Economy of pain: When to regulate offensive expression

© The Author(s) 2018. Oxford University Press and New York University School of Law. All rights reserved. For permissions, please e-mail: [email protected]

I•CON (2018), Vol. 16 No. 3, 806–835 doi:10.1093/icon/moy060

Economy of pain: When to regulate offensive expression

Michael Ilg*

Expression that assaults another’s identity or human dignity, as with hate speech, presents a vexing case for regulatory protection. This article proposes a method of expressive elasticity as a means to balance individual protection with expressive liberty in a rigorous and neutral manner. The method of expressive elasticity treats harm to individual recipients as a func­tion of social utility. To maintain neutrality on the content of ideas, social utility is defined as the optimal number of individual participants in public discourse. Offensive expression should be regulated only when it reasonably lessens the likely amount of future contributors to the marketplace of ideas. The object is to restrict offensive expression that signals a hostile or prejudiced public space that a reasonable recipient would recoil from. Elasticity generally refers to the responsiveness of one variable to a change in another variable. Elasticity, as util­ized in this article, is a view to the responsiveness of individual participation to the amount of external harm received. When individuals can absorb the external harm and their poten­tial participation remains unchanged, there should be no government intervention. Elasticity should be recognized as inversely related to the private costs imposed on offensive expression. When a speaker is publicly reviled because of a hateful message, then the private marketplace has appropriately indicated that the targeted person is not the one devalued by society. When the private costs for hateful expression are insufficient, however, public sanction may be nec­essary to ensure optimal levels of future participation.

1. IntroductionMost liberal societies restrict egregious expression that assaults an individual or group’s human dignity. While legal protections of individual dignity have ancient roots, as with defamation laws, these protections are largely a modern development. The clearest example of a dignity protection may be found in the hate speech laws

* Associate Professor, Faculty of Law, University of Calgary. Email: [email protected].

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

Economy of pain: When to regulate offensive expression 807

adopted by nearly every liberal democracy,1 and the speech codes governing most uni-versity campuses,2 both of which generally seek to prohibit expression that denigrates or vilifies individuals from a historically vulnerable group.

The modern receptiveness to dignity protections extends from an understanding of the grave psychological harm that may be wrought by offensive expression.3 A combi-nation of awareness and empathy has produced a model of government and judicial oversight of expression that is referred to in this article as the equality model. The equality model represents the balance between expressive liberty and equality struck by most liberal democracies,4 with the notable exception of the United States.5 The near uniformity in the adoption of hate speech regulations by liberal democracies

1 Canada, Great Britain, Australia, Germany, France, Italy, Finland, Norway, Sweden, and the Netherlands, among others, have adopted hate speech regulations. See David O.  Brink, Millian Principles, Freedom of Expression, and Hate Speech, 7 LegaL Theory 131 (2001); David Kretzmer, Freedom of Speech and Racism, 8 Cardozo L. rev. 445 (1987). Norms against hate speech also have a strong reflection in international law. See 1965 International Convention on the Elimination of All Forms of Racial Discrimination, § 4; December 21, 1965, 660 U.N.T.S. 195 (entered into force January 4, 1969); 1966 International Covenant on Civil and Political Rights, art. 20; December 16, 1966, 999 U.N.T.S., 171 (entered into force March 23, 1976).

2 University speech codes are a common institutional policy, if a controversial one. Such speech codes have long been criticized in the media as an example of political correctness run amok, or a version of “thought police.” See Charles R. Lawrence III, If He Hollers Let Him Go: Regulating Racist Speech on Campus, 3 duke L.J. 451 (1990). See, e.g., Lee Dembart, At Stanford, Leftists Become Censors, N.y. Times, May 5, 1989, at A35, col. 2; George Will, Liberal Censorship, Wash. PosT, Nov. 5, 1989, at C7, col. 4. More recently, a think tank devoted to campus free speech issued a report which claimed, inter alia, that nearly 60 per-cent of institutions surveyed possessed codes that substantially restricted student’s rights of expression. FouNdaTioN For iNdividuaL righTs iN eduCaTioN, sPoTLighT oN sPeeCh Codes 2014: The sTaTe oF Free sPeeCh iN our NaTioN’s CamPuses (2014), http://www.thefire.org/spotlight/reports/. See also Greg Lukianoff, Feigning Free Speech on Campus, N.y. Times, Oct. 24, 2012, http://www.nytimes.com/2012/10/25/opinion/feigning-free-speech-on-campus.html [hereinafter Lukianoff, Feigning Free Speech]; Greg Lukianoff, Speech Codes: The Biggest Scandal on College Campuses Today, Forbes, Dec. 19, 2012, http://www.forbes.com/sites/real-spin/2012/12/19/speech-codes-the-biggest-scandal-on-college-campuses-today/ [hereinafter Lukianoff, Speech Codes]; Walter Olson, University Speech Codes, Reborn as Anti-bullying Rules? CATO Institute (Nov. 23, 2013), http://www.cato.org/blog/university-speech-codes-reborn-anti-bullying-rules. Campus speech codes often conflict with the freedom of expression and First Amendment guarantees. See generally Evan G. S. Siegel, Note, Closing The Campus Gates to Free Expression: The Regulation of Offensive Speech at Colleges and Universities, 39 emory L.J. 1351, 1375 n.137 (1990); Erwin Chemerinsky, Unpleasant Speech on Campus, Even Hate Speech, Is a First Amendment Issue, 17 Wm. & mary biLL rTs. J. 765, 770 (2009).

3 See, e.g., Mari J. Matsuda, Public Response to Racist Speech: Considering the Victim’s Story, 87 miCh. L. rev. 2320 (1989); Richard Delgado, First Amendment Formalism Is Giving Way to First Amendment Legal Realism, 29 harv. C.r.-C.L. L. rev. (1994); Lawrence, supra note 2, at 431.

4 The United Kingdom, Canada, Australia, New Zealand, South Africa, and all of the Scandinavian coun-tries, among many others, have accepted the mandates of International Covenant on Civil and Political Rights, art. 20(2); December 16, 1966, 999 U.N.T.S., 171 (entered into force on 23 March 1976). Convention on the Elimination of all Forms of Racial Discrimination, arts. 4(a), 4(b); December 21, 1965, 660 U.N.T.S. 195 (entered into force January 4, 1969); by making it a crime to engage in the incite-ment to racial, religious, or ethnic hatred or hostility. Frederick Schauer, The Exceptional First Amendment, in ameriCaN exCePTioNaLism aNd humaN righTs 34 (Michael Ignatieff ed., 2005).

5 “No other nation,” Charles Fried observed, “claims as fierce and stringent a system of legal protection for speech. It is the strongest affirmation of our national claim that we put liberty above other values.” Charles Fried, The New First Amendment Jurisprudence: A Threat to Liberty, 59 u. Chi. L. rev. 225, 229 (1994). See also Shauer, supra note 4, at 30 (“the American First Amendment, as authoritatively interpreted, remains a recalcitrant outlier to a growing international understanding of what the freedom of expression entails.”).

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

808 I•CON 16 (2018), 806–835

demonstrates the appeal of the equality model. Yet, freedom of expression is a cher-ished ideal in most of these same liberal democratic societies, and so the potential that equality claims may unduly chill free expression remains a persistent concern. There is a danger that trivial or strategic claims of offense that gain notoriety will undermine human rights protections for those instances that are deserving of public sanction.

Drawing upon economic concepts may provide for a more balanced means for the regulation of hate speech in three related ways. First, an economic perspective offers a clear indication of the vulnerability of expression to silencing. Most would agree that it would be repugnant for a young child to look out her front window and see vile signs condemning her existence and that of her family because of an inherent feature of their birth, background, or belief. But a person’s dignity can be lessened in many, less stark ways, including criticism. Protection of personal dignity is more problematic if it is invoked to curb criticism of a cultural practice or belief system. An extreme example would be the use of dignity claims by corporations to silence criticism of their products or activities with the threat of costly legal proceedings.6

A second advantage of an economic approach is that subjective overreach could be lessened by connecting specific harm with general social utility. Instead of a rationale based on the imagined harm to recipients, offensive expression should be evaluated on the basis of whether or not social utility is decreased. To be clear, the justification for a system of hate speech protection need not be connected with the most effective means of its articulation. While the justification for regulating expression no doubt emanates in part from empathy, empathy should not be the guiding notion of the system’s ar ticulation. Empathy is too unwieldy, and can lead to quixotic campaigns against hateful ideas in abstraction. Empathy may explain why we choose to regulate expression in the first place, but elasticity provides a means for regulating expression in a more disciplined manner.

Utility is defined here as a function of potential contributions to public knowledge and debate. The object to be protected is a neutral individual ability for participation, not a valuable idea per se. This definition of utility avoids a form of idea utilitarian-ism, wherein a dispute over claimed offense could be resolved merely on the basis of whether the majority preferred an idea, cause, or party. Society has historically ben-efited in the long term from ideas once regarded as mad or heretical.7

Third, the concept of elasticity, borrowed from economic theory, may help inform the assessment of potential contribution loss. When utility is defined notionally as potential additions to public discourse, utility is lessened when an expressive act arbi-trarily cancels out the future contribution of others. Arbitrariness is significant for there may be circumstances when expression should certainly lessen the willingness

6 Such actions are commonly referred to as a strategic lawsuit against public participation (SLAPP). The term was first coined by Canan and Pring. See Penelope Canan & George Pring, Strategic Lawsuits Against Political Participation, 35 soC. Probs. 506, 506 (1988). See also Thomas A. Waldman, SLAPP Suits: Weaknesses in First Amendment Law and in the Courts’ Responses to Frivolous Litigation, 39 UCLA L. rev. 981 (1991–1992).

7 An eloquent portrayal of the majority’s benefits from rare and non-conformist geniuses was made by John Stuart Mill. See JohN sTuarT miLL, oN LiberTy 123–126 (1863).

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

Economy of pain: When to regulate offensive expression 809

of someone to participate in public life, as with the publication of information that reveals the corruption of a public official. Expressive acts can be suboptimal if they arbitrarily lead to the future loss of individual participants and potential contributors.

Elasticity, as an economic term, generally refers to the responsiveness of one vari-able to change in another.8 Elasticity normally is a sign of a competitive market, in which consumers can opt to substitute one product with another rather than pay an increased price. Inelasticity, on the other hand, occurs when increases in price are not answered with a proportionate decrease in demand, if at all. From a view to poten-tial contributions to knowledge, inelasticity is optimal. Inelasticity would indicate that individuals are resilient under external expressive harm and are unchanged in their willingness to participate in public life. Accordingly, the state should not intervene in situations of inelasticity.

Elasticity would initially involve an inquiry into language in context. The essen-tial question is whether a reasonable person, similarly situated, would be inclined to recoil from future contributions because of the offense? A person is assumed to recoil from future contribution because the offensive expression reflects that the surround-ing community is intolerant. Perhaps the message is delivered in such a way as to give the impression that the local majority is supportive of the hateful message. Or, per-haps the message of a fringe minority is delivered in such a flagrantly aggressive way that the recipient is fearful and shocked that the community could condone such a display. Elasticity would occur, for example, when a person shrinks from sharing opin-ions or ideas in public—at the workplace or the town council meeting, for instance—because they are genuinely concerned that the quality of one’s idea will be rejected or condemned for no better reason than one’s background or belonging.

