From Sisyphus’s Dilemma to Sisyphus’s
Duty? A Meditation on the Regulation of
Hate. Propaganda in Relation to Hate
Crimes and Genocide
Jean-Frangois Gaudreault-DesBiens”
The author examines central legal and philosophical is-
sues pertaining to the eguation of hate speech. In partieular. he
evaluates the competing perspectives of the “cansationist” ap-
proach, which requires a direct causal link between the expres-
sion it purports to regulate and the harm it allegedly causes, and
the “correlationist” approach, which would regulate hate ex-
pression based on a rational correlation between the expression
and the harm. In contrast, the correlationist approach adopts a
preventive logic that seeks to structure attitudes by enforcing
positive norms. Afte examining the theoretical underpinnings
of these views, and reviewing their legal and philosophical pit-
falls-paticularly
in their extreme forms–the author ulti-
mately favours the correlationist approach to hate speech regu-
lation. Civil society and a democratic tradition vill prevent this
type of regulation from leading down a slippery slope to state
censorship. To avoid undue limitations to freedom of expres-
sion, however, only extreme hate expression should be regu-
lated, that is, abusive expression, which is distinct from offen-
sive expression in that it targets prisons rather than ideas. There
is no optimal way to balance equality and freedom of expres-
sion, nor to address the challenges that the enforcement of ham
speech regulation entails. Analogizing with the mylh of Sisy-
phus, the author refers to these challenges as the dilemma of the
“Sisyphus state”, concluding that this dilemma becomes a duty
to regulate against abusive forms of expression, because a con-
stitutional democracy cannot tolerate radical denials of the hu-
manity of some of its citizens.
Lautur examine I
rzircips
qcr-stioaa l
‘t
n/rcttiea d: la pxt-
.rs d’utn
‘t
l:
l’
philosophiquca s ulewls par la r
ganda haineus. en particulier I dizzt err tn jrnt
approche qui requie un lien caW dirtet erae l’espe
ct
le dommage, ct cetux d’une ‘p;eodLr qui tc e oa d’tta car-
r dtaion rationnre entre iexpresiva Iht.usaz ct I: dxnag,
Cene demihre approcht c .p:r u-e ILq pirtere qa
cherche a stnrcturer le atitudes dow=,
asrrea
de normcs positives. Apr8 exann des fo:ci=rnt raa zcIb
et des dangers qui guettent d”ra= d m &u pvfirrs, sur-
tout lorqu’das pren t de frma extrcir,
J
pari en favetr d la sccon- caproci.t, b e
r la cosri:la
rationnell. La socid’i cisile ct la trzditio d
erxra#i a affi-
sent a pr~idenir la dSnirestece d- cem po!sduai
&atique. Toutefois, afr d’vir d’uypsor des lns rv=i
cx-
‘clxpar=rra htr==
cessives h la libmsd d’exrrssio.z,
abusive –
offm-
sante en cc qu’elle cible des prerce pl!Zt qu:an Wes
in –
dzvrait &re n1gleantf . Ine
Ien
a
J. c de r-
idle d co ncilicr l’galit et libeT de* xpnsao
soudr let pro!wNb es soulevis er I’c pp =ia da lois porlant
un dmrna carre
sur laproagand haineus, Vliza fait fac
son devoirdi rglemnster Iexpretneasive cn ta difi.auti s
-,, il fait ar s f=e a E=a
inhientes hun tel exereica. Tel Sis
car use ditnocrari: ra pnt tio-
tiche potentieliareent infnie –
ler une n”aion r,.dicale d: I’humanii trzr.- da- cvztxr3 da
ses citoyens.
qui sw distinguz dI exr=-Da sirmp!=ne
bl- p-a y a% or da
cem-
. Assistant Professor, Faculty of Law and Institute of Comparative Law, McGill UnKersity. I would
like to thank my research assistant, Danielle 1iller, Ph.D., for her thoughtful comments, as well as for
having patiently perfected my written English. This paper was written for a panel on “Hate Speech,
Hate Crimes, Genocide’ at the international conference Hate Genocide and Human Rights Ftfoy
Years Later: Mhat Have We Learned? What Must We Do? (Faculty of Law, McGill University, 28
January 1999). This is the final version of this paper. Due to a production error, the version printed as
(2000) 46 McGill LJ. 121 was not the final version.
McGill Law Journal 2001
Revue de droit de McGill 2001
To be cited as: (2001) 46 McGill LJ. 1117
Mode de r6f6rence : (2001) 46 R.D. McGill 1117
1118
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 46
Hate speech raises fundamental issues from legal, philosophical, and epistemo-
logical standpoints. It prompts us to think about individual and collective incarnations
of hatred, how we apprehend this social phenomenon, and most important, how we
characterize its dissemination. This article meditates on both the limits and the prom-
ises of regulations of hate propaganda, and more generally, of law itself. But why a
“meditation”? Essentially because meditating implies a certain openness to unfore-
seen questions that arise, pgle-mgle, in the course of the meditation itself. This ex-
plains why this written “meditation” is structurally and formally different from a more
traditional essay, where one reaches c by way of a and b, and where one generally
wishes to convince someone of something. As such, an essay may be seen as an at-
tempt to tame doubt and to provide certitude, whereas a meditation raises questions
more than it does anything else.
While “hate speech” may aptly describe any word or utterance intended to injure,
degrade, denigrate, or ridicule people on the basis of a distinguishing feature, this ex-
pression fails to capture what often leads to hate crimes and to genocide: the system-
atic, rather than individual, use of hateful discourse and the systemic nature of hatred
that sometimes ensures its social acceptability. In such cases, hate speech, or hate
propaganda, as I prefer to call it, is ingrained in a system where the social degradation
of the Other plays a central role in political discourse. In fact, hate propaganda con-
tributes in and of itself to the creation of an imaginary Other. Dehumanized and de-
personalized, depicted as threatening and as a potential enemy, this Other is indeed
likely to become the enemy for those influenced by such propaganda. Huge conse-
quences may follow. Depending on the circumstances, hate propaganda may create an
environment where hate crimes are considered no different from other crimes. In such
a case, what inspires them, that is, hatred, becomes obscured, thereby trivializing their
qualitatively different nature. Worse than their being trivialized, however, hate crimes
may sometimes become socially acceptable, and ultimately, desirable. From individ-
ual and isolated hate crimes committed in the private sphere, we move to more wide-
spread incitement to hate, and finally we reach the realm of publicly-supported mass
hate crimes–or genocide as public policy, to put it bluntly. This explains the nexus
between hate propaganda, hate crimes, and genocide, a nexus that has too often sur-
faced during the twentieth century, especially, but not exclusively, in countries where
democratic traditions were young and civil society weak. This nexus may be ac-
knowledged by the socio-political realm, but not always by the legal one.
The juridical apprehension of the nexus between hate propaganda, hate crimes,
and genocide stirs controversy in intellectual circles, especially in North America.’ In
‘I am referring more specifically to the debate that rages in the United States between some liberal
and libertarian jurists, on the one hand, and critical race theorists, as well as some feminist legal
2001]
J.-F GAUDREAULT-DESBIENS – REGULATING HATE PROPAGANDA
1119
particular, the question of the relative performativity” of some forms of expression, in-
cluding racist speech, remains at the centre of philosophical and epistemological de-
bates about the appropriateness of regulating these forms of expression. Can speech
alone be characterized as assaultive, causing harm in and of itself?. If yes, should we
regulate it? Assuming that we should, how do we do so?
