libel is Barrick Gold, a 2004 judgment of the Ontario Court of Ap-
peal.38 In this case the plaintiff, a publicly traded mining company and
one of the worlds leading producers of gold,39 successfully obtained a de-
fault judgment at trial against the defendants, a private individual and
his alter ego corporation. The personal defendant, Mr. Lopehandia, who
apparently believed that one of the plaintiffs gold mines in Chile belonged
to him, had engaged in a sustained online campaign over a number of
years in which he posted increasingly outlandish and defamatory mes-
sages regarding the plaintiff on various online bulletin and message
boards.
The trial judge, Justice Swinton, aptly described the defendants post-
ings as emotional, often incoherent, rambling and highly critical.40 Mr.
Lopehandia accused the plaintiff of, inter alia, fraud, tax evasion, money
laundering, manipulation of world gold prices, misrepresentation to gov-
ernment officials, pursuing organized crime, attempted murder, arson,
genocide, and crimes against humanity. Since Mr. Lopehandia did not de-
fend the action, the main question to be decided by Justice Swinton was
the quantum of damages to be awarded. In this regard, the court awarded
$15,000 in general damages and no aggravated or punitive damages. In so
ruling, the court reasoned that while the Internet empowered Mr. Lope-
handia to spread his many messages around the world, the emotional
and highly intemperate nature of the impugned statements would not
have been taken seriously by reasonable readers.41
Though the plaintiff in Barrick Gold had been seeking $250,000 in
punitive damages, Justice Swinton declined to make an award under this
heading. The court noted that in considering whether to award punitive
damages, various factors must be examined including the level of blame-
37 Ibid at para 182, citing Philip Lewis, ed, Gatley on Libel and Slander, 8th ed (London,
UK: Sweet & Maxwell, 1981) at para 1451.
38 Supra note 9.
39 Barrick Gold Corp v Lopehandia, 2003 CarswellOnt 6075 at para 4 (Sup Ct), revd
(2004), Barrick Gold, supra note 9.
40 Ibid at para 8.
41 Ibid at para 38.
12 (2010) 56:1 MCGILL LAW JOURNAL ~ REVUE DE DROIT DE MCGILL
worthiness of the defendant’s conduct, the financial or other vulnerability
of the plaintiff, and the need for deterrence.42 With regard to blamewor-
thiness, the court noted that Mr. Lopehandia had persisted in his con-
duct for a lengthy period of time, despite demands from Barrick that he
desist.43 However, the court found that this factor was mitigated by the
emotional and unreasoned tenor of the messages and the fact that no
reasonable business person or investor would have taken him seriously.44
With regard to vulnerability, Justice Swinton found that the plaintiff was
not vulnerable. To the contrary, the court held that [t]his is not a case
where the defendant is abusing power – indeed, the powerful party here is
the plaintiff.45 Finally, in declining to award any punitive damages, Jus-
tice Swinton held that an award of $15,000 plus costs was a sufficient de-
terrent in the case of most individuals sued for libel by a corporate plain-
tiff.46
On appeal, the plaintiff was successful in setting aside the quantum of
damages awarded by the trial judge. In a majority decision written by
Justice Blair, the Ontario Court of Appeal increased the general damages
awarded to the plaintiff to $75,000 and added $50,000 in punitive dam-
ages. The theme that ran throughout Justice Blairs majority judgment
was that defamatory communications made using the Internet are ex-
tremely dangerous to the reputations of individuals and corporations such
that they should be heavily punished by the courts through the imposition
of high damage awards.
Justice Blair first outlined a number of general considerations con-
cerning Internet defamation to guide lower courts assessment of damages
in future cases. He began by characterizing communications over the
Internet as being instantaneous, seamless, inter-active, blunt, borderless
and far-reaching.47 Moreover, since such communications are often im-
personal and anonymous, Justice Blair considered that there exists a
greater risk that … defamatory remarks [will be] believed.48 In sum, the
42 Ibid at para 46.
43 Ibid at para 47.
44 Ibid at para 49.
45 Ibid at para 50 [emphasis added].
46 Ibid at para 53.
47 Barrick Gold, supra note 9 at para 32.
48 Ibid at para 31. For a criticism of this view, see Judge, supra note 9 ([i]dentity, includ-
ing the speaker’s credentials and any bias, has been an important criterion of credibil-
ity, and continues to be so in the age of the Internet. Speakers can be credible when
they communicate online, but the Internet context has not introduced an entirely new
epistemological framework for parsing credibility. The traditional notions that credibil-
ity turns on such factors as the identifiability of the speaker, the speaker’s authority
13
REPUTATION AND FREE EXPRESSION IN CASES OF INTERNET DEFAMATION
court concluded that defamation over the Internet had a greater potential
to damage reputation as compared with its less pervasive cousins.49 As
such, [t]he mode and extent of publication is … a particularly significant
consideration in assessing damages in Internet defamation cases.50 In ar-
riving at these conclusions, the majority of the court quoted extensively
and approvingly from an article by Lyrissa Barnett Lidsky51 in which the
author underscores the Internets tremendous power to harm reputation
due to its extraordinary capacity … to replicate almost endlessly any de-
famatory message.52
Having articulated the general principles governing the calculation of
damages in cases of cyber-libel, the court then went on to apply those
principles to the case at bar. As noted above, this involved an almost ten-
fold increase in the quantum of damages ordered by the lower court. In
increasing the general damages awarded to $75,000, Justice Blair reiter-
ated the distinctive capacity of the Internet to cause instantaneous, and
irreparable, damage to the business reputation of an individual or corpo-
ration by reason of its interactive and globally all-pervasive nature and
the characteristics of Internet communications.53 In awarding $50,000 in
punitive damages, the majority of the Court of Appeal held, inter alia,
that Justice Swinton had misconceived the balance of power between Mr.
Lopehandia and Barrick Gold. In particular, Justice Blair held that:
Barrick is not the powerful party in the context of the Internet.
The impact of the Internet is to neutralize whatever “power” Barrick
may have had, in terms of a communication battle with Mr. Lope-
handia. In reality it is Barrick that is vulnerable to publications of
this nature, and Mr. Lopehandia who is abusing his power. The
Internet is one of the most powerful tools of communications ever in-
vented and … it is potentially a medium of virtually limitless inter-
national defamation.54
In sum, the analysis employed by the majority of the court in Barrick
Gold was focused on the Internet as an extremely dangerous tool that can
be abused by individuals of limited means, to ruinously defame otherwise
powerful entities such as multinational gold mining corporations.
and experience, other writings by the speaker, and any bias still hold true on the Inter-
net as they did for print media at 157).
49 Barrick Gold, supra note 9 at para 34.
50 Ibid.
51 Lyrissa Barnett Lidsky, Silencing John Doe: Defamation & Discourse in Cyberspace
(2000) 49:4 Duke LJ 855 at 862-65.
52 Ibid at 863-64.
53 Barrick Gold, supra note 9 at para 44.
54 Barrick Gold, supra note 9 at 62 [references omitted].
14 (2010) 56:1 MCGILL LAW JOURNAL ~ REVUE DE DROIT DE MCGILL
The majoritys decision in Barrick Gold has become the leading Cana-
dian common law authority on the quantification of damages in cyber-
libel cases and has been explicitly followed by a number of lower courts in
order to justify increased damage awards.55
II. Why This Matters: Constitutional and Normative Implications
The broad generalizations deployed by the Canadian courts in the
emerging cyber-libel jurisprudence regarding the grave danger to per-
sonal reputation posed by Internet communications, and the stark legal
consequences that flow from them, are of concern because such an ap-
proach sharply disrupts the delicate balance between freedom of expres-
sion and the protection of reputation that lies at the heart of defamation
law.
In Hill, Justice Cory, writing for a majority of the Supreme Court of
Canada observed that [t]here can be no doubt that in libel cases the twin
values of reputation and freedom of expression will clash.56 Justice Cory
cited the holding of Justice Edgerton in Sweeney v. Patterson,57 to the ef-
fect that whatever is added to the field of libel is taken from the field of
free debate.58 Ultimately, Justice Cory held that the right to free expres-
sion and the right to the protection of ones good reputation are two inter-
ests that are equally important and must be carefully balanced by the
courts in crafting the contours of the tort of defamation.59 By associating
55 See Warman v Grosvenor (2008), 92 OR (3d) 663 at para 77 (Sup Ct) [Warmen]; Griffin,
supra note 10 at para 97; WeGo Kayaking Ltd v Sewid, 2007 BCSC 49 at para 90, 154
ACWS (3d) 863; Newman v Halstead, 2006 BCSC 65 at para 256, 146 ACWS (3d) 153;
Spiros, supra note 10 at para 96; Hay v Partridge, 2004 NUCJ 3 at para 3 (available on
CanLII). But see Reaburn v Langen, 2008 BCSC 1342, 61 CCLT (3d) 227, in which the
court awarded $22,000 in damages stemming from the publication of an article in the
Kootenay Chronicle newspaper and on the newspapers website. The court found the
fact that the website was relatively unknown to be a factor that tended to reduce the
quantum of damages (ibid at 83). The court did not refer to the Barrick Gold decision or
any other cases involving defamation over the Internet.