Yet, a community should be given the opportunity to respond to an offensive speaker, and ideally provide sufficient sanction so that others are aware of the conse-quences of such expression and, more important, signal to vulnerable recipients that it is the speaker and not themselves who the majority rejects. An example of sufficient private costs would be the case of the public figure or celebrity who loses an enviable position, status, and lucrative opportunities, after he is caught giving voice to noxious views. Elasticity would normally be inversely related to private cost, for a recipient’s sense of intimidated isolation would unlikely persist in the wake of tremendous social outrage directed at an offensive speaker. In the unfortunate circumstance when the private marketplace does not, or cannot, respond, then government sanction is needed to impose costs and signal value.

Section 2 of this article reviews the equality model, noting in particular the recent controversies associated with the model due to subjective overreach. Section 3 con-siders the social cost of subjective overreach, which explores the characteristics of expression as a form of market failure. Section 4 examines the challenges of valu-ing ideas from an economic perspective. Section 5 describes the inherent elasticity of expression. Section 6 pursues the proposed method of elasticity through a series of

8 See, e.g., heNry L. moore, syNTheTiC eCoNomiCs 66 (1929) (“Elasticity of supply is the ratio of relative change of the quantity supplied to the relative change to supply price.”).

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

810 I•CON 16 (2018), 806–835

hypotheticals, and section 7 concludes with a method intended to avoid the subjec-tive overreach of the equality model while capturing the egregious instances that free speech absolutism would ignore.

2. The equality model

2.1. Markets in goods and ideas

The economic analysis of the law has been phenomenally successful in influencing a wide range of legal topics,9 but freedom of expression has not been one.10 Of the legal topics that have been resistant to economic influence, freedom of expression is argu-ably the most ironic, given that the topic’s most hallowed metaphor is the marketplace of ideas.11 Richard Posner has claimed that a surprising lack of economic analysis of freedom of expression has been due, in part, to the reluctance of economists to engage with the “rebarbative complexities that is legal doctrine,” and that lawyers specializing in expression tend to be motivated by an “enthusiasm for a version of liberal political thought to which economic analysis may seem uncongenial.”12

Posner’s observation reflects the tenor of earlier arguments offered by Aaron Director and Ronald Coase, both luminaries of the Chicago school of law and econom-ics. Director, writing in 1964, argued that while laissez-faire never really existed in the economy at large, its one bastion of support resided in the trade of ideas.13 Director found the following statement by Justice Douglas representative of the legal hier-archy which placed ideas over goods and services: “free speech, free press, free exer-cise of religion are placed separate and apart; they are above and beyond the police power; they are not subject to regulation in the manner of factories, slums, apart-ment houses, production of oil and the like.”14 A further sign of the diminution of the trade of goods is presented in a declaration by Justice Black that when it comes to the

9 See, e.g., riChard a. PosNer, FroNTiers oF LegaL Theory 31–32 (2001) (observing that law and economics has been the most influential interdisciplinary school of jurisprudence).

10 On the relatively poor connection between law and economics and expression, see Ian B.  Lee, Can Economics Justify the Constitutional Guarantee of Freedom of Expression?, 21 CaNadiaN J. L. & JurisPrudeNCe 355, 359 (2008) (noting that the law and economics literature has not provided “a persuasive justi-fication for the constitutional guarantee of freedom of expression”). See also Eric Posner, Strategies of Constitutional Scholarship, 26 LaW & soC. iNquiry 529, 544 (2001) (“[It] may be that rational choice theory simply cannot justify constitutional rights, and rational choice thinkers ought to criticize them rather than defend them.”).

11 The marketplace metaphor is derived from Justice Oliver Wendell Holmes’s famous dissent in Abrams, in which he argued that the determination of truth is best left to the “free trade in ideas,” and that ideas had to be tested for persuasiveness in a “competition of the market.” Abrams v. United States, 250 U.S. (1919) (Holmes, J., dissenting). On the legacy of Holmes’s articulation, Felix Frankfurter wrote: “It is not reckless prophecy to assume that his famous dissenting opinion in the Abrams case will live as long as English prose retains its power to move.” FeLix FraNkFurTer, mr. JusTiCe hoLmes aNd The suPreme CourT 54–55 (1938).

12 Richard Posner, Free Speech in an Economic Perspective, 10 suFFoLk L. rev. 1, 3 (1986).13 Aaron Director, The Parity of the Economic Market Place, 8 J.L. & eCoN. 1, 2 (Oct. 1964).14 Beauharnis v. Illinois, 343 U.S. 286 (1952).

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

Economy of pain: When to regulate offensive expression 811

setting of natural gas prices that enter interstate commerce, “the alleged federal con-stitutional questions are frivolous.”15 According to Director, a regulatory dichotomy between ideas and goods is unjustified, given that:

. . . the bulk of mankind will for the foreseeable future have to devote a considerable frac-tion of their active lives to economic activity. For these people, freedom of choice as owners of resources in choosing within available and continually changing opportunities, areas of employment, investment, and consumption is fully as important as freedom of discussion and participation in government.16

Ronald Coase, writing in 1973, expressly endorsed Director’s argument and increased its rhetorical force.17 Coase was consternated by the paradoxical views on the efficacy of government regulation held by the liberal legal elite. Government was assumed to be per-fectly competent to regulate all manner of economic activity (especially in the heyday of the Keynesian economic consensus in which Coase and Director were writing), and often in the protection of consumers who “lack the ability to make the appropriate choices.”18 And yet, when it came to the market for ideas, government regulation would not be effi-cient and would likely be the result of bad motives, and consumers could “exercise a fine discrimination in choosing between alternative views placed before them. . . .”19

The paradoxical treatment of goods and ideas, according to both Director and Coase, was most likely due to the tendency to privilege one’s own profession, or trade, above that of others.20 As Coase stated bluntly:

The market for ideas is the market in which the intellectual conducts his trade. The explanation of the paradox is self-interest and self-esteem. Self-esteem leads the intellectual to magnify the importance of their own market. That others should be regulated seems natural, particularly as many of the intellectuals see themselves as doing the regulating.21

Neither Director nor Coase based their arguments in constitutional doctrine,22 and their object appeared to be the illustration of an important instance of intellectual and policy inconsistency, or simple hypocrisy. Coase concluded his piece by calling for a consistent treatment of the markets for ideas and goods, and essentially a consistent view of the efficacy of government regulation. If government is incompetent in the oversight of ideas, then this should inform the market for goods, and vice versa. Either way, the approach to different markets should be based on the same objectives and assumptions.23 It has been argued that Coase implied that his favored route would be for the liberalization of the market for ideas to be extended to that of goods.24 However,

15 Cities Services Co. v. Peerless Co., 340 U.S. 179, 189 (1950).16 Director, supra note 13, at 6.17 Ronald Coase, The Market for Goods and the Market for Ideas, 64 am. eCoN. rev. 384 (1974).18 Id. at 384.19 Id. at 384–385.20 See Director, supra note 13, at 6.21 Coase, supra note 17, at 386.22 Director, supra note 13, at 5: “I am aware that the preferred position accorded to free trade in ideas is

based on constitutional considerations, with which I am not concerned.”23 Coase, supra note 17, at 389.24 See Michael Rushton, Economic Analysis of Freedom of Expression, 21 ga. sT. u. L. rev. 693, 701 (2005).

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

812 I•CON 16 (2018), 806–835

intellectual sentiment evolved eventually in a way that Director or Coase might well have found surprising.

2.2. Shifting sympathies

While the ideological preferences described by Director and Coase were reflective of a certain period of liberal advocacy and thought, these preferences have never been static. A  common narrative for the evolution of liberal intellectual attitudes begins with the progressive era skepticism of constitutional rights, in which rights claims were perceived as obstacles or subversions to the project of a benevolent state.25 However, once the oppressive capacity, or tendency, of the state was revealed, partic-ularly with the “red scare” and the persecution of Communists, liberal progressives came to embrace a strong view of the freedom of expression as a means to curb gov-ernment excess.26 The liberal progressive compromise that resulted is surely the one assessed by Director and Coase—favoring government involvement in the economy but neutrality and non-interference over the marketplace of ideas.

At the time when Director and Coase wrote their critiques the liberal progressive compromise was arguably at its apogee, with free speech liberalism coinciding with the height of Keynesian economics. With the benefit of hindsight, it is easy to see that over time the broad group of intellectuals addressed by Coase became much more recep-tive to lessened regulation in the area of goods and increased regulation for ideas, in effect adopting the middle position contemplated by Coase. In terms of the market of goods, the decline of Keynesianism and the rise of market deregulation signaled an intellectual shift away from the faith in government regulation over the economy.27 Interestingly, law and economics encouraged this deregulation trend and rode it to prominence. The faith of intellectuals in the ability of government regulation over

25 See, e.g., david m. rabbaN, Free sPeeCh iN iTs ForgoTTeN years 3–4, 299 (1999) (relating how Progressives had a prewar faith in a “benevolent state” and a simultaneous “aversion to constitutional rights”); mark a. graber, TraNsFormiNg Free sPeeCh: The ambiguous LegaCy oF CiviL LiberTariaNism 65–74, 82–83 (1991) (describing how individual rights claims were viewed skeptically by Progressives).

26 See generally roberT CoTTreLL, roger Nash baLdWiN aNd The ameriCaN CiviL LiberTies uNioN 49–50 (2000) (indicating that the civil liberties movement was used as a defense from government action); doNaLd JohNsoN, The ChaLLeNge To ameriCaN Freedoms: WorLd War i  aNd The rise oF The ameriCaN CiviL LiberTies uNioN 197 (1963) (describing civil liberties as a defense to an oppressive wartime government); geoFFrey r. sToNe, PeriLous Times: Free sPeeCh iN WarTime 230 (2004); PauL L. murPhy, WorLd War i aNd The origiN oF The origiN oF CiviL LiberTies iN The uNiTed sTaTes 25–31 (1979); samueL WaLker, iN deFeNse oF ameriCaN LiberTies: a hisTory oF The ACLU 37–47 (1990); G. Edward White, The First Amendment Comes of Age: The Emergence of Free Speech in Twentieth Century America, 95 miCh. L. rev. 299, 312–314 (1996). But see Jeremy K. Kessler, The Administrative Origins of Modern Civil Liberties Law, 114 CoLum. L. rev. 1083 (2014) (arguing that, contrary to the common view, it was executive branch lawyers who took the lead in foster-ing civil liberties as a means of strengthening the administrative state).

27 President Nixon’s statement from 1971, “We are all Keynesians now,” may be considered ironic foreshad-owing for the decline and eclipse of Keynesianism that would be complete by the late 1970s. See generally roberT sideLsky, keyNes: reTurN oF The masTer xiv (2009) (marking Nixon’s statement as the beginning of the decline and fall of Keynesianism); JohN eaTWeLL & murray miLgaTe, The FaLL aNd rise oF keyNesiaN eCoNomiCs 9 (2011) (“by the end of the 1960s, the fault lines in [the Keynesian] consensus were becoming apparent to all.”).