Although these questions are important, because acknowledging the possibility
that speech alone may cause harm may ultimately induce us to rethink the classic di-
chotomy between speech and conduct, the legal apprehension of the nexus between
hate propaganda, hate crime, and genocide requires no specific reference to the theory
of performative speech? Indeed, we need not ask ourselves: Should we consider this
form of speech as an act or conduct because of the harm it causes in its own right?
Understanding this nexus requires, instead, a reflection on the degee of causation
constitutionally required to legitimize restrictions to freedom of expression in view of
regulating hate propaganda. Should the law necessitate a close and direct link between
the expression it purports to regulate and the harm that this expression allegedly
causes? A direct link imposes upon the regulator the burden of demonstrating the ex-
istence of “a clear and present danger” or an “imminent lawless action”, to borrow
from American legal terminology,’ as a consequence of the expression it purports to
regulate (the “causationist” approach). In other words, the message conveyed must
immediately induce one to act on it. The contrasting view holds that, for a law to be
constitutionally permissible, it is enough that it acknowledge a strong rational corre-
scholars, on the other, about the constitutionality of hate speech regulation and the interplay between
freedom of speech and equality rights.
2 In a nutshell, the notion of performativity designates the capacity, in some circumstances, of an
utterance to do something simply by being uttered. For example, when an officer of justice who is le-
gally empowered to celebrate weddings utters, during a wedding, “I declare you husband and wife,”
the immediate consequence of this declaration is precisely to give truth to the utterance that the man
and the woman now are, indeed, husband and wife. A performative utterance, that is, an utterance that
performs something–hence its characterization as a “speech act”-must be distinguished from a
merely descriptive utterance such as ‘This is a cow,” which does not perform anything and which
does not become true simply because it was uttered, since the person who makes that utterance could
wrongly take another animal for a cow. That said, relying on the notion of performativity, some argue
that the utterance of hateful words alone inflicts harm on targeted listeners. On performativity, see I.
Austin, How to Do Things with librds (Cambridge, Mass.: Harvard University Press, 1962); .1 Searle,
“What Is a Speech Act?” in M. Black, ed., Philosophy in America (Ithaca, N.Y.: Comell University
Press, 1965) 221; R. Nadeau, %cabulaire technique et analytique de l’6pistuniologie (Paris: Presses
Universitaires de France, 1999) s.at “performatif/constatatif’.
3 Indeed, it has been argued, quite convincingly I must say, that the “performative speech” theory
alone offers a rather shaky foundation for hate propaganda regulations. See NV. Sadursky, “On ‘Seeing
Speech through an Equality Lens’: A Critique of Egalitarian Arguments for Suppression of Hate
Speech and Pornography” (1996) 16 Oxford J. Legal Stud. 713.
‘For a legal history of the evolution of these tests, see L.H. Tribe, American Constitutional Lm;, 2d
ed. (Mineola, N.Y.: Foundation Press, 1988) at 841-49.
1120
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 46
lation between the expression and the harm that it presumably causes, and that it act
upon this acknowledgement (the “correlationist” approach).
While the causationist approach is based on a curative logic of imputation, the
correlationist approach is inspired by a preventive logic of risk management. In a way,
the heuristic process implied by the causationist approach requires a quasi-
photographic capturing of the transition between a primary expressive state, concep-
tualized as essentially intangible, and a secondary behavioural state, conceptualized as
essentially tangible. As Bollinger puts it, “[T]he tests of the ‘clear and present danger’
genre appear to focus on a very narrow range of potential social harm from speech-
that is, on the risk that the audience will be persuaded to turn from talk to action of a
kind deemed especially harmful to the society.”‘ As such, the causationist approach
implies a microscopic examination of a segment of social reality that excludes from
its consideration any form of harm that is not immediate and entirely tangible.’ Be-
cause it emphasizes the demonstration of a direct and almost unmediated causal link,
this approach can surely be characterized as profoundly juridical in the classical
sense, in a way akin to the approach in the law of torts.
It could be argued, however, that this profoundly juridical approach has been un-
duly influenced by a rather dated theory concerning the logic of scientific discovery,
which postulates that the pursuit of a certain scientific method may actually lead to
the discovery of absolute scientific certitudes or truths. In keeping with this view, sci-
entific knowledge comprises what has been definitely proven. For example, if a hy-
pothesis, say that mixing element a with element b causes reaction c, has been duly,
that is, scientifically, verified, it should be considered as proven and therefore abso-
lutely true. That said, most scientists would now treat the notion of absolute scientific
certitude as an oxymoron. Indeed, in line with Karl Popper’s theory of knowledge, the
scientific method is now described negatively rather than positively; instead of being
depicted as implying the (positive) verification of a hypothesis, thereby giving birth to
a scientific certitude, it is now presented as involving constant attempts to falsify the
dominant hypothesis that will remain the dominant theory only insofar as it can resist
these falsification attempts. Superficially, the difference between these two views may
seem tenuous, but it is not. Quite the contrary: focussed as it is on refutation instead of
verification, Popper’s theory renders doubt unavoidable, thereby making its manage-
ment one of the most important issues to address, if not the most important. To use
anthropomorphic language, scientific theories now live with the knowledge of their
own internal frailty. As a result, even in the “hard” sciences, the imperium of strict
‘ L. Bollinger, The Tolerant Society: Freedom of Speech and Extremist Speech in America (New
York: Oxford University Press, 1986) at 179.
‘ Hence the non-consideration of most non-physical forms of harm. Hence, also, its intrinsic inabil-
ity even to acknowledge the possible performativity of some forms of speech.
2001]
J.-F GAUDREAULT-DESBIENS – REGULA77NG HATE PROPAGANDA
1121
and linear images of causation has faded away, leaving more room for other models of
causation.
If this is acknowledged in the scientific realm, it should also be acknowledged in
the social sciences, where the part played by “givens”, assuming that such “givens”
exist, is much smaller than that played by “constructs”. This is a fortiori the case with
law, at least positive law, which can certainly not be characterized as a science.’ Thus,
when examining issues pertaining to causation, we should never forget that legal
knowledge is fundamentally different from scientific knowledge. While the idea of an
entirely “pure” fact is hardly possible, in that it would exclude any human mediation
in the construction of the problematics surrounding factuality itself, the level of factu-
ality of phenomena dealt with in the legal field is quite different from that of the pure
or applied sciences. As Feldman points out:
The phenomena identified by legal concepts such as negligence, fraud, and
rape are not like the phenomena identified by natural scientific concepts like
proton or gravity. Negligence, fraud, and rape are not human-independent phe-
nomena; they are constituted by our practices, goals, values, and beliefs!
As such, we should never forget that from a legal standpoint there is no entirely natu-
ral or objective way to look at causation. This is why it is worthless to invoke the
law’s neutrality to hide that any decision taken in the legal field as to the degree of
causation required in a particular case necessarily implies a policy choice. This choice
is made either by the legislator or by the judiciary, both of whom reflect the normative
values of the society in which the choice is made.
While causation is more than politics, in that it sometimes deals with material
facts the existence of which cannot be doubted, its political dimension should not be
obscured. Far from being a pure given, causation is first and foremost a construct.