56 Supra note 7 at para 103.
57 128 F 2d 457 at 458, 76 US App DC 23 (1942), cert denied 317 US 678, 63 S Ct 160
(1942).
58 Hill, supra note 7 at para 100. See also Cusson, supra note 14 at para 127, citing Derby-
shire County Council v Times Newspapers Ltd, [1993] AC 534 at 547, [1993] 1 All ER
1011 (HL (Eng)) and Theophanous v Herald & Weekly Times Ltd. (1994), 182 CLR 104,
124 ALR 1 at 19 (HCA).
59 Hill, supra note 7 at para 121. More recently, in WIC Radio Ltd v Simpson, 2008 SCC
40, [2008] 2 SCR 420 [WIC Radio], the Supreme Court reaffirmed the importance of en-
suring that the tort of defamation and its associated defences are crafted and applied in
a manner that respects the appropriate balance between free expression and the protec-
tion of reputation.
15
REPUTATION AND FREE EXPRESSION IN CASES OF INTERNET DEFAMATION
draconian legal consequences with a particular media of communication
such as the Internetbased on oversimplified caricatures about the dan-
gers to individual reputation posed by those mediathe courts ultimately
undermine both freedom of expression and the vindication of individual
reputation and dignity.
Consider, for example, the characterization by the courts in Christian
Labour Association and Angle of all non-password protected web postings
as indiscriminate publications that necessarily exceed any qualified
privilege that a defendant may have had. This rule treats the billions
upon billions of pages that make up the Web as alike, despite the well-
known variability in the traffic that particular pages enjoy. Whereas some
pages (e.g., Yahoo! or Google) might indeed enjoy such prolific and diverse
traffic as to make it likely that any qualified privilege is exceeded no mat-
ter what the occasion, other pages might cater to such a specific and nar-
row audience (e.g., the union Web page in Christian Labour Association or
the public school education page in Angle) that the privilege is pre-
served.60
It must be recalled that the rationale behind the defence of qualified
privilege is that certain species of communication are so necessary to the
public welfare that the damage caused by their restriction through defa-
mation actions would outweigh any private harm that might incidentally
arise from any occasional instances of defamation.61 By relying on broad
and indefensible generalizations about the indiscriminate nature of web
postings, the courts throw out the proverbial baby with the bathwater. In
cases where the traffic to a particular site is composed almost exclusively
of individuals who would have had the requisite interest in accessing a
particular posting, the imposition of a blanket rule negating the availabil-
ity of the qualified privilege defence for all web pages undermines the
very purpose for the existence of the qualified privilege defence. This, in
turn, disrupts the appropriate balance at the heart of defamation law by
60 This is so since the qualified privilege defence remains available at common law where
an impugned statement only incidentally reached individuals who did not have the
requisite interest or duty to receive the communication (Brown, Defamation, supra note
26 at s 13.7(2)(b)). See e.g. Pleau v Simpson-Sears Ltd (1977), 15 OR (2d) 436, 75 DLR
(3d) 747, (CA) (the defendant, a large department store, affixed posters on each of its in-
store cash registers warning its employees about persons who had stolen the plaintiff’s
wallet and were signing bad cheques in his name. Though the posters could be seen by
persons passing the register (including customers who had no legitimate interest in the
contents of the posters), the court held that to casual observers the publication was
merely incidental and did not vitiate the privilege that had otherwise been established
by the defendant). See also Fisher v Rankin, 27 DLR (3d) 746 at para 34, [1972] 4 WWR
705 (BCSC (AD)).
61 Huntley, supra note 15, cited in Cusson, supra note 14 at para 39.
16 (2010) 56:1 MCGILL LAW JOURNAL ~ REVUE DE DROIT DE MCGILL
over-vindicating reputation at the cost of socially beneficial categories of
free expression.
Similarly, the judicial attitude that publication over the Internet, no
matter what the context, automatically increases the quantum of dam-
ages awarded in a defamation action, is also based on an indefensible
over-generalization about the Internet that serves to disrupt the appro-
priate balance between free expression and the protection of reputation.
The doctrinal error here is the repeated reliance on the potential harm of
defamatory imputations over the Internet rather than a careful examina-
tion of the particular imputation at issue in any given case. Since many
Internet communications will tend to reach only a very small audience
(such as the union web page in Christian Labour Association), and since
many of these are rambling, incoherent and unbelievable (such as those of
Mr. Lopehandia in Barrick Gold), increased damage awards in all cases of
cyber-libel overprotects individual reputation at the cost of free expres-
sion.
If oppressive damage awards against online critics are granted as a
matter of course when no damage to a plaintiffs reputation was reasona-
bly likely to have resulted, the obvious danger that will arise is that fu-
ture critics will be intimidated into silence. This, in turn, would under-
mine one of the most promising features of the Internet, which is its well-
known capacity to democratize the marketplace of ideas62 so that less
62 The conceptual roots of the marketplace of ideas metaphor can be traced to the
famous dissent by Oliver Wendell Holmes J in Abrams v United States, 250 US 616,
40 S Ct 17 (1919). It connotes an understanding of the right to free expression as a
means of promoting free competition among ideas that vie for supremacy to the end of
attaining the truth. See Thomas I Emerson, The System of Freedom of Expression (New
York: Random House, 1970) at 627. The Supreme Court of Canada has repeatedly
noted that a vibrant marketplace of ideas is an essential component of any functioning
democracy. See e.g. Reference re Secession of Quebec, [1998] 2 SCR 217 at para 68, 161
DLR (4th) 385, where the Court held:
[A] functioning democracy requires a continuous process of discussion. The
Constitution mandates government by democratic legislatures, and an ex-
ecutive accountable to them, resting ultimately on public opinion reached by
discussion and the interplay of ideas … No one has a monopoly on truth, and
our system is predicated on the faith that in the marketplace of ideas, the best
solutions to public problems will rise to the top. Inevitably, there will be dis-
senting voices. A democratic system of government is committed to consider-
ing those dissenting voices, and seeking to acknowledge and address those
voices in the laws by which all in the community must live [references omit-
ted, emphasis added].
See also Canada (Human rights commission) v Taylor, [1990] 3 SCR 892, 75 DLR (4th)
577; R v Keegstra, [1990] 3 SCR 697, [1991] 2 WWR 1; R v Butler, [1992] 1 SCR 452, 89
DLR (4th) 449; Haig v Canada (Chief Electoral Officer), [1993] 2 SCR 995, 105 DLR
17
REPUTATION AND FREE EXPRESSION IN CASES OF INTERNET DEFAMATION
powerful voices are able to meaningfully compete in the ongoing search
for truth.63 It is relevant to note that the institution of defamation pro-
ceedings by powerful litigants so as to silence impecunious online critics
has become so common a phenomenon in the United States that it has
spawned its own memorable moniker: the cyber-SLAPP (or Strategic
Lawsuit Against Public Participation).64 By allowing wealthy parties who
have suffered no serious reputational harm (such as the plaintiff in Bar-
rick Gold) to silence their online critics through the collection of inflated
damage awards, the Canadian courts allow defamation actions involving
the Internet to take on an unfortunate speech-chilling, cyber-SLAPP qual-
ity.
Treating publication on the Internet as an aggravating factor in the
quantification of damages also serves to devalue the harm to reputation
(4th) 577; Native Womens Assn of Canada v Canada [1994] 3 SCR 627, 119 DLR (4th)
224; Harper v Canada (Attorney General), 2004 SCC 33 at para 35, [2004] 1 SCR 827.
63 Ironically, this very point was made rather forcefully by Lidsky in the article cited by
Blair JA in Barrick Gold in support of the rule that increased damage awards ought to
apply to cyber-libel cases: Lidsky, supra note 51, cited in Barrick Gold, supra note 9
([t]he chief threat posed by the new cases is that powerful corporate plaintiffs will use
libel law to intimidate their critics into silence and, by doing so, will blunt the effective-
ness of the Internet as a medium for empowering ordinary citizens to play a meaningful
role in public discourse at 945). See also David L Hudson, Jr, Blogs and the First
Amendment (2006) 11 NEXUS: J Opinion 129 (the author describes the Internet as a
First Amendment fantasyland where freedom of expression [can] reach its zenith at
129) See also Council of Europe, Committee of Ministers, Declaration of the Committee
of Ministers on human rights and the rule of law in the Information Society, CM (2005)
56 final, May 13, 2000 (the Council emphasized that the Internet can provide unprece-
dented opportunities for all to enjoy freedom of expression at I(1)). For a contrary view
on the promotion of free expression through the Internet, see e.g. Diane Rowland,
Griping, Bitching and Speaking Your Mind: Defamation and Free Expression on the
Internet (2006) 110:3 Penn State L Rev 519 ([a]lthough the Internet may provide the
perfect forum for free speech where all citizens can participate equally, unfettered by
barriers of race, class and religion, the participatory nature of the Internet and the ease
of anonymous communication may also foster anti-social, malicious and immoral behav-
iour. The impact of this aspect of Internet communication is such that others may be
deterred from entering the conversation at 520).