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

Economy of pain: When to regulate offensive expression 813

the economy is likely countercyclical: growing in the wake of crises but subsiding in times of perceived government ineffectiveness and economic stagnation. The growth in intellectual receptiveness to government regulation of expression, by contrast, may in part be due to the fluctuating sympathies of liberal legal academics.

When government neutrality protected political dissenters, who were largely on the far left of the political spectrum, liberal progressive sympathies aligned well with expressive liberty and government neutrality. However, when liberal sympathies shifted toward a concern for historically disadvantaged groups, government neutral-ity became an obstacle to be overcome.28 The concern was no longer to protect a dis-senting minority from the majority’s use of the state but to use the state to protect disadvantaged minorities from private harms. The state, according to this liberal per-spective, would have to take sides in expressive disputes.

2.3. Empathy and the jurisprudence of equality

Various theoretical rationales for curbing expression have been offered since the hey-day of the liberal progressive compromise addressed by Coase, with a notable empha-sis on expression as a means to furthering deliberative democracy.29 Nonetheless, the most influential works against offensive expression have arguably emanated from the perspective of equality. Equality concerns have been raised by critical academics in many notable issues of speech, including issues of gender and pornography30 and race and discrimination.31 In regard to hate speech, scholars and advocates associated with the Critical Legal Studies (CLS) movement were instrumental in connecting equality claims with the force of empathy.

Scholars in the CLS tradition often rejected classical liberal premises on the possibil-ity of state neutrality, while some advanced novel methodologies marked by a reliance on personal narratives and experience.32 Such personal accounts are arguably more likely to evoke the sympathy of the reader than the average law journal article on reg-ulatory minutiae, and served to demonstrate serious consequences of hate speech in

28 On the shifting nature of free speech disputes, see Kathleen M.  Sullivan, Free Speech Wars, 48 SMU L. rev. 203 (1994); Delgado, supra note 3, at 169, 170; Steven Gey, The Case Against Postmodern Censorship Theory, 145 U. Pa. L. rev. 195–196 (1996).

29 See generally Cass suNsTeiN, demoCraCy aNd The ProbLem oF Free sPeeCh (1995) (arguing for the embrace of a Madisonian vision of expression as a function of deliberative democracy); oWeN Fiss, The iroNy oF Free sPeeCh (1998) (assessing expression as means to open public debate and dialogue, in which equal partic-ipation is essential).

30 See, e.g., CaTheriNe a. maCkiNNoN, ToWard a FemiNisT Theory oF The sTaTe (1989).31 See generally Derrick A. Bell, Brown v. Board of Education and the Interest­Convergence Dilemma 93 harv.

L. rev. 518 (1980); Cheryl Harris, Whiteness as Property, 106 harv. L. rev. 1709 (1993); Derrick A. Bell, Whose Afraid of Critical Race Theory 1995 u. iLL. L. rev. 893 (1995); riChard deLgado & JeaN sTeFaNCiC, Words ThaT WouNd (2004).

32 See, e.g., PaTriCia J.  WiLLiams, The aLChemy oF raCe aNd righTs (1992) (which includes many personal recounts of confronting discrimination or racism). But see Richard Delgado, On Telling Stories in School: A Reply to Farber and Sherry, 46 vaNd. L. rev. 665, 669 (1993) (observing that approximately only a quarter of the works in the annotated Critical Race Theory bibliography correspond to a narrative mode). See also riChard PosNer, overComiNg LaW 329 (1995) (criticizing critical feminist and critical race scholars, and Williams in particular, for methods argued to be closer to fiction than science).

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

814 I•CON 16 (2018), 806–835

emotional terms. CLS scholars evocatively described the emotional and psychological trauma that can occur through discrimination and the communication of offensive expression. For example, Matsuda argued that negative stereotypes and unconscious discrimination mean that “racial inferiority is planted in our minds as an idea that may hold some truth.”33 Or, as put by Delgado: “[i]incessant depiction of a group as lazy, stupid, and hypersexual – or ornamental for that matter – constructs social real-ity so that members of that group are always one-down.”34

As an example of the impact of empathy on the articulation of the equality model, when the Supreme Court of Canada handed down its landmark decision on the free-dom of expression and hate speech restrictions, 35 R. v. Keegstra, lessons derived from empathy were central to the majority’s reasons. Chief Justice Dickson, writing for the majority, in R. v. Keegstra, referenced prominent scholars associated with CLS, includ-ing Mari Matsuda and Richard Delgado, as evidence of the “growing body of aca-demic writing in the United States which evinces a stronger focus upon the way in which hate propaganda can undermine the very values which free speech is said to protect.”36 Matsuda was further referenced by the Court for the notion that personal harms from hate speech may be independent of majority views. As the Chief Justice wrote: “Even if the message of hate propaganda is outwardly rejected, there is evi-dence that its premise of racial or religious inferiority may persist in a recipient’s mind as an idea that holds some truth, an incipient effect not to be entirely discounted.”37 The concern for the psychological and emotional harm suffered by recipients of hate-ful expression was the primary pillar, in the Supreme Court of Canada’s rationale articulation of the equality model. Indeed, concern for recipients of hate serves as the first of two main rationales for the validity of the regulation.38 Moreover, even when elaborating on the second rationale of societal values, Dickson reverted immediately39

33 Matsuda, supra note 3, at 2320, 2339.34 Delgado, supra note 3, at 171–172.35 In Canada the criminal provision governing hate speech is § 319(2) of the Criminal Code, which makes

it an indictable offense to willfully promote hatred against any identifiable group. There are a number of exceptions and defenses, including private conversations, good-faith religious debate, and truthful statements. None of these defenses are available under the civil proceedings conducted under human rights regimes, although the sanctions are not penal. The federal human rights provision was the Canada Human Rights Act § 13(1) (S.C. 1976–1977, c. 33), which declared it to be a discriminatory practice to communicate or distribute, via telecommunication, any material likely to expose an individual to hatred or contempt for the reason of their belonging to an identifiable group.

Section 13 has subsequently been repealed following a number of controversial and well-publicized cases. See discussion infra section 2.4. Nonetheless, most provinces maintain human rights codes that either closely mirror the federal version or are even broader, and so the regime in Canada remains ef fectively unchanged. For example, Saskatchewan Human Rights Code § 14(1)(b) prohibits the display or publication of any material “that exposes or tends to expose to hatred, ridicules, belittles or otherwise affronts the dignity of any person or class of persons on the basis of a prohibited ground.” The Supreme Court unanimously upheld the provision in Saskatchewan (H.R.C.) v. Whatcott [2013] 1 S.C.R.

36 R. v. Keegstra [1990] 3 S.C.R. 697, 741.37 Id. at 747–748.38 Id. at 746–748.39 After only one paragraph Chief Justice Dickson returned to the topic of individual, as opposed to social,

harm. Id. at 747.

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

Economy of pain: When to regulate offensive expression 815

to justifications based on the individual recipient. The call of empathy has clearly played an essential role in the equality model as articulated in Canada.40

Since Keegstra, social science research has continued to explore nuanced psycho-logical and social manifestations of the harm from hate speech.41 A  deeper under-standing of personal and social consequences may help confirm the need to have hate speech protections in general, but it arguably is of little assistance in regulating at the margins—where the concern is determining when a trivial, disingenuous, or strategic complaint should not succeed.

2.4. Flaws in the equality model—A road paved with good intentions

The margins of offensive expression require attention because the judicial approach in many countries remains squarely concerned with what expression qualifies as offen-sive in the eyes of the majority. For example, James Weinstein has offered a strong critique of modern English jurisprudence on free speech for its tendency to deem insulting speech in public discourse to be deserving of sanction merely because it may discomfort or agitate onlookers. Weinstein observes: “That the core free speech right recognized in English free speech jurisprudence is both too weak and too indetermi-nate to adequately protect the public expression of ideas that ‘offend, shock, or dis-turb’ dominant opinion.”42 Similarly, the following list of controversial instances of regulatory overreach was marshalled in a book review:

In 2009, a member of the Belgian Parliament was convicted of distributing leaflets with the slogans: “Stand up against the Islamification of Belgium,” “Stop the sham integration policy” and “Send non-European job-seekers home.”

In 2006, protesters were convicted of distributing leaflets to Swedish high school students saying homosexuality was a “deviant sexual proclivity,” had “a morally destructive effect on the substance of society” and was responsible for the development of H.I.V. and AIDS.

In 2008, a French cartoonist was convicted of publishing a drawing of the attack on the World Trade Center in a Basque newspaper with the words: “We have all dreamed about it. . . . Hamas did it.” The European Court of Human Rights affirmed all three convictions, rejecting defenses based on freedom of speech.

In Poland, a Catholic magazine was fined $11,000 for inciting “contempt, hostility and mal-ice” by comparing a woman’s abortion to the medical experiments at Auschwitz.

The Dutch politician Geert Wilders was temporarily barred from entering Britain as a “threat to public policy, public security or public health” because he made a movie that called the Koran a “fascist” book and described Islam as a violent religion.

In France, Brigitte Bardot was convicted of publishing a letter to the interior minister stating that Muslims were ruining France.

40 The Canadian model established in Keegstra was unanimously reaffirmed by the Supreme Court in 2013 in Whatcott, supra note 35.

41 For example, the minority stress literature within social psychology examines how the social stigmatiza-tion of belonging to a minority group can produce excess levels of mental stress. See, e.g., Ilan H. Meyer, Prejudice, Social Stress, and Mental Health in Lesbian, Gay, and Bisexual Populations: Conceptual Issues and Research Evidence, 129(5) PsyChoL. buLL. 674 (2003); Bruce Link & Jo Phelan, Conceptualizing Stigma, 27 aNN. rev. soC. 363 (2001).

42 James Weinstein, Extreme Speech, Public Order, and Democracy: Lessons from the Masses, in exTreme sPeeCh aNd demoCraCy 73 (Ivan Hare & James Weinstein eds., 2009).

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

816 I•CON 16 (2018), 806–835

And Canada’s human rights tribunal, which has harassed magazines for anti-Muslim state-ments and for republishing the famous Danish Muhammad cartoons, has been so hostile to freedom of the press that efforts are under way to repeal that nation’s hate speech laws.43

As indicated in Canada’s recent repeal of federal human rights provisions, contro-versial cases of censorship overreach can undermine support for the system. In open societies, instances of censorship and government overreach can easily come to light. What is far less clear, and difficult to establish, are any tangible benefits of the equality model. Consider, for example, that the introduction of a collection of country reports on the status of hate speech regimes contains the rather disconcerting disclosure that there has been no evidence that such laws have had any beneficial impact whatso-ever.44 Similarly, a more recent study of Australian civil hate speech laws finds that despite these protections, “ongoing and significant levels of hate speech” remain.45 Nonetheless, the Australian study also contained the important finding that minority group respondents expressed overwhelming support for the legal protections, taking comfort in their mere existence, as a symbolic value.46 Although this symbolic value is an intangible benefit, it is certainly worthy of preservation, especially if the tangible costs of chilling expression do not overwhelm the intangible gains of regulation.