This is why legal reflection should never abdicate outright to “scientific” evidence,
whether it comes from the hard or social sciences. The law must remain master of the
processes through which it apprehends issues pertaining to evidence. It should not
systematically and naively yield to other forms of “authority”, even scientific, ac-
cepting these constructions as “givens” for legal purposes. This should be borne in
mind when we reflect on competing logics such as the curative logic of imputation
and the preventive logic of risk management.
A number of theoretical and practical consequences flow from the initial ac-
knowledgement of the perils associated with blind adherence to “scientific data” or
“scientific causation” and their use as a possible justification for the elaboration of a
7Law can be characterized in many ways, but certainly not as a science when “law” describes posi-
tive law. This is not to say, however, that law cannot be scientifically studied, viz, from an external
perspective.
8 ILL. Feldman, “Objectivity in Legal Judgment” (1994) 92 Mich. L Rev. 1187 at 1188-89.
1122
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 46
given norm. For example, it becomes clear that the strict causal link that the curative
logic of imputation requires between the expression to be regulated and the harm al-
legedly caused by that expression forces the potential victims of hate propaganda to
bear or absorb all risks. This may be justified sometimes in the more individualistic
context of tort law, but is it justified in the more societal context of constitutional law?
This causationist approach stands in contrast to the correlationist approach. By
considering social reality from a macroscopic angle, the correlationist approach ac-
knowledges less tangible forms of harm, and more important, considers as sufficient
the demonstration of a strong rational correlation between a given expression and the
harm it allegedly causes. The correlationist approach allows for a higher level of def-
erence towards regulatory attempts to manage the risks of harm that presumably flow
from certain forms of expression because it is based on a theory that acknowledges
links that are rational and plausible from a global societal angle rather than only di-
rect, legal-causationist links. As such, the correlationist approach is inspired by a pre-
ventive logic that can only be implemented if the traditional legal concept of causation
is expanded, or more accurately, imploded. Pushed to its extreme, however, the cor-
relationist approach sometimes refuses unduly to acknowledge the risks associated
with censorship, thereby potentially threatening all speakers who try to convey un-
popular messages. This is a risk that the slippery slope argument captures. But when
used to condemn any attempt at regulating hate propaganda, the same argument fails
to acknowledge that the correlationist approach may simultaneously acknowledge the
risk associated with censorship while acknowledging the risk associated with the ab-
sence of any legal constraints on some forms of speech. When both forms of risks are
considered, the correlationist approach may well involve a much more complex
analysis than its causationist counterpart. It may also inform a different approach to
the interplay between freedom of expression and equality, the two rights at stake in
the regulation of hate propaganda.
The curative logic of imputation, as well as the preventive logic of risk manage-
ment, are illustrated respectively by the American and Canadian positions on the con-
stitutionality of laws prohibiting hate propaganda
While these two logics may explain differences in how countries approach the
regulation of hate propaganda from a domestic constitutional standpoint, the distinc-
tion between these logics can also prove a useful tool with which to assess potential or
actual obstacles to the enforcement of international norms prohibiting hate propa-
9 In fact, it is to be noted that the American position on the unconstitutionality of hate propaganda
regulation is exceptional; Canada’s position, for its part, is in line with the rest of the world. See e.g.
I. Coder, “Hate Speech, Equality, and Harm under the Charter Towards a Jurisprudence of Human
Dignity for a ‘Free and Democratic Society”‘ in G.-A. Beaudoin & E. Mendes, eds., The Canadian
Charter of Rights and Freedoms, 3d ed. (Scarborough: Carswell, 1996) 20-1.
2001]
J.-F GAUDREAULT-DESBIENS- REGULATING HATE PROPAGANDA
1123
ganda. For example, aside from the fact that a direct prohibition of hate propaganda
would contradict a basic tenet of the First Amendment creed, that is, viewpoint neu-
trality, the causationist approach adopted by American law in its analysis of speech-
related harms precludes the United States from fully subscribing to these international
norms, and as a result, from participating in their enforcement. Indeed, how can the
United States subscribe to international norms that purport to outlaw hate propaganda
if its own domestic law prevents its government from doing so? How can the Ameri-
can government accept that legal consequences be drawn from the fact that a system-
atic campaign of hate propaganda often, but not always, serves as a precursor to geno-
cide, or in other words, that such a campaign only may serve this purpose? How can
the United States then participate in the enforcement of international norms that are
based on a prior acknowledgement by the international community of a rational cor-
relation between a given form of expression and some of its consequences-the
“mildest” including discrimination, hostility, and ostracism directed at the stigmatized
group and the worst being genocide? In short, the United States is extremely reluctant
to do so.” This reluctance is troublesome because the United States is the only super-
power in the post-cold war era and so it often finds itself in a position to influence the
outcome of events. This reluctance is especially troublesome in light of the ethnic
cleansing and genocide in Bosnia and Rwanda, and the ethnic cleansing that recently
took place in Kosovo. It is even more troublesome considering the well-documented
role that hate propaganda played in the first two cases.”
At an even deeper level, any inquiry into the nexus between hate speech, hate
crimes, and genocide forces us to reflect on the realization of basic human rights fifty
years after they were solemnly entrenched in the Universal Declaration of Human
Rights.'” More specifically, it raises the following question: To what extent can the
regulation of hate propaganda be linked to the realization of human rights? In the next
few pages, I will modestly propose some intellectual parameters within which to ap-
proach this question. Addressing it will inevitably demonstrate how the two funda-
mental rights at stake in the regulation of hate propaganda, freedom of expression and
equality, interact. I will use two well-known cultural icons as springboards for my ob-
” See eg. U.S. formal reservations on the application of the International Com’ention on the Elimi-
nation of All Fonns of Discrnination, GA Res. 2106(XX), UN GAOR, 21 December 1965, 660
U.N.T.S. 195, arts. 4,7, and of the International Covenant on Civil and Political Rights, 19 December
1966, 999 U.N.T.S. 171, art. 20, Can. T.S. 1976 No. 47 (entered into force 23 March 1976, accession
by Canada 19 May 1976).
” See K. Mahoney, ‘Iate Speech-Ethnic Cleansing-in the Balkans” (Intemational Conference
Hate Genocide and Hwnan Rights Fifty Years Later: What Have 1%I Learned? What Must M? Do?,
Faculty of Law, McGill University, 28 January 1999) [unpublished]; NV.A. Schabas, “Hate Speech in
Rwanda. The Road to Genocide” (2000) 46 McGill LJ. 141.
12 GA Res. 217(m), UN GAOR, 3d Sess., Supp. No. 13, UN Doc. A/810 (1948) 71.
1124
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 46
servations: first, the mythical story of Sisyphus, and second, the motto of the French
Revolution and now of France, “Liberty, Equality, Fraternity”.
First, let me deal briefly with Sisyphus, to whom I will return at my conclusion.
In Greek mythology, Sisyphus has been condemned by the gods to endlessly pushing
a stone up a mountain, only to see it fall down again and again upon reaching the top.
Sisyphus’s role in Greek mythology strangely evokes the state’s position when faced
with regulating phenomena such as hate propaganda in the techno-scientific era. If
Western governments choose to regulate and enforce laws against hate propaganda,
they face an endless struggle uphill, which is exacerbated by recent technological de-
velopments, such as the Internet, that allow for easier and faster dissemination of in-
formation-any information-to all parts of the world. As a matter of fact, today’s
dynamic of “de-territorialization”” destabilizes the state’s power and ability to regu-
late the flood of available information. Undoubtably, this must induce profound
changes in the way norms emerge and are enforced. Whether we like it or not, states
are ill-equipped to come to terms with this phenomenon of de-territorialization. In
fact, numerous information providers are literally shielded from state normativity as a
result of the intrinsic limitations on the latter’s extraterritorial jurisdiction. Moreover,
new information technologies preclude the state from effectively imposing any water-
tight control over anti-egalitarian expression such as hate propaganda, the regulation
of which was already hard to enforce prior to the emergence of these technologies.