64 A cyber-SLAPP case typically involves anonymously posted online criticism of a corpo-
ration or public figure. The target of the criticism files a frivolous lawsuit, often but not
always framed as a defamation action, in order to issue a subpoena to the website or
Internet Service Provider (ISP) in question, discover the identity of the anonymous
critic, and thereby intimidate them into silence. For an extensive list of American cy-
ber-SLAPP cases, see cyberSLAPP.org, Dont chill online freedom of expression,
online:
ber-SLAPP cases, see Joshua R Furman, Cybersmear or Cyber-SLAPP: Analyzing
Defamation Suits Against Online John Does as Strategic Lawsuits Against Public Par-
ticipation, Comment (2002), 25:1 Seattle UL Rev 213; Sean P Trende Defamation,
Anti-SLAPP Legislation, and the Blogosphere: New Solutions for an Old Problem
(2006) 44:4 Duq L Rev 607.
18 (2010) 56:1 MCGILL LAW JOURNAL ~ REVUE DE DROIT DE MCGILL
and individual dignity that arises in cases involving non-Internet media.
Consider the hypothetical case of a plaintiff who, during the course of a
staff meeting, is falsely alleged by a co-worker to have been asleep on the
job. She loses a promotion, the respect of her co-workers, and her health
deteriorates. Now consider a case where an individual posts a comment
on an online forum devoted to discussing a particular publicly traded
company in which he falsely asserts that the company has lost money in
the previous quarter. No one believes him and the companys stock price
is unaffected. The application of the rule that publication over the Inter-
net increases the quantum of damages shifts the focus away from harm to
the plaintiffs reputation and toward irrelevant considerations about the
danger presented by the medium in which the words were conveyed. This,
in turn, might result in the artificial inflation of the quantum of damages
in the online scenario relative to the workplace setting in a manner that
does not reflect the true gravity of the injuries suffered by the two plain-
tiffs. Such a result would undermine the protection of reputation and in-
dividual dignity, one of the twin values at the heart of defamation law.
III. Contextualizing the Emerging Cyber-Libel Jurisprudence
While the Internet and the many communications applications that it
supports are still quite new, the mode of analysis employed by the Cana-
dian courts in cases of cyber-libel to date, is not. Rather, the jurisprudence
sketched out above can be understood as the by-product of an analytical
paradigm that has seamlessly woven itself into the very fabric of defama-
tion law over several centuries. This paradigm has emerged out of the
longstanding distinction between libel and slander, and involves the use
of extremely broad and unsustainable generalizations about the media
through which impugned statements are transmitted in order to deter-
mine how particular defamation cases are assessed. To fully appreciate
the paradigmatic link between todays cyber-libel jurisprudence and the
archaic libel-slander dichotomy, the nature, history, and justifications for
this rather reviled legal distinction must be briefly canvassed.
Libel and slander represent two distinct common law regimes for the
adjudication of defamation actions. The rules that constitute the tort of li-
bel are heavily tilted in favour of aggrieved plaintiffs, which has led some
to characterize it as a form of no-fault liability.65 In order to establish a
prima facie cause of action, a plaintiff need only show that the words
complained of (1) are reasonably capable of defamatory meaning, (2) refer
to the plaintiff, and (3) have been published to at least one third party.
65 Cusson, supra note 14 at para 34. See also Grant, supra note 13 at para 28 (the Court
characterizes the tort as one of strict liability).
19
REPUTATION AND FREE EXPRESSION IN CASES OF INTERNET DEFAMATION
The plaintiff is under no burden to demonstrate that the impugned
statement is untrue or that the defendant was at fault in publishing those
words. Most significantly, having proven each of the above three ele-
ments, the plaintiff need not demonstrate that any loss was suffered at all
in order to collect damages. The tort of libel can therefore be said to pre-
sume falsity, fault, and damages.66
By contrast, where an impugned statement is governed by the law of
slander, the rules are more heavily tilted in favour of defendants. Unless
the impugned words can be slotted into one of four specific exceptions that
constitute slander per se,67 plaintiffs in slander actions must prove that
they suffered special damages in order for the claim to be actionable.
Given that what is at issue is damage to reputation, an inherently nebu-
lous concept that is by nature difficult to prove, this evidentiary obligation
tends to either completely negate or greatly reduce the value of many
such claims.68 The exceptions to the obligation to prove special damages,
which evolved in a rather haphazard manner over time, depend on the
substance of the slanderous imputation. They are: (1) words disparaging
the reputation of the plaintiff in the plaintiffs trade or profession, (2)
words imputing the commission of a criminal offence, (3) words imputing
a loathsome or contagious disease, and (4) words imputing unchastity
to a woman.69 Given the relative narrowness of these exceptions, the stra-
tegic advantage to defendants flowing from the characterization of im-
pugned words as being slanderous rather than libellous, is significant and
often determinative.
For present purposes, what is most significant about these parallel re-
gimes is that the sole determinant of whether a plaintiffs claim is gov-
erned by the plaintiff-friendly rules of libel or the defendant-friendly rules
of slander, is the medium in which the impugned words were transmitted.
Initially the distinction was made between messages communicated
orally, which were governed by the law of slander, and messages in writ-
ing, which fell under the law of libel.70
66 Ibid; see also Patrick Milmo & WVH Rogers, eds, Gatley on Libel and Slander, 10th ed
(London, UK: Sweet & Maxwell, 2004) at para 3.6.
67 Brown, Defamation, supra note 26 at s 8.5.
68 See Richard C Donnelly, Defamation by Radio: A Reconsideration (1948) 34:1 Iowa L
Rev 12 ([b]ecause of the inherent difficulty in proving special damage, recovery for
defamation frequently depends upon the taxonomy adopted at 13).
69 Brown, Defamation, supra note 26 at s 8.5; Milmo & Rogers, supra note 66 at 79.
70 Allen M Linden & Bruce Feldthusen, eds, Canadian Tort Law, 8th ed (Markham, Ont:
LexisNexis Butterworths, 2006) at 774-75.
20 (2010) 56:1 MCGILL LAW JOURNAL ~ REVUE DE DROIT DE MCGILL
A practical problem with the application of this dichotomy, readily ap-
parent even centuries ago, was that many defamation cases involve com-
plaints about communications transmitted through media that are nei-
ther speech nor writing (e.g., non-verbal gestures), or contain elements of
both (e.g., speeches read aloud from written scripts).71 This practical diffi-
culty has produced jurisprudence that is rife with inconsistency, unpre-
dictability, and unprincipled distinctions. For example, the giving of an
impromptu speech to an audience was often held by the courts to be slan-
der, whereas the reading out of the very same defamatory words from a
prepared text to an audience was regarded as being libellous, whether or
not the audience knew of the script.72 Whereas placing a lamp in front of a
house so as to signify a brothel73 or burning a persons likeness in effigy74
were regarded as being libellous, a store owner who publicly prevented a
customer from leaving a store in order to effect a search for suspected
shoplifting had merely slandered the aggrieved customer.75
These practical difficulties emerged out of a fundamental flaw that lies
at the very heart of the libel-slander dichotomy. In short, it was never
reasonable to attach so much legal significance to the medium through
with a particular communication was transmitted. This approach
spawned an analytical paradigm that has perpetually distracted judges
away from looking at the true context in which particular words were
communicated in order to discern what effect they likely had on the plain-
tiffs reputation and dignity, and whether such an effect was legally justi-
fiable under the circumstances. Instead, this approach has directed courts
to classify and compare media based on necessarily vague generalizations
about the theoretical danger to reputation that they posed or were as-
sumed to pose. This analytical methodology necessarily reduced the rich
and diverse media of communication into absurd and cartoonish carica-
tures.
71 See Julie C Sipe, Old Stinking, Old Nasty, Old Itchy Old Toad: Defamation Law,
Warts and All (A Call for Reform) (2008) 41:1 Ind L Rev 137 ([u]sing two descriptors,
both written and oral, to demarcate the defamation universe utterly fails to define
communications that are neither written nor oral, and it fails to define precisely com-
munications that are both at 145).
72 Brown, Primer, supra note 12 at 82, citing Forrester v Tyrrell (1893) 9 TLR 257 (CA);
Robinson v Chambers (No 2) [1946] NI 148 (KB); Land v Delta Airlines, 250 SE 2d 188,
147 Ga App 738, (Ct App, 1978).
73 Jeffries v Duncombe (1809), 2 Camp 3, 170 ER 1061 (KB).
74 Eyre v Garlick (1878), 42 JP 68 (QB).
75 Bennett v Norban, 151 A 2d 476, 396 Pa 94 (1959); see also Cook v Cox (1814), 3 M & S
110 at 114, 105 ER 552 (KB) (the court included in the category of slander an act or
gesticulation, such as holding up an empty purse, or the like).
21
REPUTATION AND FREE EXPRESSION IN CASES OF INTERNET DEFAMATION
Take for example the oldest medium in defamation law: the use of oral
speech. It is clear that the capacity to cause harm in this way is wildly
variable depending on all of the relevant circumstances. On the one hand,
being targeted in a speech by a respected politician before an audience
of thousands might be far more damaging to an individual’s reputation
than having the same words uttered by an (apparently) deranged
speaker in Hyde Park. On the other hand, a carefully targeted rumour
spread quietly in a workplace could destroy a career, while a drunken
toast given to a large audience at a wedding might beno matter how vi-
cious the barbs it containssimply laughed off as innocuous tomfoolery.