Preserving the intangible benefits of hate speech regulation may well depend upon a consideration of how to limit the chilling of expression under hate speech regula-tions, especially when government assumes an active role, and the cost, of investigat-ing and prosecuting claims of offense. In order to demonstrate the uniqueness of the marketplace of ideas and the vulnerability of expressive supply, the following section examines the Danish cartoon controversy alluded to in the above list of offenses to expression that referred to Canada. To begin, it should be noted that only three pub-lishers reprinted the Danish cartoons in Canada, and none of these was a mainstream outlet. One, a small university student newspaper, was almost immediately barred from distributing copies on campus. Security guards were even directed to gather up loose copies that had already been distributed on campus.47 The other two publishers had complaints lodged against them with human rights commissions. One of these disputes was settled in mediation.48 The lone holdout was the publisher of a small,

43 Michael McConnell, You Can’t Say That: “The Harm in Hate Speech” by Jeremy Waldron, N.y. Times suNday book revieW, June 24, 2012., at BR14.

44 saNdra CoLiver (ed.), sTrikiNg a baLaNCe: haTe sPeeCh, Freedom oF sPeeCh aNd NoN-disCrimiNaTioN 7 (1992) (“Most papers share the view that laws which restrict free expression do not reduce hatred or violence. Others point to the lack of empirical evidence as to the relationship between hate speech and acts of vio-lence. . . . Beyond the declaratory effects of laws which outlaw racist statements, what evidence can be pointed to that, at least in isolation, suppression has deterred racism, intolerance and bigotry? There is evidence in this book of the abuse of restrictions which would justify the conclusion that little is gained and much is put at risk by punishing the expression of ideas however loathsome.”).

45 Katharine Gelber & Luke McNamara, The Effects of Civil Hate Speech Laws: Lessons from Australia, 49 L. & soC’y rev. 631, 658 (2015).

46 Id. at 654–655.47 H. Wade MacLauchlan, The Cartoon Controversy and the Learning Environment, uPei aLumNi NeWsLeTTer (Feb.,

2006), http://remember.upei.ca/newsletter/2006/02/cartoon-controversy-and-learning-environment.48 See Alta. Rejects Muslims’ Complaint Against Danish Cartoons, NaTioNaL PosT, Aug. 6, 2008, http://www.

canada.com/ottawacitizen/story.html?id=f3e1ae31-33c4-48b8-a2a8-4604184fc690.

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

Economy of pain: When to regulate offensive expression 817

now defunct newspaper, the Western Standard, published by Ezra Levant, a lawyer and well-known free speech advocate. After a year-long investigation, the Alberta Human Rights Commission held a hearing and then ultimately concluded that publication of the images did not rise to the level of restricted hate speech.49 Levant claimed that his defense cost him thousands of dollars and 100 hours of time.50 Throughout the entire episode the images were readily available online.

The harassed Canadian magazine referred to by McConnell was undoubtedly Maclean’s. Maclean’s, a mainstream media outlet, published an article Why the Future Belongs to Islam, a book excerpt from the author Mark Steyn.51 The Canadian Islamic Congress claimed that the article exposed Muslims to hatred and vilification, and issued complaints to human rights commissions in various Canadian jurisdictions. Each of the three human rights bodies that considered the complaints rejected them; but not before the BC Human rights tribunal issued a decision that commented at length on the offensive content of the article.52 Another commission declared that it lacked jurisdiction to hear the matter but nonetheless felt compelled to issue a public statement that condemned Maclean’s article as Islamophobic.53

Unsurprisingly, the lawyer who represented the complainant declared that he was “delighted” by the strong stance taken against the offensive content.54 Furthermore, a law student who worked on the file for the Islamic Congress (and who was also curiously able to deliver testimony at a tribunal hearing) boasted afterward that the complaint had achieved its objective, “to increase the cost of publishing anti-Islamic material.”55 The cost of Maclean’s defense, estimated by the student, was approxi-mately $2 million.56 Whatever the actual financial cost to Maclean’s, the total social

49 Id.50 Graeme Morton, Muslim Leader Drops Ezra Levant Cartoon Complaint, NaTioNaL PosT, Feb. 12, 2008, http://

www.nationalpost.com/news/canada/story.html?id=303895.51 Mark Steyn, The Future Belongs to Islam, maCLeaN’s, Oct. 20, 2006, http://www.macleans.ca/culture/

the-future-belongs-to-islam/.52 See Charlie Gillis, Section 13: How the Battle for Free Speech Was Won, maCLeaN’s, June 19, 2012, http://

www2.macleans.ca/2012/06/19/five-years-two-tribunals-a-raft-of-secret-hearings-a-supreme-court-challenge-how-the-battle-for-free-speech-was-won/ (noting that even supporters of hate speech provisions were “astounded by the spectacle of a state tribunal reviewing a news  maga-zines’ content”). See also B.C. Human Rights Tribunal May Never Live Down Embarrassing Steyn Fiasco, vaNCouver suN, Oct. 13 2008, http://www.canada.com/vancouversun/news/westcoastnews/story.html?id=6debcaed-7a0a-4021-8079-9492163e7cd4.

53 Press Release, Ontario Human Rights Commission, Commission Statement Regarding Issues Raised by Complaints Against Maclean’s Magazine (Apr. 9 2008), http://www.ohrc.on.ca/en/news_centre/com-mission-statement-concerning-issues-raised-complaints-against-macleans-magazine (“While freedom of expression must be recognized as a cornerstone of a functioning democracy, the Commission strongly condemns the Islamophobic portrayal of Muslims, Arabs, South Asians and indeed any racialized com-munity in the media, such as the Maclean’s article and others like them, as being inconsistent with the values enshrined in our human rights codes”).

54 Joseph Brean, Rights Body Dismisses Maclean’s Case, NaTioNaL PosT, Apr. 9, 2008, http://www.national-post.com/news/story.html?id=433915.

55 Mohamed Marwen Meddah, Ezra Levant Sued in ‘Jihad Chill’ Case, oTTaWa CiTizeN, Jan. 26, 2010, http://blogs.ottawacitizen.com/2010/01/26/ezra-levant-sued-in-jihad-chill-case/.

56 Id.

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

818 I•CON 16 (2018), 806–835

cost from the complaint would also have to include the substantial government costs involved in the investigations and hearings, as well as any chilling effects on expres-sion. Concerns of chilling of expression are particularly acute in situations involv-ing human rights regimes. Censorship is a penalty or a tax applied upon expression. Human rights regimes, such as those in Canada and the United Kingdom, entail that government undertake the full investigative and prosecutorial costs on behalf of soci-ety, while the defendant must bear the costs wholly. Therefore, it is arguably all too easy or tempting for a strategic public participant to impose a cost upon a perceived adversary in debate, by lodging a human rights complaint that the government will fund and which the adversary must bear individually.57

The above circumstances are indicative of how costly even constitutionally permis-sible expression can be under the equality model. It bears stressing that in none of the above examples was the ultimate result a determination of prohibited expression, and no expressive act was punished by the courts. Nonetheless, the above examples certainly produced serious costs for those who voiced the expression, whether in money, time, energy, stress, or social stigma. Under the equality model, even when there is no public liability ultimately found, these costs and inconveniences are a form of public tax that is extracted by government agencies against accused speakers. In the United States, the public costs for the above instances of expression would certainly have been zero.58

Perhaps it is telling that while no mainstream Canadian publishers opted to repub-lish any of the Danish cartoons, to express solidarity in the cause of freedom of expres-sion as many European newspapers had done, at least two relatively sizable American publications, the Philadelphia Inquirer and the New York Sun, did so. The Philadelphia Inquirer experienced a small, short, and peaceful demonstration in front of its head-quarters in response to its republication of one of the Danish cartoons.59 Compare these trivial private costs, of the inconvenience and perhaps some poor public percep-tion in witnessing a smattering of protesters before the publisher’s headquarters, with the costs that could have been levied had but one of the few protesters been able to trigger a humans rights investigation.

Considering the economic dynamics of expression helps to demonstrate how vulner-able its supply may be. While undue harm to a recipient deserves our sincere empathy and vigilance, undue harm to a speaker should evoke our sense of alarm, for some expres-sion is so susceptible to silencing that one instance of overreach conducted in the name of well-intentioned empathy may well threaten untold instances of idea expression that

57 An Australian study found no evidence of chilled expression in the face of that country’s hate speech laws. However, the Australian model is a notably different from its Canadian and UK counterparts, in that hate speech complaints proceed as a private civil action, in which a complainant must bear the expense of bringing an action. See Gelber & McNamara, supra note 45.

58 There may be private costs, as with the student newspaper editor who was fired for the decision to repub-lish the Danish cartoons, ostensibly for breaching editorial protocol for inflammatory material. E.g., Student Editor Sacked over Prophet Cartoons, The guardiaN, Mar. 17, 2006, http://www.theguardian.com/media/2006/mar/17/cartoonprotests.race.

59 David Folkenflik, U.S Media Avoid Publishing Controversial Cartoons, NPr.org (Feb. 27, 2006) (noting the small, civilized protest in front of the Philadelphia Inquirer’s Headquarters after the paper published one of the cartoons), http://www.npr.org/templates/story/story.php?storyId=5193569.

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

Economy of pain: When to regulate offensive expression 819

might have served society. Economic insight on expression as a public good can assist in establishing how significant and concerning the public costs of expressive regulation are.

3. A precarious expressive marketplaceFreedom of expression has often been described as a public good.60 A public good is a problem of market failure, in which the private marketplace does not produce an opti-mal amount of an activity. In the case of public goods, the social benefits of an activity outweigh the individual incentives to provide it, and so a suboptimal outcome results, in which private actors would be better off with a service or good that no single actor is willing or capable of providing.

The public good characteristics of expression may be seen in that that information is often non-excludable and non-rivalrous61—that is, there is often little or no cost to individuals in absorbing a new piece of information, and that knowledge may be shared equally without diminishment. Market failure arises with expression in that there are wide benefits that accrue to society with the production and distribution of information, but because some types of information do not reward individual con-tributors directly or commensurately in relation to the social benefit they provide, valuable information will tend to be underproduced.

The nature of expression as a public good has been used to explain a variety of consti-tutional doctrines.62 First and foremost of these explanations is the extraordinary degree of liberty afforded to expression in comparison with other activities, and certainly in com-parison with the trade in goods and services. The public good qualities of expression may also help rationalize the hierarchy of certain core types of expression, in which expression defined as political, scientific, or artistic is afforded much more protection from regulation than is commercial expression. An economic rationale for this hierarchy would indicate that commercial expression, such as advertising, will normally see a far greater percentage of the benefits claimed by the speaker, who is also the producer of the advertised item.63

Similarly, the public goods rationale has also been applied to the US Supreme Court’s seminal decision in New York Times Co. v. Sullivan, which famously introduced the “actual malice” standard for defamation.64 Essentially, the great disparity between

60 See Daniel A.  Farber, Free Speech Without Romance: Public Choice and the First Amendment, 105 harv. L. rev. 558 (1991); Hugo M. Mialon & Paul H. Rubin, The Economics of the Bill of Rights 10 am. L. & eCoN. rev. 1, 5 (2008); Thomas miCeLi, The eCoNomiC aPProaCh To LaW 318 (2004); roberT CooTer, The sTraTegiC CoNsTiTuTioN 437 (2002).