Without a doubt, insofar as the control of information is concerned, today’s regulatory
state looks more and more like Sisyphus with his stone. Hence the Sisyphus state’s
first dilemma is “to regulate or not to regulate”. Assuming a positive answer, the di-
lemma then becomes whether to enforce the regulation.
While the challenges mentioned above must certainly be considered when re-
flecting on the utility of state regulation as a means of preventing the harm caused by
hate propaganda, they cannot be invoked alone either to dismiss any further use of
such regulation or to abandon the very idea of enforcing the already existing norms
against that form of speech. That a legal norm appears ineffective does not mean that
it is actually ineffective. Legal norms do not operate in a social vacuum; thus, their ef-
fectivity may be assessed from microscopic as well as macroscopic standpoints. As
noted by legal feminists and critical race theorists, acknowledging this multiplicity of
perspectives has important consequences. For example, reflection on a legal norm’s
layers of effectivity may push some to rethink their adherence to a causationist ap-
proach for assessing the constitutionality of that norm. More important, it may induce
them to recognize that some forms of expression have harmful effects on certain
” I borrow this expression from R.-J. Dupuy, “Le d6doublement du monde” (1996) 100 Rev. D.I.P.
313 at 317.
2001]
J.-F GAUDREAULT-DESBIENS – REGULA7NG HATE PROPAGANDA
1125
groups and that there is no watertight distinction between expression and conduct.” In
any event, simple acknowledgement of the relevance of such questioning could
change the way we frame debates about the regulation of these forms of expression.
At the very least, it surely affects the way in which we conduct any serious discussion
about the interplay between freedom of expression and equality.
Some epistemological obstacles, however, must be removed before such discus-
sion can occur.” For example, a purely ideological portrayal of the interplay between
freedom of expression and equality may impinge on a complex understanding of this
relationship, thereby systematically allowing freedom of expression to prevail over
equality, or vice versa. As a result, depicting this interplay as an unsolvable conflict
where one value always trumps the other is of no use whatsoever. Indeed, from a
legal-philosophical perspective, there is no clearly obvious or “given” solution to the
appropriate balance between freedom of expression and equality, since neither of
these two rights can be realized in the abstract, despite their status as normative ideals.
Bearing this in mind, any answer to the problem must go beyond the realm of pure
abstraction, which often obscures strictly ideological motives anyhow, to a kind of
“idealistic pragmatism” that considers not only the immediate legal constraints appli-
cable but also the whole socio-legal context in which this balance must be struck. In
any event, two things are certain. First, no solution can ever pretend to be optimal, in
the sense of pleasing everyone, nor can it pretend to be self-evident, in the sense, in
and of itself, of “naturally” convincing all. Second, any solution is likely to be condi-
tioned by the internal and external legal cultures,’6 as well as by the political cultures,”
“I think it is intellectually absurd to say without any nuance, however, that “speech is act”, and this
even if one accepts the possible relative perfonnativity of some forms of expression. The analysis
must always be contextualized.
“- Gaston Bachelard defines the concept of “epistemological obstacle” as a generic concept that
designates all causes of inertia, stagnation, or setback in knowledge. See G. Bachelard, Laformation
de l’esprit scientifique: contribution a une psycanalyse de la connaissance objective, 12th ed. (Paris:
Vrin, 1960) at 14. Frangois Ost and Michel van de Kerchove are more specific, describing it as follows:
L”obstacle 6pistmologique” est une categorie qui, plac~e au fondement d’une thdorie,
tente de la garantir contre les mises en question et infirmations dont cie paurrait faire
robjet: tant6t simple concept indiment dlargi … tant6t vdritable “lltanschatung”,
l’obstacle survalorise la th~orie, refoule les questions irrsolues qu’elle suscite par de-
vers elle et la dote d’un pouvoir explicatif exag&6rment dtendu (Jalons pour une theorie
critique du droit (Brussels: Facults universitaires Saint-Louis, 1987) at 121-22).
16 “External legal culture” refers to the lay people’s legal culture, whereas “internal legal culture” re-
fers to the legal culture of the members of the legal community. See L.M. Friedman, The Legal Sys-
temz: A Social Science Perspective (New York& Russell Sage Foundation, 1975) at 223.
” See generally J.-F Gaudreault-DesBiens, “Du droit et des talismans: mythologies, mtaphores et
libert6 d’expression” (1998) 39 C. de D. 717 [hereinafter “Du droit et des talismans”].
1126
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 46
at play, and by the audiences that the interpreter seeks to address.” This is true for the
interplay between freedom of expression and equality, as it is for the specific question
posed by the constitutional understanding of juridical attempts to regulate hate propa-
ganda or other forms of anti-egalitarian expression.
One could legitimately ask, however, why there is no obvious solution to the
problem posed by the regulation of hate propaganda. The answer lies first in the
Manichaean and abusive use of the two previously identified logics-the curative
logic of imputation and the preventive logic of risk management-in the debate sur-
rounding regulation. Pushed to its extreme, the curative logic of imputation prevents
almost any juridical acknowledgement of the harm caused by hate propaganda. This
is, in essence, the position adopted by libertarian jurists who will systematically op-
pose the regulation of hate propaganda unless presented with direct and irrefutable
evidence of its deleterious effects. Their fear of losing the smallest part of liberty
pushes them to deny any legitimacy to state regulation of such propaganda, thus ob-
scuring considerations pertaining to equality. On the other side, when pushed to its
extreme, the preventive logic of risk management overstates the systemic nature of the
harm caused by hate propaganda or any other form of anti-egalitarian expression,
thereby rendering its own philosophy the only determining variable for the solution to
the problem. From this vantage point, hate propaganda’s systemic nature will be in-
voked to legitimize a quasi-total control of anti-egalitarian forms of expression, at the
risk of obliterating all considerations pertaining to freedom of expression. Both sides,
however, are wrong; social reality is too complex to be imprisoned by any “pure”
logic designed to be universally applied. Indeed, the way these two logics are applied
owes much to another logic, binary logic.
Given the latter’s imperium, and the staunch idealism that inspires the protago-
nists of the debates, negotiating the competing claims of the debate is no easy task.
Some share a blind faith in the so-called free market of ideas’ self-regulating powers,
while others entertain idealistic expectations about the effectivity of state law, or more
generally, of institutional law, in correcting social inequalities. Each party to the de-
bate demonstrates an exaggerated skepticism about the other’s arguments. In the end,
neither the idealist arguments nor the skeptical ones are very convincing. For exam-
ple, if it is true that regulating hate propaganda may contribute to the goal of substan-
tive equality, it is not accurate to state that such regulation will solve the problem of
inequality. Conversely, if it is true to say that some state regulation of a given form of
expression may in some circumstances lead to totalitarianism-the slippery slope ar-
gument-it is erroneous to state that it necessarily leads to it, especially where a civil
” Following Chaim Perelman, legal interpretation can be depicted as an attempt to answer the ex-
pectations of a certain number of audiences. For a work that applies Perelman’s theories to Canadian
constitutional adjudication and then refines it, see A. Lajoie, Jugements de valeurs (Paris: Presses
Universitaires de France, 1997).