In short, the extent to which any particular communication is likely to
have damaged a plaintiffs reputation will vary dramatically depending on
innumerable factors that have little to do with the medium in which the
words were communicated.
This obvious truth has long been recognized by common law courts
and commentators. For example, in the 1812 case of Thorley, Chief Jus-
tice of the Common Pleas James Mansfield held that he could not upon
principle, make any difference between words written and words spoken,
as to the right which arises on them of bringing an action.76 In this re-
gard he noted:
[I]t is argued that written scandal is more generally diffused than
words spoken, and is therefore actionable; but an assertion made in
a public place, as upon the Royal Exchange, concerning a merchant
in London, may be much more extensively diffused than a few
printed papers dispersed or a private letter: it is true that a newspa-
per may be very generally read, but that is all casual.77
Similarly, a Select Committee of the House of Lords concluded in 1843
that the distinctions between libel and slander as well as the distinction
between regular slander and slander per se, which are quite peculiar to
the Law of England, do not rest on any solid Foundation.78 As such, the
esteemed committee concluded that wherever an Injury is done to Char-
acter by Defamation there ought to be Redress by Action.79
When one of the witnesses before the Select Committee was asked
whether there should be a legal distinction between words that were spo-
ken and words that were written, the witness John Borthwick, Esq.a
seasoned Advocate at the Scottish Bartestified:
76 Supra note 6 at 364.
77 Ibid at 365.
78 UK, HC, Report from the Select Committee of the House of Lords appointed to consider
the law of defamation and libel and to report thereon to the House, with the minutes of
evidence taken before the committee, and an index, sess 1843 (513) at iii.
79 Ibid at iv.
22 (2010) 56:1 MCGILL LAW JOURNAL ~ REVUE DE DROIT DE MCGILL
I do not think there ought; and, in confirmation of my Opinion, I
may perhaps be allowed to refer, it being Matter of moral as well as
legal Principle, to the Authority of the great David Hume, who says,
that in particular Circumstances even a Whisper may fly as quickly
and be as pernicious as a Pamphlet.80
More than a century later, Professor Prosser disparaged the libel-slander
dichotomy in part by noting the wildfire spread of oral rumor and gossip
in a small town 81 and the persistence of oral gossip about a citizen’s
misconduct thirty years after the event.82
Professor Prosser was also among the many who have pointed out that
the libel-slander dichotomy itself is less the product of principle than of
unfortunate historical accident.83 Before the seventeenth century, jurisdic-
tion over defamation law in England was parceled out among a number of
competing bodies including local, seigneurial, common law, ecclesiastical,
and royal Star Chamber courts.84 Each of these developed its own rules in
respect of defamation claims that were reflective of its particular jurisdic-
tional role and constraints.85 A perpetual point of jurisdictional tension
that left a lasting mark on defamation law was between the ecclesiastical
and the common law courts (i.e., between the Church and the State). The
ecclesiastical courtswhich had for a long period of time jealously
80 Ibid at 141.
81 William L Prosser, Libel Per Quod (1960) 46:5 Va L Rev 839 at 843.
82 Ibid. See also Nicholas St John Green, Slander and Libel, Book Review of A Treatise
on the Wrongs called Slander and Libel, and on the Remedy by Civil Action for those
Wrongs, 2d ed by John Townsend (1872) 6:4 Am Law Rev 593.
83 JH Baker, An Introduction to English Legal History, 3d ed (London: Butterworths,
1990) at 508; Prosser, Libel Per Quod, supra note 81 ([o]f all of the odd pieces of bric-
a-brac upon exhibition in the old curiosity shop of the common law, surely one of the
oddest is the distinction between the twin torts of libel and slander. It is a distinction
unknown elsewhere in the civilized world. Arising out of old and long forgotten jurisdic-
tional conflicts, and frozen into its present form in the seventeenth century by the rising
tide of sentiment in favor of freedom of speech and of the press, it remains a senseless
thing, for which no court and no writer has had a kind word for upwards of a century
and a half at 839 [footnotes omitted]). See also Veeder, supra note 2 at 546; Donnelly,
Proposals, supra note 4; Toelle, supra note 6 at 17; E Hall Williams, Committee on
the Law of Defamation: the Porter Report, Reports of Committees (1949) 12 Mod L Rev
217 at 219-20; Williams v Riddle, 140 SW 661 at 661, 145 Ky 459 (Ct App 1911); Grein
v La Poma, 340 P 2d 766 at 768, 54 Wash 2d 844 (1959) ([i]t is … apparent that the
hodgepodge of the law of slander is the result of historical accident for which no reason
can be ascribed … There ought not to be any distinction between oral and written defa-
mation); Restatement (Second) of Torts 568 cmt b (1977) ([t]his anomalous and
unique distinction is in fact a survival of historical exigencies in the development of the
common law jurisdiction over defamation); MS Marks, Damages for Defamation by
Radio, Notes and Comments (1946) 25:2 Chicago-Kent L Rev 142 at 142.
84 See Veeder, supra note 2.
85 Sipe, supra note 71 at 147.
23
REPUTATION AND FREE EXPRESSION IN CASES OF INTERNET DEFAMATION
guarded their jurisdiction over defamation law using the powerful threat
of excommunication as leveragetraditionally punished defamation with
orders for penance.86 This impelled the common law courts to restrict
their own jurisdiction, through the law of slander, to cases where tempo-
ral as opposed to spiritual damage could be actually proven in the form
of pecuniary loss. This is the historical origin of the obligation to prove
special damages in cases of slander. The categories of slander per se,
which were developed by the courts in a haphazard and ad hoc fashion,
were instances wheredue to the nature of the imputation at issue
temporal damage was presumed.87
Beginning with the well-known case of De Libellis Famosis in 1605,88
the dreaded royal Star Chamber89 regulated written imputations through
the law of libel by importing Roman criminal law. Though initially avail-
able only as a criminal remedy against seditious defamation,90 the law of
libel quickly evolved so as to encompass civil actions, and was not re-
stricted to imputations made against the state or state officials.91 This
new body of law, whichunlike the common law of slanderconsidered
all defamatory imputations to be actionable, was heavily influenced by the
invention of what was then a powerful new medium of communication:
the printing press. This technological advancement, combined with the
ongoing phenomenon of the blood duel as a method for vindication of
reputation, was regarded by the monarchy as being particularly danger-
ous to the stability of the State.92 In this regard it is important to note the
manner in which the law of libel, like the emerging cyber-libel case law
catalogued above, represented a stern legal response to the emergence of
a feared new technology for mass communication that threatened to dis-
86 Veeder, supra note 2 at 557. The usual ecclesiastical punishment for the offence was an
acknowledgment of the baselessness of the imputation, which was made by the defamer
(usually while wearing a white sheet) in the vestry room of the church in the presence
of the clergyman and church wardens, and an apology to the person defamed (ibid at
551-52).
87 Prosser, Libel Per Quod, supra note 81 at 841; Cecil Raitt, Torts-Defamation
Defamation by Broadcast or Defamacast is Actionable Per Se, Note (1963) 9:2 Wayne
L Rev 391 at 391.
88 (1605), 5 Co Rep 125a at 125a, 77 ER 250 (KB), cited in Hill, supra note 7 at para 117
(the case involved an anonymous defendant who had traduced and scandalized the
late archbishop of Canterbury and the then bishop of London).
89 Veeder, supra note 2; supra note 5 and accompanying text.
90 The punishment for libel varied according to the quality of the offence and included
fine or imprisonment, and if the case be exorbitant, by pillory and loss of [the of-
fenders] ears (Veeder, supra note 2 at 565).
91 Hill, supra note 7 at para 113; John W Broadfoot, Defamation in Radio and Television
Past and Present Comment (1963) 15:2 Mercer L Rev 450 at 451-52.
92 Hill, supra note 7 at para 113; Veeder, supra note 2 at 561-62.
24 (2010) 56:1 MCGILL LAW JOURNAL ~ REVUE DE DROIT DE MCGILL
rupt existing power relations in society.93 When the common law courts
finally gained exclusive jurisdiction over all of defamation law in the late
seventeenth century,94 the Frankenstein-like hodgepodge of complicated
and contradictory rules that emerged out of these old jurisdictional con-
flicts (including the libel-slander dichotomy) became a single body of un-
wieldy law.95
Yet despite its accidental origins and the persistence of withering
criticism, the libel-slander dichotomy endured throughout most of the
common law world into the twentieth century. At that point, many be-
lieved that the emergence of mass media such as radio and television
which appeared to truly defy any facile categorization as libel or slander
would finally ring the death knell for the distinction.96 But, often citing
the weight of historical precedent,97 courts and commentators generally
declined to jettison the distinction and instead took up the mantle and
struggled with the awkward task of slotting twentieth century mass me-
dia into one of the two archaic categories by analogy to either writing or
speech.98
93 Viewed in this light, the words of Blair JA take on a new dimension:
Barrick is not the powerful party in the context of the Internet. The impact
of the Internet is to neutralize whatever power Barrick may have had, in
terms of a communication battle with Mr. Lopehandia. In reality it is Barrick
that is vulnerable to publications of this nature, and Mr. Lopehandia who is
abusing his power. The Internet is one of the most powerful tools of commu-
nications ever invented and … is potentially a medium of virtually limitless
international defamation (Barrick Gold, supra note 9 at para 62).