61 See Ejan MacKaay, Economic Incentives in Markets for Information and Innovation, 13 harv. J.L. & Pub. PoL’y 867, 890–896 (1990) (describing the non-rival and non-exclusive properties of information).

62 Mialon & Rubin, supra note 60, at 8–9.63 See Posner, supra note 12, at 22, 40.64 New York Times v. Sullivan 376 U.S. 254 (1964). See generally Farber, supra note 60, at 569–570 (“if

newspapers were liable for all inaccuracies, they would strike the wrong balance between the costs of inaccurate information and the benefits of producing additional accurate information.”); meLviLLe b. Nimmer, Nimmer oN Freedom oF sPeeCh: a TreaTise oN The Theory oF The FirsT ameNdmeNT 2 (1984) (arguing that the highly protective balance set forth in the New York Times was justified because society as a whole has an interest in the dissemination of information).

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

820 I•CON 16 (2018), 806–835

liability for falsehoods and the great social benefits in truth indicates that the press would be too cautious and report too little of public value because the risks of error are borne disproportionately by the publisher. Recall the Maclean’s example from above. The $2 million cost of proceedings boasted of would have been borne by the publisher alone, whereas the social benefits of news reporting are spread across soci-ety at large. In a simplistic fashion one could imagine a rough calculation: that the dispersal of correct and useful information on a political matter results in gain to a publisher of +5, in the combined benefit to its stakeholders and potential future sales increase as a respected news source, while society at large could be +1000 if people act on the information to produce social gains. However, if the information is incor-rect and the publisher is held liable then the publisher could be −200. Assuming such, then the next opportunity of political revelation may be avoided out of caution. In the event, while the publisher will be 0, rather than −200, society could be −1000.

3.1. Elasticity and chilling expression

Richard Posner used the concept of economic elasticity to interesting effect when dealing with different types of expression and public good characteristics. According to Posner, ideas which have ready substitutes are much more likely to be responsive to the price imposed on expression by government sanction or censorship.65 The ability of substitution in the market for goods generally determines its elasticity—a rise in price of a good that has readily available substitutes will likely mean that consumers opt for competing products. The demand for a readily substituted good will thus be quite elas-tic. In the alternative, when a good has no near substitute or when consumers treat the good as indispensable, then a price increase will not decrease the demand for the good proportionately. Demand for this unique good would be relatively inelastic.

Posner reasons that most types of information will be subject to very elastic demand, and therefore will be much more vulnerable to regulation.66 People may not care that they cannot read a certain fringe or radical publication, and while they may easily substitute the loss of this fringe source with myriad other alternatives, this may not necessarily account for the potential loss of useful information that occurs in the sub-stitution.67 Posner treats expressive regulation as a form of tax that increases the price of expression, which not all producers can bear equally. For instance, commercial expression is said to be more resilient to taxation because there is a more direct benefit that the producer can recoup; while artistic or political benefit often remains diffuse and available to everyone.68

An interesting example Posner utilizes is that of James Joyce’s Ulysses.69 Ulysses initially gained notoriety as an obscene work, and was banned in some jurisdictions,

65 See Posner, supra note 12, at 19 (“The more elastic the demand and supply, the more the tax will cut back the product’s output.”).

66 See Posner supra note 12, at 19–20.67 Id. at 24.68 Posner, supra note 12, at 40.69 Id. at 24.

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

Economy of pain: When to regulate offensive expression 821

including the USA and the UK.70 Posner’s point is that the fringe status of a work likely means that there is little profit margin in publication, and certainly not in an amount that could absorb the tax of censorship.71 That Joyce was supported for much of his career by a patron arguably adds to the argument on the economic fragility of fringe works of genius.72 In terms of elasticity, the greater replaceability in the minds of a cur-rent and potential audience should equate to a greater vulnerability of expression and greater regulatory restraint. Posner observes that while established and financially sub-stantial outlets like the New York Times or CBS are likely able to withstand some height-ened costs for a few contested publication items, a fringe press could conceivably be threatened by the costs of censorship, so that Ulysses would not be distributed at all.73

Posner’s rationale based on elasticity could certainly be applied to the above men-tioned Maclean’s case. Indeed, the American author of the contested article, Mark Steyn, commented afterward that he would have been much more vulnerable to the complaints against him had not Maclean’s been part of, and supported by, a huge media conglomerate.74 While Steyn’s comments may support Posner’s point on the greater financial resiliency of large firms over small, it also implies that there is greater trepida-tion among government censors in venturing to sanction mainstream outlets, which may voice their dissatisfaction far more strongly and widely than others. It is surely no coincidence that calls to repeal the hate speech sections of Canadian human rights legislation gained momentum during and after the Maclean’s case, when persistent and critical attention was paid to the issue of censor overreach by the national media.75

4. The elusive economic valuation of expressionThe most prominent attempt to apply an economic analysis to expression arguably remains Posner’s.76 Posner’s method offers a form of cost-benefit analysis for expres-sion, drawing upon Justice Learned Hand’s formula for expression from United States v. Dennis,77 which in turn mirrored Hand’s more famous formula for determining tort liability.78 Hand’s algebraic formula indicates that the regulation of expression, or

70 However, the trial decision of Judge Woosley would become a landmark on the censorship of the obscene, indicating that the determination of the obscene had to be conducted for a work in its whole, not in a fragmented view. United States v. One Book Called “Ulysses,” 5 F. Supp. 182 (S.D.N.Y 1933), aff ’d, United States v. One Book Entitled Ulysses by James Joyce, 72 F.2d 705 (2d Cir. 1934).

71 Posner supra note 12, at 20.72 See generally PaTriCk ParriNder, James JoyCe 199 (1984) (noting that Joyce had the good fortune to attract

a patron, Harriet Shaw Weaver, who supported him from 1917 until his death.).73 Posner, supra note 12, at 20.74 Maclean’s Wins “Hate Speech’”Challenge in BC, viCToria Times via CaNada.Com (Oct. 11, 2008), http://www.

canada.com/victoriatimescolonist/story.html?id=6934db81-e85b-4ceb-a350-91ca65c8bba9.75 See, e.g., Jonathan Kay, Good Riddance to Section 13 of the Canada Human Rights Act, NaTioNaL PosT, June 7,

2012, http://fullcomment.nationalpost.com/2012/06/07/jonathan-kay-good-riddance-to-section-13- of-the-canadian-human-rights-act/.

76 eyaL zamir & barak mediNa, LaW, eCoNomiCs & moraLiTy 178 (2010) (describing Posner’s work as “the most sophisticated economic analysis of free speech to date.”).

77 183 F.2d 201, 206 (2d. Cir. 1950), aff ’d, 341 U.S. (1951); Posner, supra note 12, at 8.78 United States v. Carroll Towing Co., 159 F.2d 169, 173 (2d Cir. 1947).

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

822 I•CON 16 (2018), 806–835

censorship, should occur only if B < PL, where B is the cost of regulation (including loss of valuable information), P is the probability that the expression will result in harm, and L is the magnitude of the harm.79 Posner further refines Hand’s formula into: V + E < P × L/(1 + i)n,80 wherein V is for value of the social loss of the regulation, in addition to E, the likelihood of judicial error that will lead to overregulation. Additionally, the cumulative harm of L is discounted for the present value of future harm.81

One can readily see that the elements in the proposed equations are highly specula-tive and arbitrary. For example, the notion of probability when dealing with future and large-scale social upheaval can surely never become mathematically precise. What is the probability that expression calling for the violent overthrow of a liberal democratic government in favor of a communist regime will achieve its object? What would be the social loss from prohibiting the distribution or staging of Shakespeare’s Richard II? Posner acknowledges the arbitrariness of evaluating expressive content82 and instead indicates that his method can serve as a form of heuristic.83 Posner’s method has attracted serious criticisms, which are dealt with extensively elsewhere.84 For the purposes of this inves-tigation, the focus is on the general challenges of a cost-benefit analysis for expression.

4.1. Markets, pragmatism, and expression

An interesting feature of Posner’s approach is how it corresponds with, and seems to extend out of, Posner’s stated legal and judicial philosophy of pragmatism.85 Posner’s approach is pragmatic in both a weak and a strong sense. In the weaker sense, Posner’s approach is pragmatic in advocating for a case-by-case and contextual judicial assess-ment of expression disputes.86 In the stronger sense, Posner’s approach adheres to the tenets of the philosophy of pragmatism,87 as espoused by the likes of John Dewey, William James, and Charles Pierce, and as especially as espoused judicially by Justice Oliver Wendell Holmes Jr.88 The core belief of the philosophy of pragmatism is that truth is not objective but rather is evolving and socially derived. The pragmatism of Holmes may be gleaned from his well-known notions that while new political

79 Posner, supra note 12, at 8.80 Id.81 Id.82 Id. at 9.83 Id. at 8.84 See Peter T. Hammer, Free Speech and the “Acid Bath”: An Evaluation and Critique of Judge Richard Posner’s

Economic Interpretation of the First Amendment, 87 miCh. L. rev. 499 (1988); zamir & mediNa, supra note 76, at 187–190.

85 See, e.g., Richard A. Posner, What Has Pragmatism to Offer Law?, 63 s. CaL. L. rev. 1656 (1990).86 Posner, supra note 12, at 6.87 See, e.g., WiLLiam James, WhaT PragmaTism meaNs 63 (1906) (“The true is the name of whatever proves

itself to be good in the way of belief, and good, too, for definite, assignable reasons.”).88 See Cass R. Sunstein, Lochner’s Legacy, 87 CoLum. L. rev. 879 (1987) (indicating that Holmes’s famous

dissent seemed to exude personal forbearance, and the suppression of his personal predilections, for the sake of allowing social forces to contest each other in the safe confines of the law. What might seem more personally compelling to Holmes was secondary to the systemic need to provide the space for ideas and factions to battle it out in the private realm for supremacy.).

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

Economy of pain: When to regulate offensive expression 823

majorities should get their say,89 they should not be able to lock in their sense of truth to exclude dissent and future attempts at unsettling the accepted state of truth.90

Posner displays an affinity with the philosophy of pragmatism by proposing that the value of an idea is contingent and open to evolving receptiveness—as with a sci-entific or mathematical proposition that may “be so arcane that its actual audience is very small, may have a vast potential audience, comprising all the people who will eventually receive the information (greatly simplified, or perhaps impounded in some good or service).”91 That the value of an idea should not be in the hands of the present is consistent with pragmatism, and has great intuitive appeal considering how many great advances in learning and culture have originated in ideas that once were not only fringe but were roundly rejected or feared by the masses.92

The means by which Posner avoids the pitfalls of a present majority valuation of expres-sion, in an ingenious move, is to identify value as a matter of lost potential audience.93 That an idea be given the chance at gaining influence, the chance to affect the minds and temperaments of others, no matter how remote its likelihood of impact, is an admirable premise and one that is in line with the pragmatic tenet that truth evolves, and that radical dissent in one generation may become received wisdom for the next. Yet, as appealing as it may sound to ensure an idea’s chance of influence, it must be given procedural purpose. For Posner, the pragmatic modesty on the future value of ideas is furthered by the ideal of content neutrality. Accordingly, prohibitions on specific content are the most troubling restrictions because they attempt to extinguish an idea’s dispersal altogether. Far less con-cerning for Posner are restrictions that merely limit one outlet for an idea.94

Posner examines the infamous Skokie parade proposal as an example of his method in potential application. A  group of neo-Nazis planned to parade in their regalia through a predominantly Jewish neighborhood, whose residents included many holo-caust survivors.95 The planned expression was obviously extremely odious, and for this reason a permit was refused by local officials on content grounds, which subse-quently was deemed an unconstitutional violation of viewpoint neutrality.96 Posner’s

89 As Holmes famously wrote: “[i]f in the long run the beliefs expressed in proletarian dictatorship are des-tined to be accepted by the dominant forces of the community, the only meaning of free speech is that they should be given their chance and have their way.” Gitlow v. New York, 268 U.S. 652, 673 (1925) (Holmes, J., dissenting).