2001]
J.-F GAUDREAULT-DESBIENS – REGULATING HATE PROPAGANDA
1127
society and democratic tradition exist. This kind of intellectual process works by in-
flating a specific example, either positive or pathological, so as to transform it into a
rule that is universally applicable. In the end, none of the arguments and counter-
arguments invoked in the debate about anti-egalitarian expression are compelling in
themselves. In fact, the argument that most effectively refutes objections in principle
to any form of state control of hate propaganda is fundamentally an ethical one. It
posits that hate propaganda conveys an unambiguous message of contempt and deg-
radation that denies the humanity of those it targets.” Put otherwise, this form of ex-
pression denies its victims the right to participate as equals in social life. At a societal
level, the question then becomes whether a democratic society is bound to tolerate
some of its members actively inciting their fellow citizens to disrespect and demean
other members of the very same society, and ultimately, to inflict harm on them. Con-
versely, is a democratic society ethically obliged to help, even in the symbolic realm,
those people whose very humanity is radically denied? According to David Kretzmer,
Ifihis argument stresses the symbolic importance of restrictions on racist
speech. … it does not necessarily assume that the prevention of racist speech
will result in fewer people subscribing to racist ideas. … it does not emphasize
the indignity caused by the exposure of target populations to racist speech.
rather it stresses the indignity of living in a society in which such speech is
protected. The thrust of this argument is that a society committed to the ideals
of social and political equality cannot remain passive: it must issue unequivocal
expressions of solidarity with vulnerable minority groups and make positive
statements affirming its commitment to those ideals. Laws prohibiting racist
speech must be regarded as important components of such expressions and
statements.
More than any other, this argument convinces me that a limited regulation of hate
propaganda, and of other forms of anti-egalitarian speech in general, is not inherently
problematic. This is not to say, however, that such regulation is entirely unproblematic.
While regulation of hate propaganda may have a considerable symbolic impact, it
can also entail long-term social consequences. As Richard Delgado points out, “The
establishment of a legal norm creates a public conscience and a standard for expected
behavior that check overt signs of prejudice:’ Legislation aims first at controlling only
the acts that express undesired attitudes. But “when expression changes, thoughts too
in the long run are likely to fall into line”‘” Some object to such arguments on the
grounds that they legitimize the idea of governmental thought control, since the pur-
pose of hate propaganda regulation would not be the prevention of harm, but rather
‘9 C.R. Lawrence I, “If He Hollers Let Him Go: Regulating Racist Speech on Campus” [1990]
Duke LJ. 431 at 461.
‘ D. Kretzmer, “Freedom of Speech and Racism” (1987) 8 Cardozo L Rev. 445 at 456.
2 1 R. Delgado, “Words That Wound: A Tort Action for Racial Insults, Epithets, and Name-Calling”‘
(1982) 17 Harv. C.R.-C.L. L. Rev. 133 at 149 [footnotes omitted].
1128
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 46
the elimination of any anti-egalitarian thought. In other words, those who support the
regulation of hate propaganda are reproached for desiring a prior control of attitudes
or beliefs, rather than simply a downstream control of the expression of hateful atti-
tudes or beliefs.’ This line of argument is far from convincing. Without elaborating
more than necessary on a question pertaining to legal ontology, is not the ultimate
objective of most legal norms to influence not only immediate behaviours but also at-
titudes and beliefs, or put otherwise, intellectual presuppositions that condition be-
haviour and, as a result, effect respect for the norm? Consider the classic example of
the obligation to stop at a red light. What explains that most people stop? Is it the con-
stant fear of being caught? If so, the strict effectivity of a norm depends on the state’s
coercive mechanisms. Or rather is it a mix of the fear of triggering these coercive
mechanisms and the internalization of the content of this norm because of its intrinsic
value? In the same vein, most people respect legal prohibitions against fraud not only
for fear of being caught, but also because they have internalized these prohibitions. In
fact, the objection expressed above seems to rely on a rather reductionist vision of the
law’s effectivity; not only does it deny any legitimacy to positive norms purporting,
even minimally, to structure attitudes; it also denies that even the most innocuous
norms may have that effect or purpose. This position is premised on the belief that
positive norms are intrinsically neutral, an ideal that may be legitimate but is almost
never realized in fact. Indeed, since all positive norms embody public policy objec-
tives to various degrees, the law, more than being simply procedural, necessarily pro-
duces meaning for the purpose of moulding attitudes and thoughts in one way or an-
other. The result-the attitudes and thoughts so moulded-may certainly be subject to
criticism, but it is hard to see why the process itself would be fundamentally flawed.
Legal norms are signs that contribute to shape each individual’s thoughts and atti-
tudes, just as other social “signs” do.
Furthermore, simply equating the impact the law may have on people’s con-
science with some Orwellian form of thought control seems even more dubious con-
sidering the social context from which this argument stems. Without denying the need
for vigilance, the notion that a real “thought control” system could be implemented in
a society with a long-standing democratic tradition and civil society seems far-
fetched, since any civil society would oppose such a system. The counter-argument is
that, while anti-hate legislation is a mild form of thought control, it constitutes a dan-
gerous first step in the direction of a stronger one. This is a variation on the slippery
slope argument. I need not repeat that such an argument is not conclusive in itself,
since it fails to take into account the specific socio-legal fabric of the society in which
such anti-hate laws are enacted. Once again, the existence of a democratic tradition
and strong civil society would probably prevent slippage towards totalitarianism or
22 See e.g. L. Alexander, “Banning Hate Speech and the Sticks and Stones Defense” (1996) 13
Const. Commentary 71 at 79.
2001]
J.-E GAUDREAULT-DESBIENS – REGULAT7NG HATE PROPAGANDA
1129
authoritarianism. This is not to say that democratic societies are not vulnerable to fas-
cist impulses. It is only to point out that, although valuable for reminding us of that
vulnerability, metaphors such as the slippery slope should be used with caution, since
they tend to drain complexity from any analysis. As such, they act as epistemological
obstacles, and more specifically, as “verbal” ones, to use Gaston Bachelard’s classifi-
cation.
Bachelard characterizes “verbal” obstacles as encompassing all metaphors that
pretend to explain complex phenomena while actually obscuring their complexity.
These metaphors therefore act as intellectual “sponges”, sucking complexity from the
analysis and pre-packaging the phenomena they are supposed to describe in such a
way as to give the impression that they describe them accurately and completely.’ The
intellectual process that leads to the transformation of useful metaphors into episte-
mological obstacles is often triggered by a search for analytical certainty. The process
implies a shift from the relatively weak status of metaphor–that is, simply a figure of
speech-to the stronger one of “ideal reality”,’ a concept that roughly describes the
factualization of some representations in the realm of social relations. Not only do
these representations reflect social relations; they also constitute them: once believed,
they cease to be representations, strictly speaking, and enter the content of social rela-
tions. What may have been a normative representation of how things could or should
be will ultimately be transformed into a material fact. For example, the free market of
ideas, which is a legitimate ideal but certainly not an empirical reality, becomes in the
minds of its believers an empirical reality that serves as the incorrigible’ intellectual
foundation upon which they base their concept of freedom of expression. Similarly,
the slippery slope metaphor, which is meant to warn against a possible danger, often
leads its believers to see actual danger in contexts that do not empirically warrant such
fear. Without factual indication to that effect, and notwithstanding the socio-political
culture of the country, they imagine that regulating hate propaganda could actually
lead to totalitarianism. In other words, actual danger will be assumed whatever the
circumstances,’ preventing interpreters from pursing any thorough inquiry on the ac-
tual circumstances in which regulation may take place. In conclusion, there may be a
slippery slope to the reckless use of the “slippery slope” metaphor!