94 The distinction between libel and slander was first drawn expressly by Chief Baron
Hale in King v Lake (1679), Hardr 470, 145 ER 552 (the Court of Exchequer held that,
subject to a few established exceptions, words spoken would not be actionable on their
own, but once written, malice would be presumed and actionable).
95 Sipe, supra note 71 at 147; Reed R Callister, Defamation: Elimination of Distinction
between Libel and Slander as to the Showing of Special Damages, Notes (1959) 33:1 S
Cal L Rev 104 at 105-106.
96 See e.g. Stuart Sprague, Freedom of the Air (1937) 8:1 Air L Rev 30 at 43; Raitt, supra
note 87 at 393; Andrew J Newhouse, Defamation by Radio: A New Tort (1938) 17:4 Or
L Rev 314 at 319; EEM, Television DefamationLibel or Slander?, Note (1956) 42:1
Va L Rev 63 ([t]o continue to apply the archaic distinctions between libel and slander
to modern media of communication is manifestly ridiculous at 74); Marks, supra note
83 at 142.
97 See e.g. Locke v Gibbons, 299 NYS 188 at 192, 164 Misc 877 (Sup Ct 1937) [Locke];
Jones v Jones, (1916) 2 AC 481 at 493 (HL (Eng)).
98 See Melissa A Troiano, The New Journalism? Why Traditional Defamation Laws
Should Apply to Internet Blogs in Bankruptcy 2.0(05): Chapters, Changes, and Chal-
lenges, Comment (2006) 55:1 Am U L Rev 1447 at 1465; Keith Siver, Good Samaritans
in Cyberspace (1997) 23:1 Rutgers Computer & Tech LJ 1 at 4.
25
REPUTATION AND FREE EXPRESSION IN CASES OF INTERNET DEFAMATION
The results of this comparative exercise were often inconsistent and
incoherent. For example many American jurists and commentators, espe-
cially in the first half of the twentieth century, adopted a rigidly formalis-
tic approach, pursuant to which the key criterion in determining the ap-
plicable legal regime for a given medium was whether it involved or could
somehow be tied to writing, in which case the rules of libel would apply, or
to the transmission of sounds, in which case rules of slander would apply.
This analysis produced the distinction between those cases where an im-
pugned statement made during a radio or televised broadcast was made
based on a script, to whichdue to the supposed connection to writing
the rules of libel would apply, and broadcasts of extemporaneous com-
mentary, which were governed by the rules of slander.99
Other American jurists and commentators contended that the control-
ling criterion in classifying statements made using broadcast technology
as being either libel or slander, ought to be the permanence of the medium
in question.100 Under this analysis, if a communication was merely fleet-
ing in nature it was likely less harmful and, as a result, it was argued
that the rules of slander ought to apply to radio or televised broadcasts.101
This meant that radio and television defamationeven if broadcast to
millionswas far easier to defend than simple written defamation. In
recognition of the obvious illogic and unfairness of this approach, many
courts and commentatorsespecially in the latter half of the twentieth
centuryfinally settled upon the capacity for mass dissemination, and
the presumed harm to reputation flowing from such dissemination, as the
controlling criterion in determining whether a statement made using a
particular medium ought to be considered libel or slander. Pursuant to
this analysis, all radio and television broadcasts were to be treated under
99 With regard to radio broadcasts, see Meldrum v Australian Broadcasting Co Ltd, [1932]
VLR 425 (VSC); Locke, supra note 97, affd 2 NYS 2d 1015, 253 AD 887 (1938); Hart-
mann v Winchell, 73 NE 2d 30, 296 NY 296 (Ct App 1947); Charles Parker Co v Silver
City Crystal Co, 116 A 2d 440, 142 Conn 605 (Sup Ct Err 1955). In terms of televised
broadcasts, see Remington v Bentley, 88 F Supp 166 (NY Dist Ct 1949) (the court held,
I accept the [radio] analogy to the extent that it applies to extemporaneous oral expres-
sion, and I feel that the additional factor of pictorial representation along with the
statements adds no more to the form of defamation than would the circumstance of a
great audience in a stadium or the like listening to the spoken word at 169).
100 See e.g. Ostrowe v Lee, 175 NE 505, 256 NY 36, Cardozo CJ (1931) ([w]hat gives the
sting to the writing is its permanence of form. The spoken word dissolves, but the writ-
ten one abides and perpetuates the scandal at 506).
101 Ibid.
26 (2010) 56:1 MCGILL LAW JOURNAL ~ REVUE DE DROIT DE MCGILL
the more stringent rules of libel.102 This common law development has
been codified in most of Canadas provincial defamation statutes.103
Though the difficulties in slotting television and radio broadcasts into
the old categories of libel and slander now appear to be more or less re-
solved, the emergence of Internet communications technologies clearly
has the potential to reopen these old wounds and shine a light once again
on the archaic libel-slander distinction. As a result of the vast and ever-
expanding diversity of applications that are supported by Internet tech-
nology, such a debate could prove far more intractable than that pertain-
ing to radio and television in the twentieth century.104 Fortunately, in part
because the libel-slander distinction has been statutorily abolished in sev-
102 See e.g. Shor v Billingsley, 158 NYS 2d 476, 4 Misc 2d 857 (Sup Ct 1957) as discussed in
Leo L McCormick, TortsDefamationDefamatory Telecast as Libel Rather than
Slander (1957) 26:1 U Kan City L Rev 69; First Independent Baptist Church of Arab v
Southerland, 373 So 2d 647 (Ala 1979); Matherson v Marchello, 473 NYS 2d 988, 100
AD 2d 233 (Sup Ct App Div 1984); McLaughlin v Rosanio, Bailets & Talamo, Inc, 751 A
2d 1066 at 1076, 331 NJ Super 303 (Super Ct App Div 2000). See also Donnelly, Pro-
posals, supra note 4 at 612; Raitt, supra note 87.
103 In those jurisdictions that have abolished the distinction between libel and slander,
there is no need to expressly provide that broadcasting is governed by the rules of libel
(see supra note 12 and accompanying text). Of the three provinces that have not abol-
ished the distinction (Ontario, British Columbia, and Saskatchewan), two (British Co-
lumbia and Ontario) have statutorily deemed broadcasting to be libel. See Libel and
Slander Act, RSO 1990, c L12, s 2 ([d]efamatory words … in a broadcast shall be
deemed to be published and to constitute libel) and Libel and Slander Act RSBC 1996,
c 263, s 2.
104 An example that would represent just the very tip of the iceberg is the debate over
whether blogs, which are but one of the innumerable forms of expression supported by
the Internet, ought to be covered by the laws of libel or slander. It has been argued that
blogs ought to be subject to the law of slander, see Glenn Harlan Reynolds, Libel in the
Blogosphere: Some Preliminary Thoughts (2006) 84:5 Wash U L Rev 1157. In this re-
gard he notes that, unlike newspapers, blog errors can be corrected within minutes
(ibid at 1163). Moreover, blogs according to Reynolds, are unlike traditional mass media
news outletswhich are subject to the law of libelin that they exist in a low-trust
culture such that readers do not generally view particular blogs as being authoritative
(ibid at 1159). Finally, Reynolds argues thatdue to the universal accessibility of the
Internetindividuals that are defamed in blogs have a much greater ability to respond
to accusations made against them than is the case in traditional high cost media such
as newspapers, radio, or television (ibid at 1166). Anthony Ciolli argues the opposite
view, see Defamatory Internet Speech: A Defense of the Status Quo (2007) 25:4 QLR
853. He suggests that Reynolds wrongly assumes that the blogosphere is a culturally
homogeneous community, when in reality the blogosphere is a highly fractured entity
where no one core set of values or norms has been universally accepted (ibid at 854).
Moreover, Ciolli criticizes Reynolds for assuming that the blogosphere is an insular
community and for his suggestion that blogs are not as authoritative as traditional
news media (ibid at 854-55, 859-62). He thus suggests that the law of libel ought to ap-
ply to Internet defamation, including that which takes place in the blogosphere (ibid).