90 Abrams v. United States, 250 U.S. 625, 626 (1919) (Holmes, J., dissenting) (“When men have realized that time has upset many fighting faiths, they may come to believe even more than they believe the very foundations of their own conduct that the ultimate good desired is better reached by the free trade in ideas—the best test of truth is the power of the thought to get itself accepted in the competition of the market, and that is truth is the only ground upon which their wishes safely can be carried out.”).

91 Posner, supra note 12, at 12.92 See, e.g., Mill supra note 7.93 Posner, supra note 12, at 8.94 Id. at 16–18, 30.95 Id. at 30.96 On the Skokie parade and freedom of expression implications, see generally Irving Louis Horowitz, First

Amendment Blues: On Downs, “Nazis in Skokie,” 1986 am. b. FouNd. res. J. 585 (1986); Irving Louis Horowitz & Victoria Curtis Bramson, Skokie, the ACLU and the Endurance of Democratic Theory, 43 LaW & CoNTemP. Probs. 328 (1979); aryeh Neier, deFeNdiNg my eNemy: ameriCaN Nazis, The skokie Case, aNd The risks oF Freedom (1979).

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

824 I•CON 16 (2018), 806–835

analysis of the Skokie example illustrates that a balanced approach to expression is difficult to achieve.

Posner seeks to avoid content valuation, but he also states that expressive liberty absolutism is ridiculous, for saving millions at the expense of only the slightest curtail-ment of expression is more than justified.97 Posner presumes that the probability of harm would be quite high in the Skokie case, which is only reasonable given the emo-tional intensity or trauma Nazi paraphernalia could invoke for holocaust survivors.98 More interesting is the other side of the equation, where Posner assigns the proposed parade a very low (V) value.99 In terms of a cost-benefit analysis, it would then appear that the avoided harm far outweighs the loss from restricting expression. In order to maintain content neutrality, and not simply to declare the content prohibited because it is manifestly vile, requires Posner to declare a very low potential lost audience.

In part, Posner ultimately assigns the parade a low social value score because: (i) the expressive type was mainly emotive and not plausibly intended for the dissemina-tion of ideas,100 and (ii) because the rejected parade permit application only involved a narrow geographic restriction.101 Posner’s reasons reveal how interpretative and open to manipulation a view to audience is. First, as others have observed, it is highly arbitrary to deem the intended audience to be one municipality alone, when the hate mongers could have sought to attract news coverage and national attention with their odious parade.102 It is conceivable to render many expressive acts in the hope of an eventual global audience—should this be sufficient to produce a huge factor of V, which only the largest levels of social harm could outweigh? Was the intended audi-ence for Ulysses a small literary elite, or every literate person?

Furthermore, ex ante censorship in one jurisdiction should simply be read as “restricting one mode of expression in one small area of the country.”103 Yet, this begs the question—if one jurisdiction restricts the content why would not other jurisdic-tions not do likewise? Conceivably one municipality would not wish to appear more receptive than its neighbors to demonstrations of vile and hateful expression. If one municipality refuses the neo-Nazi parade, it is certainly possible that others would do likewise, thereby rendering the prohibition de facto national in scope.

Considering a hypothetical with similar audience proportions but with a far less emotionally charged message helps to illustrate the malleability of Posner’s method. Suppose a small group of locals decide to protest outside the gated environs of an exclu-sive community that houses the oligarchs who rule over the nearby city and extract great wealth in doing so. To begin with the harm side of the equation, challenging the gilded position of the elite can hardly be equated with the emotional trauma of holocaust survivors in confronting a Nazi parade. However, even on this side of the

97 Posner, supra note 12, at 6.98 Id. at 30–31.99 Id. at 30.100 Id. at 31.101 Id. at 30.102 See Hammer, supra note 84, at 521.103 Posner, supra note 12, at 30.

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

Economy of pain: When to regulate offensive expression 825

equation it may be open to manipulate the measure of impact and harm. The pro-test against inequality may well instill serious fear in the targeted inhabitants. Even if these inhabitants are generally unsympathetic, their response should be accounted for in Posner’s method.

The more serious problem posed by the oligarchy protest occurs on the value side of the equation. Mirroring the Skokie case, it may be said that the expression is both emotive and geographically limited, and therefore deserving of a very small value score. Surely the protesters could not be thought to be trying to convince their imme-diate audience—it is not often that the displayed dissatisfaction of the poor convinces the rich and powerful to surrender up their position voluntarily. Furthermore, if the municipal council, which unsurprisingly reflects the interests of the oligarchs, pro-hibits protests outside the community gates, this could be described as a very narrow geographic restriction. The protester’s idea could surely get out in different modes or in different locales, so there is no real potential audience loss, or V value, in the pro-hibition. Yet, this scenario seems as if it would surely distort some basic essence of political protest and dissent.

To use the lack of receptiveness of an immediate audience to classify expression as mainly emotive and less deserving of protection is a frightening proposition. It could mean there could be no protest in Alberta’s oil sands, for instance, because oil com-pany executives may not be open to persuasion. More important, it risks distortion of a protest message by relegating the message to other modes or venues. The essence of Posner’s analysis of value seems to be based on the notion that if one tiny locale restricts the expression, it does not mean that idea cannot get out at all. But what if proximity is part of the message? It might be reasonably supposed that the oligarchy protesters desire not only to criticize the state of inequality in their society but also to demonstrate before the most recognizable symbol of this inequality, perhaps in a desire to embolden others. To lessen the value of the protest by indicating that it can proceed in a different manner could potentially rob an expressive act of a core portion of its intended meaning.

A further difficulty with Posner’s method is that it may unduly promote the so-called heckler’s veto,104 in which the threat of a violent reaction and the need for public order is utilized to silence a speaker. With the Danish cartoon controversy, for example, a low potential audience loss in the censorship of one tiny publication could be identified, joined with the claim that so many different outlets for the expression existed. On the harm side of the equation, on the other hand, a high probability of substantial harm is easy to discern, especially with the benefit of hindsight, as the controversy descended into widespread violence and loss of life. However, hindsight of violence is not necessary to demonstrate the heckler’s veto in the case of the Danish cartoons. For example, when

104 The heckler’s veto generally refers to the potential for an agitated listener to silence a speaker by providing the appearance of imminent violence or disorder. The phrase “heckler’s veto” is commonly assumed to derive from Justice Black’s dissent in Feiner v. New York 340 U.S. 315, 326–329 (1951) (Black, J., dissent-ing). However, Professor H. K. Kalven was credited with coining the memorable term. See Owen M. Fiss, Free Speech and Social Structure, 71 ioWa L. rev. 1405, 1416–1417 (1986).

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

826 I•CON 16 (2018), 806–835

Yale University Press consulted with experts on the risks of republishing the cartoons in an academic book on the controversy, the responses were adamant enough on the chance of triggering global violence that the editors decided to pull the images from the book.105 Regardless of the merits of the re-publication decision, this example indicates that the prospect of violence can dramatically overwhelm the cost-benefit analysis for expression, especially when alternative sources of idea distribution can be identified.106

5. The elasticity of expressive supplyIf calculating harm on a societal level allows the majority to silence discomforting individuals, but acting in the name of individual psychological harm opens up the prospect for the administration of hurt feelings and bad taste, achieving a balanced method appears to be a worthwhile but difficult task. The proposed method of elastic-ity concentrates on individual recipients of offensive expression, as with the equality model, but does so in a manner that is arguably much more constrained and intelli-gible. The first step is to center the method on a concern for individual recipients as a matter of social utility. Sanctions should be invoked only when an instance of offen-sive expression decreases social utility by lessening the amount of potential contribu-tors to public discourse. This stipulation of social utility plays two important roles: (i) it avoids a primary focus on the emotional or psychological harm a recipient has received, and (ii) it embraces intellectual dynamism and the pragmatic recognition that knowledge and truth are constantly evolving. The object for protection is poten-tial participation, not the present valuation of expressed content. Therefore, it is the possibility of knowledge through individual participation or contributions that mat-ters, not the actual increase of anticipated knowledge or usefulness. Knowledge can-not be assigned a value in an ex ante fashion, and the attempt to do so would violate the content neutrality premise of elasticity.

Though it may sound harsh to render concern for an individual’s psychological or emotional well-being a feature of social utility, it should be stressed that this is a fea-ture of implementation, not justification. A society that adopts a regulatory regime to curb extreme instances of offensive expression will no doubt be inspired in large part by feelings of empathy. Great social strides have often depended greatly on human empathy. Yet, the why of regulation should not overwhelm the how in the case of expression. Again, while the justification for regulating expression no doubt emanates from empathy, empathy should not be the guiding notion of the system’s articulation. Elasticity provides a means for regulating expression in a more disciplined manner, even as empathy explains why we choose to regulate expression in the first place.

To prevent problems of idea utilitarianism, wherein majority preferences could pro-hibit individual dissent on the basis of widespread dislike or discomfort, social utility

105 See Patricia Cohen, Yale Bans Images of Muhammad in New Book, N.y. Times, Aug. 12 2009, http://www.nytimes.com/2009/08/13/books/13book.html.

106 Yale Press’s editorial decision to remove the cartoons from the book in press was apparently eased by the fact that the images were available readily online. See Cohen, supra note 105.

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

Economy of pain: When to regulate offensive expression 827

is defined to require content neutrality. Knowledge evolves, and so it is modest and prudent to not extinguish an idea that is today radical but tomorrow may be received wisdom. In keeping with Holmes’s marketplace of ideas metaphor, in which ideas must compete to convince the audience of their validity, optimality is defined as the greatest possible number of idea contributions to compete as truth. Sub-optimality, accordingly, occurs when one instance of hateful expression causes the loss of one or more other contributions.

Expressive optimality may be identified by modifying a concept of elasticity bor-rowed from economic theory. It is possible to recognize sub-optimality when the supply of potential contributions becomes elastic. Supply as used here is intended to denote conceivable participation in the public exchange of ideas. Supply is foreclosed for an individual when that individual no longer believes he or she can contribute publicly because their ideas, regardless of their merit, cannot overcome the contempt felt for the author’s identity. Maintaining the supply of participation can be thought of as a feature of reasonable resiliency, or elasticity.