23 See Bachelard, supra note 15 at 78-79.
, M. Godelier, L’idel et le inatirie” pensde, 4conondes, socits (Paris: Fayard, 1984) at 172.
2I use the word “incorrigible” in its epistemological sense, meaning “immune from refutation”.
See W.P. Alston, Episteinc Justification: Essays in the Theory of Knowledge (Ihaca, N.Y.: Corell
University Press, 1989) at 286. On the free market of ideas metaphor as a potential epistemological
obstacle, see “Du droit et des talismans”, supra note 17 at 733-43.
6 It is ironic to note that the “slippery slope” metaphor being founded on an assumed danger, that of
that hate propaganda may indeed cause harm to
totalitarianism, denies another potential danger, i.
some groups and individuals. Would the causationist approach operate selectively?
1130
MCGILL LAW JOURNAL / REVUE DE DROITDE MCGILL
[Vol. 46
That being said, it should be clear to my readers that I do not subscribe to these
“all or nothing” approaches that play such a huge role in debates about hate propa-
ganda, and more generally, anti-egalitarian expression. Nevertheless, once we have
concluded that the regulation of hate propaganda is not inherently problematic, at
least in principle, the potential scope of such regulation remains to be determined.
Leaving aside the indirect effect positive norms may have through their symbolic and
mind-structuring impact, we must decide which forms of anti-egalitarian expression
should be subject to a norm intended to be directly effective. Some parameters may
lead us to a relatively satisfying answer.
First, we should not forget the limits of state, or institutional, law in preventing
social phenomena that are systemic in nature such as racism, or in eradicating related
epiphenomena such as racist expression. Although we must remember the role that
anti-egalitarian expression plays in the reproduction of discriminatory social systems,
we should not forget that it would be inappropriate-and unrealistic-to view the
problem from the standpoint of an absolutist logic purporting to prevent or eliminate
all risk of harm.
Second, while it is not only legitimate, but also necessary, to look at the harm
anti-egalitarian expression causes from the victims’ perspective, we should not fall
prey to complacency or intellectual abdication. The multiple forms of anti-egalitarian
expression that exist are neither equally harmful nor performative; we must not, there-
fore, lose sight of the link between the norm that the state is drafting and the broader
public policies involved when identifying the specific forms of anti-egalitarian ex-
pressions to discourage. As much as the victims’ perspective must be actively listened
to rather than passively heard, this listening process should enhance, not trump, the
values underlying a free and democratic society, including freedom of expression.
Third, one must bear in mind that legal control of anti-egalitarian forms of ex-
pression may entail unwanted consequences, both at the enactment stage and during
implementation. The worst of these is probably the elevation of those who convey
anti-egalitarian views to martyr status. As Richard Abel correctly points out, “The
greatest perversion … is that law, far from silencing harmful speech, rather encour-
ages, valorises, and publicises it, transforming offender into victim and offense into
romantic defiance “‘” Rhetorically, this is illustrated by what I call the “Galileo syn-
drome”, which is the propensity to systematically invoke the example of Galileo be-
fore the Inquisition, or some equivalent, in order to delegitimize any attempt at regu-
lating anti-egalitarian expression and, a fortiori, enforcing such regulation.2 Aside
27 R. Abel, Speech and Respect (London: Stevens & Sons, 1994) at 107.
2’ See J.-F. Gaudreault-DesBiens, La liberti d’expression entre ‘art et le droit (Quebec: Presses de
l’Universit Laval, 1996) at 258.
2001] J.-F GAUDREAULT-DESBIENS – REGULA7NG HATE PROPAGANDA
1131
from being paradoxical, if not absurd,’ the problem with this kind of comparison is
that it trivializes freedom of expression itself. Indeed, by relying on quantitative rather
than qualitative logic, it basically considers any non-violent eructation an “expres-
sion”, even if it is entirely irrational. Indeed, irrationality is a hallmark of anti-
egalitarian expression.’ While the rationality of an expression should not be in and of
itself a determining factor in deciding whether it deserves constitutional protection,”
and while it may be pragmatically defensible to reject any a priori distinction between
less performative forms of expression at the stage of circumscribing the scope of free-
dom of expression for constitutional purposes,” it is nonetheless important to ac-
knowledge this trivializing effect.
Last, it should never be forgotten that there is no entirely optimal way to reconcile
equality and freedom of expression; only a complex apprehension of the problems
raised by anti-egalitarian expression may lead to a relatively satisfying compromise
where one right will not trump the other. The preceding comments thus induce me to
say that any regulation of anti-egalitarian expression, even the most abhorrent, can
only pursue limited and well-defined ends. It follows that a legal norm purporting to
be directly, as opposed to symbolically, effective, could and should only target ex-
treme forms of anti-egalitarian expression. The confinement of positive law to the
realm of extremes seems, in any case, inevitable considering the limited effectivity of
state regulation of expression, the possible unwanted consequences of such regula-
tion, and more important, the difficulty of drawing a clear line between the forms of
anti-egalitarian expression that should be proscribed and those that should not.
We inevitably come back to the problems raised by state regulation of forms of
expression that may bear multiple, and even contradictory, meanings. How can the
border between what should and should not be proscribed be established? The ethical
Indeed, it is quite strange to hear hate-mongers tried for the dissemination of hate comparing their
“plighf’ to that of Galileo before the Inquisition when their own views of what a society should be
have much more in common with the Inquisition’s than with Galileo’s.
Jean-Paul Sartre argued, rightly as far as I am concerned, that anti-Semitism implies a devaluation
of both words and rationality. See J.-P. Sartre, Rflaxions sur la question juive (Paris: Gallimard,
1954) at 22. Obviously, this is not to say that anti-Semitic expression, however irrational, is not ra-
tionalized by its proponents. Evidence of this can be found in their use of pseudo-scientific theories,
Le. of superficially rational discourse, to legitimize their views.
3′ E.g. art is not always rational, but it is commonly accepted in Western democracies that it de-
serves constitutional protection. On art and expression, see N. Carroll, Philosophy of Art: A Contem-
porary Introduction (London: Routledge, 1999) at 59-106.
This is precisely what the Supreme Court of Canada did in its jurisprudence on s. 2(b) of the Ca-
nadian Charter of Rights and Freedoms, Part I of the Constitution Act, 1982, being Schedule B to the
Canada Act 1982 (U.K), 1982, c. 11 [hereinafter Charter], when it refused to accept arguments that
purported to exclude hate propaganda from the ambit of this subsection. See R. v. Keegstra, [1990] 3
S.C.R. 697,117 N.R. 1.
1132
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 46
argument that I found to be compelling enough earlier to justify, at least in principle,
regulation of some anti-egalitarian expression leads us down an interesting path. I ar-
gued that it is not conceivable for a democratic society to remain mute when con-
fronted with discourse that denies the very humanity of entire groups based on some
defining feature of their identity, or incites their exclusion or, worse, their elimination.