27
REPUTATION AND FREE EXPRESSION IN CASES OF INTERNET DEFAMATION
eral Canadian provinces, which now treat all defamation as libel,105 and in
part because all parties and the courts seemed to have simply assumed
without discussion that Internet defamation should be governed by the
rules of libel (hence the term cyber-libel), this debate has not yet come to
pass.106
Though the awkward debate as to whether Internet defamation is li-
bel or slander may have been temporarily averted, the unfortunate ana-
lytical methodology associated with this archaic common law dichotomy
has, in any event, infected the emerging Canadian cyber-libel jurispru-
dence. As we have seen, in both the areas of the qualified privilege de-
fence and the quantification of damages, the courts have repeatedly relied
105 See supra note 12 and accompanying text.
106 The closest that we have come to such a debate has been in the context of determining
whether a publication over the Internet constitutes broadcasting under the Ontario
Libel and Slander Act (supra note 103). Under s 1(1), broadcasting is defined as the
dissemination of writing, signs, signals, pictures and sounds of all kinds, intended to be
received by the public either directly or through the medium of relay stations, by means
of, (a) any form of wireless radioelectric communication utilizing Hertzian waves, in-
cluding radiotelegraph and radiotelephone, or (b) cables, wires, fibre-optic linkages or
laser beams. If Internet publication was covered by this definition then, pursuant to s
5(1), a plaintiff would be barred from bringing an action unless the plaintiff has, within
six weeks after the alleged libel has come to the plaintiffs knowledge, given to the de-
fendant notice in writing, specifying the matter complained of. In Bahlieda v Santa
(2003), 64 OR (3d) 599, 16 CCLT (3d) 108 (Sup Ct) (cited to OR) the court granted the
defendants motion for summary judgment on the basis that publication over the Inter-
net did constitute broadcasting and the requisite notice had not been given. In so con-
cluding, the court canvassed competing expert opinions as to the nature of the Inter-
net. One expert argued that publication using the Internet was not a broadcast since,
unlike traditional broadcast media, it employs pull technology (ibid at paras 37-38).
This means that the technology is pulled into the users computer when the data is
downloaded. By contrast, traditional radio and television employ push technology,
with the programs being unilaterally produced and pushed to a waiting audience that
simply turns on the radio or television (ibid at para 37). The other expert argued that
Internet is akin to radio and television broadcasting in terms of (1) the immediacy of ac-
cess, (2) its transient nature, (3) electronic distribution in the same manner over the
same physical plant, and (4) the potential to be received by hundreds of thousands of
people (ibid at para 43). The court concluded that [t]he purpose of broadcasting defini-
tion is to single out information which is transmitted to mass audiences, where maxi-
mum harm to reputation can be done (ibid at para 51). Since [t]he Internet, some-
times more than traditional broadcast media … reaches a mass audience, thus placing
material on the Internet, via a website … constitutes broadcasting within the meaning
of the Libel and Slander Act (ibid at para 52). However, this ruling was overturned on
appeal on the basis that the matter was not appropriate for disposal by way of sum-
mary judgment due to the need to resolve conflicting expert evidence (Bahlieda v Santa
(2003), 68 OR (3d) 115, 233 DLR (4th) 382 (CA)). Though the issue has come up on a
number of subsequent occasions, it has never been conclusively resolved, see e.g.
Janssen-Ortho Inc v Amgen Canada Inc (2005), 256 DLR (4th) 407, 199 OAC 89 (Ont
CA); Warman v Fromm, 2008 ONCA 842, 62 CCLT (3d) 246, leave to appeal to SCC re-
fused, 32990 (April 23, 2009).
28 (2010) 56:1 MCGILL LAW JOURNAL ~ REVUE DE DROIT DE MCGILL
upon menacing caricatures of Internet media (e.g., instantaneous, omni-
present, borderless and far-reaching)107 so as to apply a set of legal rules
that are plainly tilted in favour of aggrieved plaintiffs (i.e., no qualified
privilege defence and increased damage awards). In my view, the Internet
demagoguery that has marked the emerging cyber-libel jurisprudence
simply would not have arisen had the common law courts not been so per-
ennially fixated on assessing particular media for their potential to harm
reputation, and then dealing with particular cases differently based on
that assessment.
If the twin values of free expression and protection of reputation are to
be appropriately reconciled in the digital age, the common law courts in
Canada must finally abandon their tendency toward media demagoguery,
and begin to focus on the entire context in which particular defamatory
words were conveyed. Fortunately, the law of defamation in Quebec af-
fords an excellent demonstration of just how such an approach should
proceed.
IV. A More Civil Approach: Defamation Law in Quebec
A brief sketch of defamation law in the Province of Quebec not only
provides further support for the thesis that the hostility toward Internet
publications in the emerging cyber-libel jurisprudence is an outgrowth of
an analytical approach that has its roots in the historical development of
libel and slander law, it also provides a fruitful template for reform of the
common law. Since the law of defamation in Quebec (a civil law jurisdic-
tion) developed largely free from the historical shackles of the common
law libel-slander dichotomy, the courts and the provincial legislature have
never distinguished among different defamatory imputations based upon
the medium used in particular cases. This, in turn, has translated into a
cyber-libel jurisprudence (though that term is not employed in Quebec)
that remains largely free from the use of broad generalizations about the
dangers posed by Internet defamation and the use of legal rules that tend
to favour plaintiffs over defendants. The analysis under Quebec civil law,
on the whole, remains highly contextual and case-specific, where both the
existence of liability and the quantum of damages awarded in any given
case are determined on the basis of all of the relevant circumstances.
Though Quebec civil law does not provide for a specific form of action
for defamation, the general rules that apply to questions of civil liability
as laid down in the Civil Code of Qubec (specifically article 1457)108 do
107 Barrick Gold, supra note 9, cited in Warman, supra note 55 at para 77.
108 Art 1457 CCQ provides, [e]very person has a duty to abide by the rules of conduct
which lie upon him, according to the circumstances, usage or law, so as not to cause in-
29
REPUTATION AND FREE EXPRESSION IN CASES OF INTERNET DEFAMATION
permit the collection of damages for an interference with reputation.109
Pursuant to these general rules of civil liability, a plaintiff must establish
on a balance of probabilities, the existence of an injury, a wrongful act or
fault, and a causal connection between the two.110 For a plaintiff to es-
tablish that they suffered an injury in a defamation case, the plaintiff
must prove that the impugned words were defamatory. Defamatory words
are those that, from the perspective of a reasonable person, would have
brought discredit to the plaintiffs reputation.111 The determination of
fault is a question of fact that must take into account all of the relevant
circumstances.112 A wrongful act may be either malicious or the result of
simple negligence without any intent to harm. In either instance, a civil
fault entitling the victim to reparation will arise.113
Unlike the situation at common law, there is no formal distinction
drawn under Quebec civil law in its treatment of written and oral imputa-
tions. The absence of any such distinction, or indeed any particular atten-
tion given to the medium through which the defamatory words were
communicated, is evident from the following summary of the law by
Beaudoin and Deslauriers:
Any injury to reputation, be it verbal (speech, songs, mimicry) or
written (letter, legal proceeding, caricature, portrait, etc.), public
(newspaper or magazine articles, books, radio or television commen-
tary) or private (letter, pamphlet, report, brief), whether merely abu-
sive or also defamatory, or whether arising from an assertion, impu-
tation, or insinuation, constitutes a fault which, if it gives rise to
damage, must be punished by pecuniary compensation.114
jury to another. Where he is endowed with reason and fails in this duty, he is responsi-
ble for any injury he causes to another person by such fault and is liable to reparation
for the injury, whether it be bodily, moral or material in nature (paras 2-3).
109 See Prudhomme v Prudhomme, 2002 SCC 85 at para 32, [2002] 4 SCR 663
[Prudhomme]; Nicole Vallires, La presse et la diffamation : rapport soumis au minis-
tre des Communications du Qubec (Montral: Wilson & Lafleur, 1985) at 43; Houde c
Benoit, [1943] BR 713 at 720; Socit Radio-Canada c Radio Sept-les Inc, [1994] RJQ
1811 at 1818, [1994] RRA 444 (CA) [Sept-les].
110 Ibid.
111 Prudhomme, supra note 109 at para 34; Sept-les, supra note 109.
112 Prudhomme, supra note 109 at para 38.
113 See Jean Pineau & Monique Ouellette, Thorie de la responsabilit civile, 2d ed (Mon-
tral: Thmis, 1980) at 63-64.
114 Jean-Louis Beaudoin & Patrice Deslauriers, La Responsabilit civile, 5th ed (Cowans-
ville, Que: Yvon Blais, 1998) at para 475 [references ommitted] cited and translated in
Bou Malhab c Diffusion Mtromdia CMR Inc, 2006 QCCS 2124 at para 98, [2006] RJQ
1145.
30 (2010) 56:1 MCGILL LAW JOURNAL ~ REVUE DE DROIT DE MCGILL
Though the common law defences of qualified privilege and fair comment
are not available under Quebec civil law per se, the criteria for these de-
fences are circumstances that must be considered in assessing fault.115
Thus, a defendant can argue that no wrongful act arose on the basis that
they were communicating their opinion on a matter of interest to the pub-
lic, or that there was a duty to communicate the impugned words to the
recipients in question.116
The calculation of damages, like the determination of whether suffi-
cient fault to sustain liability is present, is a highly contextual analysis
that eschews any fixed formulation.117 In addition to any material injury
that can be proven (e.g., loss of income, decreased clientele, etc.) a plaintiff
in a defamation action can collect damages for moral prejudice, which is
roughly the equivalent of general damages at common law. In this regard,
Beaudoin and Deslauriers note:
[T]he analysis of factors influencing the assessment of moral preju-
dice is complex. The first is the seriousness of the act. Is it merely a
discourteous or impolite comment, or is it a full-fledged attack? The
defamers intent, while of no importance in establishing the existence
of a wrong, may be relevant when it comes to assessing the preju-
dice … The dissemination of the defamation is also of some conse-
quence. Widespread publicity should logically justify a more gener-
ous award than publicity within a small circle, unless the circle was
carefully targeted. The condition of the parties, the consequences of
the act for the victim and those close to him, the duration of the at-
tack, and the permanence or impermanence of the effects are also to
be considered. Finally, the existence of an apology or retraction …
may constitute a factor mitigating the prejudice.118
Thus, while the relative scope of publication is a factor in the calculation
of damages, the particular medium that might have been employed is not
itself a relevant consideration.