Elasticity refers to the responsiveness of one variable to a change in another. Elasticity is usually found in competitive markets, as an increase in price will usually decrease the demand for a product. Whereas elasticity is desirable for a competitive market in goods, as it indicates that competitive substitutes are available, for expression it would be the opposite. When potential supply is inelastic, it means that potential individual contributions are unresponsive to increases in offensive expression. Offensive expres-sion is a cost imposed on recipients and potential contributors. Elasticity occurs when reasonable recipients of offensive expression would curtail their participation in soci-ety in response to the harms imposed or the threats latent in a particular environment. Harms and threats that result in elasticity, and a retreat from public space, should not be tolerated and should trigger government intervention.

5.1. Context and a reasonable retreat

The first step of the inquiry under elasticity is to question whether the offensive con-tent would likely lead a reasonable person situated in the position of the targeted recipient to retreat from future participation. The specifics of the individual recipient’s capacities should not be examined, for this would again entail a present and majoritar-ian valuation. The object of elasticity is not to protect present value in the marketplace but to preserve the potential for open discourse with the fullest participation possible. Concentrating on what would cause a reasonable person to retreat from public partic-ipation when faced with offensive expression has several advantages. First, the view to a reasonable retreat permits for content neutrality. Those whose expression soci-ety values most, such as those who translate comic books into film, should not count more in this assessment than dissenters or those who are currently silent.

Second, the actual present amount of voice loss is not the focus of elasticity but rather future potential public discourse and the hateful expression that may lessen it. Hateful expression could target those who are unlikely or unable to contribute to future public discourse, perhaps because of recipient’s disability, language barrier, or

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

828 I•CON 16 (2018), 806–835

old age, and yet such expression is certainly destructive of public discourse. Not only would a reasonable person retreat when faced with the same hateful expression, but he or she would also retreat from a system that would tolerate such expressive attacks on those who are uniquely vulnerable. Third, and most important, a focus upon a rea-sonable retreat would indicate that the standard for triggering government interven-tion would be far beyond mere offense and hurt feelings. Under elasticity, as opposed to the equality model, it is not simply the individual harm received that is considered, but what a reasonable person does in the face of the harm. Significantly, it is only when we examine reactions, or inaction, in the private marketplace of ideas that we can even identify a prima facie situation of elasticity.

Recalling the Danish cartoon controversy provides an example of how the market-place reaction provides a crucial safeguard against the chilling of expression and the abuses of the equality model. For while the cartoons were clearly offensive to many, or even blasphemous, this alone could hardly cause a reasonable person to shrink from public participation. To begin with, given the small initial circulation in print, no doubt many who were offended by the images would have had to voluntarily discover them online, lessening any sense of personal intimidation. Moreover, the cartoons were widely condemned by world leaders, and, again, many, if not most, mainstream media outlets refused to re-publish the cartoons. Therefore, it is difficult to consider that a majority anywhere, whether a Muslim majority country or not, had endorsed the car-toon contents in a manner that would be indicative of a scorn for Muslims. Indeed, it could be said that many of those who were offended by the cartoons were exercised to protest and debate, in the classic liberal antidote of greeting hateful expression with even more expression. In terms of elasticity, the cartoon controversy manifestly would have not warranted any government intervention.

The private marketplace can indicate majority support for an offended recipient in non-participation or expressions of solidarity, as with the cartoon controversy, but it may also play a more active role in punishing or taxing hateful expression. It is signif-icant to note that elasticity may be inversely related to the amount of cost imposed on offensive expression in the private marketplace. If it is found that prima facie individu-als will reasonably shrink from future participation because of offensive expression aimed at them, this surely is based in large part, if not exclusively, on the susceptibility of humans to the opinion of others. The prima facie harm, and tendency toward a public retreat, then can often be overcome by the reaction of others; especially if the reaction confirms that the majority is indeed receptive to the targeted recipients, and that it is the cruel and offensive speaker who is excluded.

An example of a tax in the private marketplace of ideas could involve a celebrity discovered giving voice to vile invective about a vulnerable minority group. Assume that no member of the minority group was addressed personally, but that he or she could have learned of the communication through celebrity gossip publications or online news. After this revealed behavior it would be surprising if the celebrity did not immediately express abject remorse, either as a genuine sentiment or as a public relations strategy. The key question under elasticity, however, would be whether there were substantial consequences for the offensive celebrity speaker?

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

Economy of pain: When to regulate offensive expression 829

It is not difficult to imagine that the celebrity could have many lucrative opportuni-ties lost because of his or her newly toxic public persona. If the private marketplace reacted in such a way, to the extent that the celebrity surely would pay millions of dollars to undo his or her words, then it is difficult to see how an individual in the tar-geted minority group could feel threatened and isolated, when the majority seemingly aligns with the minority at the expense of the celebrity. Additionally, the cost imposed by the private marketplace would render an additional government sanction superflu-ous. The private cost of the offensive expression in this example would be more than enough to assuage any offended recipient that the majority does not agree with the speaker, and to signal to other like-minded offensive speakers the potential costs.

The example of an unhinged celebrity provides an easy example of private market cost, but perhaps another instance of a publicity-garnering offensive expression could better illustrate the steps of the elasticity method. Suppose a talented college athlete entered a professional sports league known for its exuberant embrace of traditional conceptions of sexuality, and that this athlete publicly came out as gay.107 Suppose that an issue of expression arose because this athlete had to confront extremely offen-sive signs denigrating his or her sexuality and place in the sport. The social utility in voice loss is indicated in this example through the potential loss of an example, role model, or spokesperson for others.

If the athlete were confronted with messages that demeaned his or her sexual ori-entation and his or her right to play the sport because of it, then the situation would likely qualify on the first step of prima facie elasticity. If few to zero gay athletes in the sport’s history have felt comfortable in acknowledging their sexuality, then it would appear that the background context indicates that there is a strong pressure against the transparent participation of all individuals. Perhaps it could be said that, unlike racial or ethnic minorities, persons from some minorities could conceal their differ-ence publicly and proceed with the athletic portion of their participation undetected. Yet, the object of the elasticity method is the avoidance of voice loss. Minority entrants to the sports league might conceal their identity, but this would mean that they could not speak from the position of their own identity; their public participation would have to be divorced from the full range of their own opinions and personal experience. There is voice loss regardless of physical presence or the appearance of participation.

The second step of inquiry would look to the marketplace response. If the message is held by a lone taunting fan, who is almost immediately shouted down by a stadium full of spectators, the presumptive elasticity of supply would be overcome by the mar-ketplace. The player could not reasonably feel that the lone sign-holder represented an unwelcoming public, since the nearest majority so clearly sided with him or her. Yet, in another variation, the likelihood of withdrawal would arguably be increased if the

107 This hypothetical obviously draws on actual events. See, e.g., Chris Connolly, Mizzou’s Michael Says He’s Gay, esPN.Com (Feb. 10, 2014), http://espn.go.com/espn/otl/story/_/id/10429030/michael-sam-mis-souri-tigers-says-gay; Jason Collins Comes Out: NBA Player Becomes First Openly Gay Professional Male Athlete in Major Sport, huFFiNgToN PosT, Apr. 29, 2013, http://www.huffingtonpost.com/2013/04/29/jason-collins-comes-out-gay_n_3178401.html.

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

830 I•CON 16 (2018), 806–835

message was the attempted intimidation or bullying by a teammate(s), which would only confirm the concerns or fears of other minority athletes.

Even if the athlete visibly ignored the insult, through force of character or will, this should not diminish the needed attempt to incorporate a price on this expression. The externality of this offensive expression should be priced so that others, includ-ing potential speakers of offense and especially other minority group members, are aware of how society values such expression. A model of elasticity is concerned with future contribution loss; and, therefore, the strength of a particular recipient should not lessen the tax needed for certain expressive harms, for the next speaker may not be so resilient.

The following section demonstrates and applies the proposed method.

6. Hypothetical applications of the model

6.1. Hypothetical One

Imagine that a lone individual handed out pamphlets in a town square. The pamphlets contained spurious and hateful claims about same-sex relationships and claimed that same-sex couples were unfit to raise children. The cover of the pamphlet did not contain any offensive material, but the title alone, which began “the truth about X,” would have alerted most that conspiratorial and inflammatory content was likely con-tained inside. Every single pamphlet was thrown unopened and unread into a nearby recycling receptacle. Does this situation merit government interference?

This situation could not possibly qualify under the first stage of potential elasticity. How could potential contribution loss occur if no member of the public actually reads the message? Moreover, even if a member of the targeted minority group had read the content this would still not qualify. In context, it is one lone individual distributing the offensive pamphlets, and there is no imposition or intimidation joined with the mes-sage. Additionally, in this situation the offense had to be voluntarily entered into by the minority individual, assuming that the cover is fairly clear that offensive material lies inside. Unless distributed offensive material was rumored to be about a specific person, there would seemingly be little temptation to view vile material for the sake of experiencing offense or emotional pain. Finally, looking to the marketplace response, in which anyone who was witnessed receiving a pamphlet could then be seen imme-diately discarding it, there could hardly be an impression created that the majority is sympathetic to the message.

This hypothetical is intended to demonstrate an essential distinction between a method of elasticity and the equality model. The above expression would likely be punishable under the equality model, which, again, permits the punishment of pub-lication of a detestable idea, period. Under the equality model, as long as anyone can conceivably come across the hateful content, it is sufficient to punish. Recorded mes-sages, websites with disclaimers and click agreement front pages, and a message in a bottle—these may all contain hateful words that may reach no one; and yet, under the

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

Economy of pain: When to regulate offensive expression 831

equality model, government agencies can pursue and sanction the authors of such ineffective, or even unheard, expression.108

6.2. Hypothetical two

Instead of pamphlets, imagine that a lone individual in the square is protesting on the issue of same-sex relationships with the use of large signs that are visible from some distance. The threat of elasticity rises in this hypothetical through the introduction of involuntari-ness. People out shopping or enjoying the fresh air may come across ideas they find offen-sive and without warning. There would be the potential for people to avoid the offense by walking away, but the potential for harm remains. Context of the offense is obviously key.

If most townspeople are expected to gather in the town square over a holiday week-end, and the town has witnessed increasing levels of homophobia culminating in the passage of anti-gay laws, the prohibition of political symbols in the square would argu-ably affect certain protestors much more than others. When the majority is seething against a minority the state should protect the display of minority flags, rather than encourage majority prejudice by barring all flags for the sake of limiting the one that has a special meaning for the vulnerable. So, a lone speaker that dissatisfies the public is a poor guide.

From a view to elasticity, it would be likely that prior context could determine whether it is reasonable for a  minority group member to shrink from participation because of the display of a few signs in the town square. The assessment could begin by inquiring whether the proselytizing individual is a fringe speaker who discom-forts and offends nearly everyone else. This initial question would essentially address whether the community has an established tradition of tolerance. Potential contri-bution could be presumed to be inelastic if, for example, everyone in context would reasonably know that the speaker is a fringe and unrepresentative crank that does not speak for anyone, much less the majority.

If the majority is safely on the side of the targeted minority, then the lone speaker might well be ignored in the town square without private consequence, and there would not be the need to discern a private marketplace cost imposed. Although some individuals would undoubtedly give voice to their disgust at the expression, which would surely strengthen the inelasticity of minority members who witness this, it might not impact the speaker overly much. If the speaker is insistent then it may also be that he or she would be resistant to multiple forms of sanction in any event, for they likely have little social standing at stake.