Not only is such discourse offensive; in many respects it is abusive as well. Indeed, in
addition to advocating abusive behaviour towards the targeted individuals or groups, it
abuses democratic societies’ tolerance by denying the intrinsic value of the social and
cultural pluralism that these societies cherish. Paradoxically, such discourse demands
the right to deny the plurality of individuals and groups within society in the name of
the pluralism of ideas. It is this abusive expression, conceptually distinct from offen-
sive expression, that the state may legitimately try to regulate. The distinction between
one form of expression and the other is admittedly not always clear, but absence of
intrinsic clarity is the lot of many other distinctions in the legal field. Nothing there-
fore justifies treating this distinction differently from other ones of the same nature.
When manipulating this distinction, however, one must tune out the sirens of
dominant discourses that, for ideological reasons, obscure the complexity of the
problems raised by state regulation of anti-egalitarian expression. Opponents of regu-
lating hate propaganda, for example, systematically depict those supporting such
regulation as asking for the censorship of merely offensive expression, thus trivializ-
ing the harm that abusive anti-egalitarian expression may cause.” Conversely, propo-
nents of such regulations often confuse offence and abuse in their desire to dismantle
the tools that (re)produce inequality. Hence the importance, as noted earlier, of re-
fraining from an uncritical acceptance of the victims’ perspective.’ In fact, if I had to
find a formula to capture the essence of the distinction between offensive expression
and abusive expression, I would propose this one: offensive expression targets ideas,
while abusive expression targets human beings. In this light, an offence implies, at
worst, that the values of a group are confronted. For example, the depiction of Jesus
Christ in Martin Scorcese’s Last Temptation of Christ or that of Mahomet in Salman
Rushdie’s Satanic Verses certainly clashes with some Christians’ or Muslims’ beliefs.
But these works, as offensive as they may be for some, do not advocate that Christians
or Muslims are inferior beings who not only can, but must, be excluded and degraded
as individuals and as a group. Racist hate propaganda, in contrast, advocates the need
33F. Schauer, “The Sociology of the Hate Speech Debate” (1992) 37 Vill. L. Rev. 805 at 814.
E.g. in its interpretation of the equality right guaranteed by s. 15 of the Charter, supra note 32, the
Supreme Court of Canada relativizes, without in any way obscuring it, this perspective by assessing
the harm allegedly suffered from the subjective-objective perspective of the “reasonable victim”. See
Law v. Canada (Minister of Employment and Immigration), [1999] 1 S.C.R. 497 at paras. 59-61, 170
D.L.R. (4th) 1.
2001]
J.-E GAUDREAULT-DESBIENS – REGULATING HATE PROPAGANDA
1133
and the right to perform such exclusion and degradation, as do some forms of hard-
core pornography.’
Some would fault the distinction between offensive expression and abusive ex-
pression for being unduly Eurocentric in that it relies on a rationalist philosophy. Per-
sons whose beliefs are offended, it could be argued, may be said to be victimized in
their humanity, since these beliefs are partly constitutive of who they are. This may be
so, but it does not change my answer. it would be antithetical to the idea of democ-
racy to uphold the right not to be offended. Indeed, the emphasis that democracy
places on freedom, in both its liberal and republican forms, fosters an atmosphere of
tempered relativism which, in addition to inducing tolerance, encourages the ques-
tioning of orthodoxies and the raising of doubt. In this light, as was pointed out by
Homi Bhabha, Salman Rushdie’s greatest sin is having proposed an alternative read-
ing of the Koran that went against Iranian mullahs’ orthodox interpretation.’
That said, the reference I make to democracy clearly indicates the intellectual
framework in which my reflection takes place-that of a democratic society em-
bodying the Enlightenment’s ideals, with their defects but also with their positive
qualities. Such philosophical “ethnocentrism” is inevitable, and it may well be that
some values are irreconcilable. Without any imperialist vocation, this ethnocentrism
denies neither diversity nor Otherness; it simply acknowledges that it is only from our
own standpoint that we can participate in a true and authentic dialogue with the Other.
It also posits reciprocity in openness and compromise as an essential condition for
such a dialogue to take place. Last, it acknowledges that in any interpretive commu-
nity, the act of judging is, for historical reasons, constrained within a certain intellec-
tual framework that is futile to hide. This philosophical ethnocentrism, which in a way
relates to Richard Rorty’s epistemological relativism,” is therefore entirely opposed to
radically relativist schools of thought which, to use Claude Lefort’s words, would see
in democracy a “simple preference”.”
Ultimately, socio-legal reflection on the appropriateness of regulating hate propa-
ganda, and on the potential scope of that regulation, leads me to believe that regula-
” If I use here the word “some” in relation to hard-core pornography, it is because pornography
raises more complex issues than racist hate propaganda. Indeed, since it participates in the polysemic
phenomenon of sexual representation, the meaning of pornography, hard-core or not, is not ahys as
clear or obvious as racist hate propaganda. For sure, it may be exclusion or degradation, but it may
also be something else.
“See H. Bhabha, 77e Location of Culture (London: Routledge, 1994) at 226.
R. Rorty, Objectivity, Relativism, and Truth: Philosophical Papers, vol. I (Cambridge: Cambridge
University Press, 1991) at 30.
C. Lefort, “La libert
l’heure du relativisme” in Les usages de la liberte: tertes des confrences
et entretiens organisis par les trente-deuxiinres Rencontres Intemationales de Geneve 1989
(Neuchatel: tditions de la Baconnire, 1999) 237 at 241.
1134
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 46
tion must be confined to anti-egalitarian expression that is both extreme and harmful.
In other words, the scope of such regulation must be restricted to expression which, to
borrow the language of the Quebec Charter of Human Rights and Freedoms, has “the
effect of nullifying or impairing [the right to full and equal recognition and exercise of
[one’s] human rights and freedoms],”” or worse, which may be characterized as direct
attempts at inducing people to act so as to nullify or impair those rights.”‘ Although
this solution is unsatisfactory in many respects, it is the only one that permits a certain
reconciliation of equality and freedom of expression, rather than allowing one to
trump the other. One could argue that proscribing only the worst epiphenomena of a
systemic problem of hate propaganda risks lulling us into the belief that the entire
problem has been definitively solved. In the same vein, one could also argue that, un-
der the proposed solution, more pervasive and discrete anti-egalitarian expression re-
main untouched-for example, the “mild” stereotypes that permeate mass culture.
Even worse, to target only the most obvious and extreme anti-egalitarian expression
could lead some to believe that everything not expressly condemned by the law is so-
cially acceptable and thus legitimate.
While not unfounded, these observations do not take into account three important
factors that reduce the potentially negative effects of a confinement of the regulation
of anti-egalitarian expression to its most extreme and harmful forms. First, other rules
participate in the fight against “ordinary” anti-egalitarian expression, such as the anti-
discrimination norms proscribing sexual or racial harassment, or the norms acknowl-
edging the right to dignity. Second, the symbolic-normative impact of the regulation
of anti-egalitarian expression, even if restricted to the most extreme cases, may largely
exceed the cases so targeted. Last, debates on the regulation of anti-egalitarian expres-
sion may themselves shape the legal-normative consciousness of individuals and
groups, to the point of triggering the emergence of new “social” or informal norms,”
or of modifying older ones. 2 Viewed this way, the enactment of positive norms pur-
porting to regulate anti-egalitarian expression could simply amount to an institution-
alization of already effective informal ones.