The relative lack of emphasis placed on the medium of communication
under the civil law can be observed directly in Quebecs emerging Internet
115 Prudhomme, supra note 109 at paras 58-63.
116 Ibid.
117 Aubry v ditions Vice-Versa Inc, [1998] 1 SCR 591 at para 72, 157 DLR (4th) 577.
118 Jean-Louis Baudouin and Patrice Deslauriers, La responsabilit civile, 6th ed (Cowans-
ville, Que: Yvon Blais, 2003) at para 551 [emphasis added, footnotes omitted], cited with
approval and translated in Doucet c Normandeau, 2006 QCCQ 3151 at para 203 (avail-
able on CanLII). See also Galarneau c Azrak, 2008 QCCQ 2647 at para 23 (available on
CanLII); Rizzuto c Rocheleau, [1996] RRA 448 at 456-57, JE 96-989 (Que CS), followed
in Abou-Khalil c Diop, 2008 QCCS 1921 at para 22 (available on CanLII) [Diop], affd on
other grounds 2008 QCCA 1606 (available on CanLII)), revd on other grounds 2009
QCCA 521 (available on CanLII).
31
REPUTATION AND FREE EXPRESSION IN CASES OF INTERNET DEFAMATION
defamation jurisprudence. Though there have been a number of such
cases to date,119 questions of liability and the quantification of damages
have never devolved into any sustained invective about the Internet as
potentially a medium of virtually limitless international defamation120
as has been the case in the common law jurisprudence. Nor have any
hard-and-fast rules that apply specifically to Internet communications
been developed. Two brief examples are sufficient to illustrate the contex-
tual approach to cyber-libel in Quebec.
In Bilodeau c. Savard121 the Quebec Court (Civil Division) awarded
$2,000 to the plaintiff, a consultant who treated compulsive gambling, as
119 See e.g. Lamarre c Allard, 2008 QCCS 5266, [2009] RJQ 89; 3095-4333 Qubec inc c
Service de transport STCH inc, 2007 QCCS 2442, [2007] RRA 731; Lacroix c Dicaire
2005 CanLII 41500 (Que CS) [Lacroix]; Sasseville c Vincent 2004 CanLII 56890 (Que
CS); Gosselin c Vincent [2004] RRA 630 (available on CanLII) (Que CS); Buchwald c
2640-7999 Qubec Inc [2003] RRA 1427 (available on CanLII) (Que CS); Association des
mdecins traitant lobsit c Breton, [2003] RRA 848, JE 2003-1339 (Que CS); Graf c
Duhaime [2003] RRA 1004 (available on CanLII) (Que CS); Caron c Rassemblement des
employs techniciens ambulanciers du Qubec (RETAQ), 2003 CanLII 738 (Que CS),
affd REJB 2004-61201 (WL Can) (Que CA).
120 See Collins, supra note 10. The broadest generalization of such a nature that has ap-
peared in the case law to date is in Lacroix where the court noted that in posting the
impugned words on a website, the defendant a utilis un moyen de communication
puissant, linternet, afin de sassurer de dtruire plus largement la rputation (supra,
note 119 at para 75). Another case in which the elevated potential of the Internet to in-
jure reputation was mentioned is Prudhomme c Rawdon (Municipalit de), 2008 QCCA
1985 (available on CanLII) [Rawdon]. In that case the Court of Appeal entertained an
application for leave to appeal from an interlocutory judgment of the Superior Court
that had dismissed the defendants motion to strike the action of the respondent, the
municipality of Rawdon. The main basis for the motion to strike was an argument to
the effect that it was contrary to the right to free expression in s 2(b) of the Canadian
Charter of Rights and Freedoms for a municipality to sue a person for defamation (Part
I of the Constitution Act 1982, being Schedule B to the Canada Act 1982 (UK), 1982, c
11 [Charter]). In dismissing the application for leave to appeal, Morissette JA noted
some pre-Charter case law that did seem to suggest that a municipality could not sue a
person in defamation. However, the court considered that the use of the Internet by the
defendant might raise a question as to whether that case law was applicable to the case
at bar. In this regard the court observed:
[T]he use and the abuse of recent technology, and in particular of the Inter-
net, by anonymous pamphleteers, may … call for a contextual reconsideration
of the limits of free of speech that is defamatory. Addressing a small crowd in
the flesh and from the north-east corner of Hyde Park is one thing, connect-
ing from a suburban basement, and by means of the Internet, with a vast
and anonymous multitude in the cyberspace may be quite another thing. It
may be that, technologically, the medium, so enhances the message as to
radically alter its impact and oblige its originator to exercise a degree of cau-
tion not expected of orators around Speakers Corner (Rawdon, supra note
130 at para 12) [emphasis added].
121 2007 QCCQ 5127 (available on CanLII).
32 (2010) 56:1 MCGILL LAW JOURNAL ~ REVUE DE DROIT DE MCGILL
a result of defamatory allegations published by the defendant, who was a
relative of the plaintiff. The impugned allegations were published by the
defendant on the plaintiffs own website for one day, and suggested that
the plaintiff had misrepresented the effectiveness of his treatment meth-
ods to the media. The court carefully examined the context in which the
impugned statement was made and concluded that while the defamatory
statement was unmeasured and motivated by malice on the part of defen-
dantwhich led to the requisite faultit had caused only minor damage
to the plaintiffs reputation, likely restricted to his own family circle.122
The fact that the message at issue was transmitted using the Internet did
not play a significant role in the decision.
By contrast, in Diop,123 the fact that the words in question were
transmitted using the Internet did play a role in the final determination
of the Quebec Superior Court. The court awarded the plaintiff, a Senegal-
ese businessman, $100,000 in moral damages and $25,000 in punitive
damages over defamatory articles published on the defendants blog.
These articles falsely alleged, inter alia, that the plaintiff had been ar-
rested at the airport in Paris, France in possession of two suitcases con-
taining eight million euros. In awarding $125,000 in damages, the court
noted that the blog in question was extremely popular; the defendant tes-
tified that he had a loyal readership of 15,000 to 25,000 in Senegal
alone.124 However, the extent of the publication was just one factor in the
overall circumstances canvassed by the court. These included the fact that
the plaintiff was a well-known and well-regarded businessman, that the
defendant had acted deliberately in publishing the impugned articles, and
that the broadly-diffused charges against the plaintiff were serious, false,
and had caused him tremendous ongoing personal embarrassment.125
What the court did not do was to simply rely on a boilerplate assumption
that all materials posted on the Internet are inherently dangerous to
reputation and thus deserving of increased monetary compensation.
As a result of the contextual approach to the establishment of fault
and the quantification of damages in defamation actions, a distinction can
be drawn between Quebec civil law and Canadian common law in the
relative emphasis placed upon the medium through which impugned
messages are communicated. While the common law, beginning with the
libel-slander dichotomy through to the present day cyber-libel jurispru-
dence, has for centuries focused on ascribing broad and often menacing
122 Ibid at paras 17-24.
123 Supra note 118.
124 Ibid at para 17.
125 Ibid at paras 18-30.
33
REPUTATION AND FREE EXPRESSION IN CASES OF INTERNET DEFAMATION
generalizations to particular media, and attaching significant legal conse-
quences as a result, there has been no such reasoning under the civil laws
more contextual methodology. With its focus on all the relevant circum-
stances in each case, and not on any untenable generalizations about the
danger to reputation posed by particular media, the approach of the Que-
bec courts to defamation law appropriately reconciles the twin values of
free expression and individual reputation, which apply with equal force in
that jurisdiction.126
V. A Prescription for Reform: A Move from Caricature to Context
Quebecs civil law approach to defamation points the way forward to a
more balanced approach to cyber-libel adjudication in Canadian common
law courts. In attempting to transplant the civil law defamation method-
ology into the common law context, the prescription is simple: courts must
treat each and every cyber-libel case on its own merits without relying on
any sweeping generalizations about the broad reach and dangers of Inter-
net publications. This applies both to the availability of the qualified privi-
lege defence and to the quantification of damages.
A. Qualified Privilege: A Caseby-Case Assessment
That an impugned publication was made using the Internet should
noton its ownhave any effect on the availability of the qualified privi-
lege defence. In determining whether the privilege associated with a par-
ticular occasion has been lost in any given case, a court should examine
the impugned Internet publication with a view to determining approxi-
mately how many times it was actually accessed and by whom, as well as
the extent to which the likely audience had the requisite interest in re-
ceiving it.