Logically, if elasticity can be implied because of prior context, as in the growth of flagrant intimidation or even violence committed against same-sex couples, then the private marketplace could not be expected to provide adequate response. The growth

108 For instance, in Taylor, a decision handed down alongside Keegstra, the Supreme Court of Canada held that handing out cards which directed people to phone a number to reach vile recorded messages con-stituted hate speech. That people had to voluntarily call the number and then remain on the line did not diminish the threat of the message apparently. Canada (Human Rights Commission) v. Taylor [1990] 3 SCR 892, 903–904.

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

832 I•CON 16 (2018), 806–835

in flagrant abuse of a vulnerable minority arguably occurs precisely because the out-rageous offenders were emboldened by the latent permissiveness and prejudice of a surrounding majority. Elasticity should be inversely related to private costs on offen-sive expression: if there is a substantial reaction from a majority, a minority individ-ual will feel less alienated from it. Not only will a prejudicial majority not impose the proper cost on expression that causes elasticity, but it may well increase the harm—perhaps if the incident of the lone individual proselytizing is met by groups yelling their support, or if others join in the display. Government intervention must correct this private subsidy of hate and do so swiftly.

6.3. Hypothetical three

Preceding examples have contemplated minority individuals encountering hateful expression in a public space, but it would be useful to also consider instances in which the offense moves toward the recipients and consider how this may affect elasticity. Suppose the proselytizing individual comes to the town square with his or her vile mes-sage knowing full well that the square will be occupied by same-sex couple celebrants of a national same-sex relationship day (imagine further that the holiday has a better title). This would likely be an example of when elasticity could not be presumed and for which the market response would be adequate. The proselytizer could yell at the crowd of thousands, and of those who noticed him or her did so in a collection of indiffer-ence, outrage, scorn, or mockery. A participant in the festivities, or an onlooker, would unlikely feel that the one raving and widely derided person was a cause to retreat from public life. There is no need to look for private market costs because there should be no elasticity; or, put differently, the market cost is sufficient because in this context the targeted group was the vast majority and their market response would likely signal to onlookers what the unwelcome minority viewpoint was.

To illustrate the different dynamic of elasticity presented by intimidation, a stark example would be of hateful messages directed at a residence.109 For most, it is likely difficult to imagine a more vulnerable location than one’s home, where one sleeps and shelters that which one cares of most in the world. It is perhaps possible to rationalize a hateful display that targets one’s home by discounting for randomness and telling one-self that it was the work of a few wretched individuals. On the other hand, it would not be irrational to be fearful of, or at least deeply disappointed in, one’s community. Someone who suffers the trauma of her home targeted by hate might not be able to return and to function publicly just as before. The event could indicate that there is a serious threat in that a hate monger has knowledge of one’s family and where they reside.

It would not be outlandish to think that a reasonable person may want a lower pub-lic profile if it meant lessening the exposure of his or her family to fear or potential

109 Cross-burning provides a particularly stark contrast of the differing regulatory approaches between the USA and Canada. See R.A.V.  v.  City of St. Paul, 505 U.S. 377 (1992). Cf. R.  v. A.B. [2012] N.S.J. No. 227 2012 NSPC 31 (recognizing cross burning as punishable as a hate crime). See generally Roy Leeper, Keegstra and R.A.V.: A Comparative Analysis of the Canadian and U.S. Approaches to Hate Speech Legislation 5 CommC’N L. & PoL’y 295 (2000); Schauer, supra note 4, at 5.

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

Economy of pain: When to regulate offensive expression 833

physical harm. Surely the terrifying nature of personalized harm, directed at where one feels most vulnerable, can overwhelm a background context of living in liberal democracy with no significant or recent history of violence against minority groups. Elasticity can occur because we are frightened by what is possible in a community we previously assumed was safe and tolerant.

If offensive expression joined with intimidation may displace the importance of background context, it is also important to recognize that not everyone has equal knowledge of, or certainty in, a society’s level of tolerance. A site housing refugees, for example, could be described as a setting of very high potential elasticity. Refugees no doubt occupy a very precarious position, between two societies, with the former often a place of serious hazard. Stemming out of this precariousness, refugees might well feel a heightened sensitivity to public pressures against their presence.

Suppose a liberal democratic government agreed on humanitarian grounds to receive a group of refugees from an international conflict area, and temporarily housed them in a certain facility or housing center.110 Imagine next that one of the first impressions that these refugees have of this host country is that of a hate group protesting with vile messages at the gates of the facility. It would be reasonable to assume that people thus confronted could develop a fragile sense of their worth or welcome in the host society. And the most vulnerable, and susceptible to future recoil from participation, in this context would likely be children; who could be more suscep-tible to the words or images that devalue their identity, background, and family.

Elasticity could occur in this situation since a small child, at a formative age, is given the lesson of being shouted at and condemned viciously because of nothing more than the identity of his or her parents. This type of circumstance is unlikely to find a subse-quent private marketplace reaction that could possibly compensate for the harshness and concentration of the message. A large group of supportive parents overwhelming the voices of the protesters would qualify as a private market response that eased the tendency of elasticity, but it is entirely unlikely without sufficient foreknowledge and a means to organize.111 Instead, this situation would be one where there is an easy finding of egregious offense but with little to no chance of private response to equalize and overcome the effect. Serious expressive harms may occur in the absence of pub-lic awareness, and the content of which almost no member of the community would condone or tolerate, had they been given the opportunity. Government sanction in this context would surely correct a glaring instance of social market failure.

110 For example, in 1997, protesters congregated outside a Toronto-area motel that at the time housed Roma refugees as they awaited the resolution of their requests for refuge status in Canada. Most of the protest-ers wore the garb associated with white supremacists. Nazi flags were waved, chants of “white power” went up, and the “Sieg Heil” Nazi salute was made. Some of the mob also held signs that read “Honk if you hate Gypsies,” “Canada is not a Trash Can,” and “You’re a cancer to Canada.” R. v. Krymowski, [2005] 1 S.C.R. 101, 104.

111 Admittedly society’s heightened protectiveness of children would mean that in many circumstances that the private market response would be very robust. Perhaps it could be said that children can present the extremes of elasticity—more vulnerable to silencing in isolated instances but also more likely to garner the strongest support and protection from both government and the public.

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

834 I•CON 16 (2018), 806–835

7. ConclusionA special concern with the vulnerability of children is one instance where elasticity finds overlap with existing constitutional doctrine in some states, as in the greater restrictions of expression permitted in schools that are justified by the aim of main-taining a stable and receptive educational environment.112 A  more recent develop-ment in this regard involves widespread attempts to restrict cyber bullying, which is a telling instance of when an intuitive and empathic desire to protect the vulner able in our midst overlaps with the objective of elasticity—individual resiliency and future participation. As mentioned previously, empathy should of course be foundational to the justification for regulation, but the borrowed economic notion of elasticity may provide a more concrete means of implementation and adjudication. Elasticity is threatened in adolescent bullying situations because vulnerable adolescents can so completely locate their sense of self-worth in a small community of opinion. Externalities in this context, when a speaker seeks a base sense of improved worth by destroying the reputation of another, are devastating. Private marketplace costs, if they occur, are unpredictable, likely too late and insufficient. For example, the crusad-ing hacker movement Anonymous may give vent to collective support for a victim and moral outrage at his or her victimization, and in doing so also impose social costs on the offensive speaker. In the most inspiring fashion, the intervention may even assist a victim contemplating suicide.113 Unfortunately, for this movement’s attention to have been drawn to a particular instance probably means that the situation is already so worthy of outrage that it may well be past the point of curing the harm to a victim’s sense of worth, and often tragically so.114

Information technology has altered fundamentally the scope of communicative interconnectedness. Substantial dangers certainly abound, but most do not qualify

112 The U.S. Supreme Court’s leading decision on freedom of expression in a school setting arguably remains Tinker v. Des Moines Indep. Cmty. Sch. Dist., 393 U.S. 503 (1969) (which famously held that students do not “shed their constitutional rights to freedom of speech or expression at the schoolhouse gate” (id. at 506). Nonetheless, the school setting is not afforded the same expressive liberty as other contexts, and school administrators may validly restrict student expression that: “materially disrupts classwork or involves sub-stantial disorder” (id. at 513) or “colli[des] with the rights of other students to be secure and to be let alone” (id. at 508). See generally Bonnie A. Kellman, Tinkering with Tinker: Protecting the First Amendment in Public Schools 85 NoTre dame L. rev. 367. For a recent decision that endorsed student advocacy in a school, see B.H. v. Easton Area Sch. Dist., No. 11–2067, 2013 WL 3970093 at 5–6 (3d Cir. Aug. 5, 2013).

113 See, e.g. Emily Bazelon, What Do Anonymous and MTV Have in Common?: They’re Trying to Protect Kids Online, sLaTe.Com (Nov. 12, 2012) (recounting how Anonymous and Rustle League intervened when a girl was being encouraged to commit suicide online, and threatened the bullies with disclosure and expressed moral support for their target. She thankfully was able to see this support.), 1.1134233http://www.slate.com/articles/news_and_politics/bulle/2012/11/anonymous_and_mtv_trying_to_protect_kids_online.html; Laura Bell, Anonymous Exposes Bullies Who Encouraged Teen Girl to Commit Suicide, JezebeL (Nov. 15, 2012), http://jezebel.com/5960767/anonymous-exposes-bullies-who-encouraged-teen-girl-to-commit-suicide.

114 See, e.g., Cevan Sieczkowksi, Amanda Todd’s Alleged Bully Named by Anonymous After Teen’s Tragic Suicide, huFFiNgToN PosT, Oct. 16, 2010, http://www.huffingtonpost.com/2012/10/16/amanda-todd-bully-anonymous-suicide_n_1969792.html. See also David Kushner, Anonymous v. Steubenville, roLLiNgsToNe.Com (Nov. 27, 2013) (“Online vigilante Deric Lostutter helped expose the cover-up in the Steubenville rape case. Now he’s facing more jail time than the convicted rapists.”), http://www.rollingstone.com/culture/news/anonymous-vs-steubenville-20131127.

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022

Economy of pain: When to regulate offensive expression 835

for intervention under a model of elasticity. A child forced to witness a hysterical mob aimed against his or her identity, or adolescents who are tormented so that they feel they have no worth and others would prefer it if they did not exist, are situations that can lead to voice loss. An online video of one lone deranged person burning a cul-tural symbol, which has been condemned by politicians universally, and protested in streets across the globe, is dangerous expression,115 but not expression that should be prohibited under elasticity. When offensive expression shutters peoples in their homes in fear, speech supply has become elastic, and government intervention is required to assuage this fear. When people move to the streets to peacefully protest against offen-sive expression, their reaction is inelastic, and the marketplace of ideas is apparently functioning quite well.

115 Yaroslav Trofimov & Maria Abi-Habib, Petraeus Says Quran Burning Endangers War Effort, WaLL sT. J., Apr. 4, 2011, http://www.wsj.com/articles/SB10001424052748703806304576240643831942006.

Dow

nloaded from https://academ

ic.oup.com/icon/article/16/3/806/5165821 by guest on 11 August 2022