” This proposition is inspired by the wording of the Charter of Human Rights and Freedoms,
R.S.Q. c. C-12, s. 10. One could argue that this section embodies what Hannah Arendt called the
“right to have rights”.
,’ See P. Bosset, “Les mouvements racistes et ]a Charte des droits et libertds de la personne” (1994)
35 C. de D. 583 at 611-13.
” I do not use here the expression “social norm” to distinguish positive legal norms from extra-legal
ones, as this expression has a very low heuristic value. See V. Olgiati, “Le pluralisme juridique
comme lutte pour le droit: la folie th~orique et mdthodologique d’une rcente proposition” (1997) 12
Can. J. L. & Soc. 47 at 53, n. 16.
4′ This is precisely the hypothesis advanced by Professor Schauer in light of the actual success of
Catharine MacKinnon’s anti-pornography campaign, in spite of the striking down of the ordinance
she had inspired. See Schauer, supra note 33 at 818-19.
2001]
J.-F GAUDREAULT-DESBIENS – REGULATING HATE PROPAGANDA
1135
If we accept that hate propaganda is likely to nullify or impair the right to full and
equal exercise of human rights and freedoms, any inquiry into the nexus between hate
propaganda, hate crimes, and genocide ultimately forces us to reflect on the realiza-
tion of human rights. Indeed, how can one truly exercise one’s rights and freedoms if
one lives in a society that tolerates expression that denies one’s equal dignity as a hu-
man being? In the end, this individual will probably feel alienated in his or her own
society, no longer comfortable exercising his or her supposedly inalienable rights and
freedoms.’3 While such situations undermine the legitimacy of these rights as useful
tools to fight discrimination and exclusion, they also demonstrate, albeit negatively,
that human rights find their real strength in their concrete and practical nature.
How can we contribute to the realization of both freedom of expression and
equality without systematically trumping one or the other, bearing in mind the limits
of law and legal discourse? The answer to this is far from obvious, as illustrated by
this discussion. At both domestic and international levels, however, inspiration can be
drawn from the ideals underlying the first three generations of human rights. These
are embodied in the French Revolution’s normative ideals, that is, “liberty” for first
generation rights, “equality” for second generation rights, and “fraternity” for third
generation rights. Notwithstanding questions associated with the justifiability of sec-
ond and third generation rights, the fact that a whole normative structure centred
around “basic rights” was built after World War H shows a concern for the realization
of such rights. It was not enough to entrench “core” civil and political rights; some-
thing more was needed to make these rights effective. Acknowledgement of this gap
eventually led to the recognition of social, economic, and cultural rights-the second
generation rights-and later, of third generation collective rights such as the right to
development, identity, and environment. Third generation rights, by emphasizing the
idea of “fraternity”, serve together with second generation rights as a reminder that
solidarity is not foreign to human rights discourse. Not only is it not foreign, but it
could be argued that the interpretation of any human right, as well as the resolution of
any conflict between various human rights, should be inspired not only by the ideal of
liberty and equality, but also by this too-often-forgotten ideal of fraternity, or if one
prefers, solidarity.
3 G. Haarsher, Philosoplde des droits de l’honune, 4th ed. (Brussels: Universitd de Bruxelles, 1993)
at 42.
“Fraternity already finds expression in the public lav, as well as the private law, of some countries.
This is obviously the case with France (see M. Borgetto, La notion defratemli en droit publicfran-
gais: le passi, le prsent et l’avenir de la solidaritJ (Paris: LG.DJ., 1993)), but also, to some extent,
with Canada. It could even be argued that fraternity is one of the implicit principles of the Constitu-
tion of Canada. On these questions, see C.D. Gonthier, ‘Liberty, Equality, Fraternity: The Forgotten
Leg of the Trilogy, or Fraternity: The Unspoken Third Pillar of Democracy” (2000) 45 McGill W.
567.
1136
MCGILL LAW JOURNAL / REVUE DE DROITDE MCGILL
[Vol. 46
With respect to the case of hate propaganda, the reconciliation of freedom of ex-
pression and equality may, perhaps, be made through reference to this normative
ideal, thereby forcing us to recognize the relational dimension of even the most indi-
vidualistic rights such as freedom of expression. As noted, acknowledging the con-
crete importance of that ideal may lead us to recognize, both in the domestic and in-
ternational realm, the legitimacy of some forms of hate propaganda regulation, and of
the enforcement of such regulation when needed. Particularly in the international
realm, acknowledging the legitimacy of such enforcement may also result in the es-
tablishment of a nexus between what is perceived to belong to traditional human
rights law and what is now characterized as humanitarian law. From it may ultimately
flow a much-needed reconsideration of the imperium of state sovereignty, as well as a
relaunching of the debate around the duty and the right of the international commu-
nity to interfere in a state’s internal affairs under certain circumstances.
I will conclude these remarks by returning, first, to Sisyphus himself, and then to
the notion of the Sisyphus state. Albert Camus once described Sisyphus’s fate as fol-
lows: “The Gods had condemned Sisyphus to ceaselessly rolling a rock to the top of a
mountain, whence the stone would fall back of its own weight. They had thought with
some reason that there is no more dreadful punishment than futile and hopeless la-
bor” 5 This last sentence raises a fundamental question about the attitude the Sisyphus
state should adopt towards the regulation of hate propaganda: Is it not a futile and
hopeless labour to regulate that propaganda while knowing very well the obstacles
that arise in enforcing that regulation? In metaphorical terms, is it worthwhile trying
to contradict the fate the Gods reserved Sisyphus? I believe that the answer should be
positive. Indeed, emphasizing the importance of solidarity as a normative ideal leads
us to assess the regulation of hate propaganda from a different angle. While there is
no doubt that the decision to regulate or not that kind of expression places the Sisy-
phus state in a dilemma because of the difficulties it is likely to experience in enforc-
ing such regulation and of the perverse effects that are associated with it, the state that
accepts solidarity as a legitimate inspiration for such regulation, however, conveys the
message that, notwithstanding these problems, it is ready to make a commitment to a
certain kind of society; one in which, for ethical reasons, the idea of constitutional
democracy cannot be reconciled with the radical denials of the humanity of some of
its citizens. This entails that indifference to these citizens’ victimization is itself intol-
erable. Hence the need to convey this message through formal legal means, the effect
of which is significant, despite being largely symbolic. Moreover, although I initially
suggested that, unlike Sisyphus, the state has a choice whether to engage itself in this
labour that some could indeed characterize as “futile and hopeless”, the principles of
democracy compel the state to do something in order to shake up its citizens’ indiffer-
” A. Camus, The Myth of Sisyphus and Other Essays, trans. J. O’Brien (New York: Alfred A.
Knopf, 1969) at 119.
20011
J.-F GAUDREAULT-DESBIENS – REGULA7NG HATE PROPAGANDA
1137
ence and to make them feel collectively and individually responsible for the fate of
their fellow citizens. It should try to make sure that, when confronted with the phe-
nomenon of hate propaganda, each citizen, for a moment, thinks of himself as an-
other, to paraphrase Paul Ricceur.”‘ As Camus wrote, “The struggle itself toward the
heights is enough to fill a man’s heart. One must imagine Sisyphus happy.” As such,
viewed under the light of solidarity, the Sisyphus state’s–or the international com-
munity’s-initial dilemma may very well become its duty.
See P Ricceur, Oneself as Awther, trans. K. Blarney (Chicago: University of Chicago Press,
1992).
Camus, supra note 45 at 123.