Courts must avoid assumingas was done in Christian Labour Asso-
ciation and Anglethat because a publication was posted on the Web it
necessarily reached an audience that did not have the requisite interest. If
the publication at issue happens to be concerned, for example, with an ob-
scure topic or was posted to an obscure website, courts should not shy
away from concluding that in all likelihood, the number of uninterested
personsas defined by the particular occasion at issuewho accessed the
page, was only reasonably incidental such that the privilege attaching to
the occasion was not exceeded.127
126 Prudhomme, supra note 109 at paras 38-44.
127 See supra note 60 and accompanying text.
34 (2010) 56:1 MCGILL LAW JOURNAL ~ REVUE DE DROIT DE MCGILL
Parties to cyber-libel actions should assist the courts in this regard by
adducing as much evidence identifying the likely scope of the impugned
publication as is practically available. Expert-opinion evidence may be of
particular utility since the mechanics of the Web and other Internet ap-
plications will often be outside the ordinary knowledge of many judges.
Educating the court about the nature and scope of the specific online pub-
lication at issue might also reduce the chance that the availability of the
qualified privilege defence will be denied based on nothing more than
broad generalizations about the dangers of Internet defamation.
B. Quantification of Damages
Just like the qualified privilege defence, the calculation of damages in
cyber-libel cases should be undertaken on a case-by-case basis, taking into
account all of the relevant circumstances. Courts should strive to eschew
any preconceived notions as to the potential dangers posed by defamatory
statements conveyed using the Internet or any other medium. Rather, the
focus should be on the reality of the defamatory statement that is actually
before the court. Did it actually reach a wide audience? If it did, was it the
sort of statement that would have been given any credence by reasonable
persons in that audience? Such an approach would implicitly recognize
the wide diversity of communications and speakers on the Internet, many
of which involve communications among very small groups of individu-
als.128
128 The trial level decision of Swinton J in Barrick Gold represents an excellent example of
the approach advocated here (see supra note 39). A similar example from England is
Keith-Smith v Williams [2006] EWHC 860 (QB) (available on BAILII) in which the
plaintiff, a prominent politician, launched successful defamation proceedings against
the defendant who had posted repeated claims about the plaintiff on a publicly accessi-
ble Yahoo! discussion board with a membership of approximately 100. In these posts
the defendant claimed that the plaintiff was a nonce, a sexual offender, a racist bigot,
a Nazi, and that he had sexually harassed a female colleague. In awarding damages to
the plaintiff in the amount of 10,000 MacDuff J reasoned:
The published statement upon which reliance is placed in the pleadings is
clearly seriously defamatory. In once sense those statements have been made
to a restricted audience. In fact, it is very likely that few people have read
those statements. However, they were available to the whole world, at least
to that part of the world which has a computer or access to a computer and
knows how to go on to the internet and to find the various sites upon it. But
fortunately, as a matter of fact, few people have likely picked up on these de-
famatory statements and I suspect that many of those who have read them
dismissed them as being the rantings of a person who was not to be believed
(at para 17).
REPUTATION AND FREE EXPRESSION IN CASES OF INTERNET DEFAMATION
35
C. Summary
The ruined reputations of innocent parties that are defamed using
Internet media should never be casually shrugged off as unavoidable road
kill on the information superhighway.129 Nor should the suppression of vi-
brant online speech, which promotes many of the core values underlying
the constitutional protection of free expression itself,130 be countenanced
by the courts based on facile generalizations about the Internet and its
well-known capacity for unleashing virtually limitless international
defamation.131 The way forward is for common law judges to jettison their
age old propensity for becoming distracted by the medium in which im-
pugned words were communicated, and to instead focus on examining the
full context of each case so as to assess the nature and scope of the as-
serted attack on a plaintiffs reputation. Such an approach, which would
adopt the spirit of the civil law of defamation in Quebec, is the best way to
reconcile the competing constitutional values of freedom of expression and
vindication of reputation in accordance with the dictates of the Supreme
Court of Canada.
Conclusion: Grant v. Torstar Corp.A Potential Turning Point?
While it does not represent a fundamental shift in the jurisprudence
analyzed above, the recent decision of the Supreme Court of Canada in
Grant may signal the beginnings of an important judicial retrenchment
from both the stance of hostility toward Internet communications in par-
ticular, and the more general fixation on the medium in which the im-
pugned communication was made in defamation cases.132
In short, the Court in Grant reassessed the defence of qualified privi-
lege and concluded that it offered insufficient protection for free expres-
129 This phraseology is borrowed from WIC Radio, supra note 59 (the Court held that [a]n
individuals 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 per-
sonal reputation be permitted to chill freewheeling debate on matters of public inter-
est at para 2).
130 The Supreme Court of Canada has held that these values are threefold: (1) seeking and
attaining truth, (2) encouraging participation in social and political decision making,,
and (3) cultivating individual self-fulfillment and human flourishing. See Irwin Toy Ltd
v Quebec (AG), [1989] 1 SCR 927 at 976, 58 DLR (4th) 577; R v Sharpe, 2001 SCC at
para 23, [2001] 1 SCR 45; Montral (City) v 2952-1366 Qubec Inc, 2005 SCC 62 at para
68, [2005] 3 SCR 141. For a discussion of the connection between the protection of free
expression under the Charter (supra note 120) and the use of Internet technology see C
Ian Kyer, Technology and Personal Freedom (2007) 56 UNBLJ 36.
131 Collins, supra, note 10.
132 Supra note 13.
36 (2010) 56:1 MCGILL LAW JOURNAL ~ REVUE DE DROIT DE MCGILL
sion and was thus inconsistent with the applicable Charter values.133
Chief Justice McLachlin, writing for the Court, found thatespecially in-
sofar as news media outlets are concernedthe need to establish a recip-
rocal duty as between the defendant and the public at large, unduly lim-
ited freedom of expression in favour of protecting reputation.134 The Court
thus created a new defence of responsible communication on matters of
public interest that will be available in respect of assertions of fact on a
matter of public interest where the defendant was diligent in trying to
verify the allegationhaving regard to an open list of enumerated consid-
erations.135
Though the defendants in Grant and its companion case of Cusson v.
Quan136 were part of a traditional media outlet, the Supreme Court of
Canada considered whether the new defence would also apply to those
who were not professional journalists. In concluding that the defence
would be open to anyone who publishes material of public interest in any
medium,137 the Court noted that
the traditional media are rapidly being complemented by new ways
of communicating on matters of public interest, many of them
online, which do not involve journalists. These new disseminators of
news and information should, absent good reasons for exclusion, be
subject to the same laws as established media outlets.138
The Supreme Court of Canadas approach to this new defence is com-
mendable in a number of ways. By explicitly making the defence available
to communications made in any medium, the Court borrowed a page from
the civil law textbook and eschewed the improper focus on the mode of
communication that has bedevilled the common law for so long. The test
that the Court adoptedwhich assesses whether the defendant acted re-
sponsibly by looking to factors such as the seriousness of the allegation,
the public importance of the matter, its urgency, the status and reliability
of the source, and any other relevant circumstances139introduces pre-
cisely the sort of medium-blind context sensitivity into the analysis that I
have advocated above. Finally, the Court should be applauded for avoid-
133 Charter, supra note 120.
134 Grant, supra note 13 at paras 38-40.
135 Ibid at paras 97, 126.
136 2009 SCC 62, 314 DLR (4th) 55.
137 Grant, supra note 13 at para 96 [emphasis added], citing Jameel v Wall Street Journal
Europe Sprl, [2006] UKHL 44 at 54, [2007] 1 AC 359.
138 Ibid.
139 Ibid at para 126.
37
REPUTATION AND FREE EXPRESSION IN CASES OF INTERNET DEFAMATION
ing the kind of demonization of Internet communications that has marred
much of the emerging cyber-libel jurisprudence to date.
However, the Court in Grant did not eliminate the qualified privilege
defence, it merely opened up a new defence for defamatory statements of
fact that are made on matters of public interest. Where the subject matter
of an online publication cannot be slotted into the nebulous public inter-
est category (i.e., a subject matter inviting public attention, or about
which the public has some substantial concern because it affects the wel-
fare of citizens, or one to which considerable public notoriety or contro-
versy has attached 140), the antiquated rules associated with the qualified
privilege defence will still apply. Similarly, because the Courts ruling did
not address the issue of damages, this area of the law remains improperly
calibrated, as I have argued above.
Ultimately, whether or not the Grant decision represents a fundamen-
tal shift in the common law courts attitude toward defamation law re-
mains unknown, both in terms of the age old propensity to focus far too
much on the medium in which the impugned publication was made, and
the more recent judicial hostility toward Internet communications. Never-
theless, the ruling suggests that with the libel-slander dichotomy receding
into the rear-view mirror in most Canadian jurisdictions, the courts may
yet begin to nurture the enfant terrible of the [common] law141 into digi-
tal age maturity.
140 Ibid at para 105, citing Brown, Defamation, supra note 27 at s 15.5(1) [footnotes omit-
ted].
141 Donnelly, Proposals, supra note 4.
