Few provisions of Commonwealth law have generated as much sustained public argument as section 18C of the Racial Discrimination Act 1975 (Cth). Inserted by the Racial Hatred Act 1995 (Cth), the provision makes it unlawful to do a public act that is reasonably likely to offend, insult, humiliate or intimidate a person or a group where the act is done because of race, colour or national or ethnic origin. To its critics the provision licenses state censorship with no place in a liberal democracy; to its defenders it is a modest civil safeguard of equal membership in one of the world’s most culturally diverse societies. The dispute has repeatedly entered national politics, most visibly after Eatock v Bolt (2011), the abandoned repeal proposal of 2014, and the 2017 parliamentary inquiry that followed complaints against Queensland university students and a newspaper cartoonist.
This essay asks whether the balance Part IIA strikes between protection from racial vilification and freedom of expression is defensible. It first sets out the architecture of sections 18C and 18D as the courts have actually construed them, then situates the scheme within a constitutional framework that protects political communication by implication rather than through a bill of rights, before weighing dignity-based against liberty-based accounts of expressive freedom and testing both against the American alternative. It concludes that the present settlement is defensible in substance, although its drafting and complaint procedures remain legitimately open to refinement.
The architecture of Part IIA: sections 18C and 18D
Section 18C(1) renders unlawful an act done otherwise than in private if two conditions are met: the act must be reasonably likely, in all the circumstances, to offend, insult, humiliate or intimidate another person or a group of people, and it must be done because of the race, colour or national or ethnic origin of that person or group. Three features deserve emphasis: the prohibition is civil, not criminal, exposing a respondent to conciliation and civil remedies but never punishment, unlike the criminal hate speech laws of many European states (Gelber 2011); it is confined to public conduct; and it is tethered to a racial ground, so offensive speech as such falls outside its reach.
The courts have further narrowed the apparent breadth of the statutory verbs. In Creek v Cairns Post Pty Ltd (2001), Kiefel J held that section 18C is concerned only with ‘profound and serious effects, not to be likened to mere slights’, a threshold applied consistently ever since (Meagher 2004). The test is objective: the question is not whether an unusually sensitive complainant was wounded, but whether an ordinary, reasonable member of the targeted group would probably have been offended, insulted, humiliated or intimidated (Eatock v Bolt (2011)). The operative law therefore addresses serious public denigration rather than hurt feelings, although the gap between text and construction, discussed below, carries costs of its own.
Section 18D then exempts anything said or done reasonably and in good faith in artistic work, in academic or scientific debate, in fair and accurate reporting, or in fair comment expressing a genuinely held belief. The Federal Court has treated it as the point at which Parliament deliberately balanced Part IIA against free expression (Bropho v Human Rights and Equal Opportunity Commission (2004)), and its significance is often underestimated: Andrew Bolt failed not because his opinions about fair-skinned Aboriginal people were unlawful as opinions, but because his columns contained significant errors of fact that deprived him of the reasonableness and good faith on which section 18D insists (Eatock v Bolt (2011); Stone 2015). Procedurally, complaints go first to the Australian Human Rights Commission for conciliation under the Australian Human Rights Commission Act 1986 (Cth); the Commission’s statistics show most racial hatred complaints are conciliated or terminated, and only a small residue ever reaches the courts (Australian Human Rights Commission 2022).
An implied freedom, not a bill of rights
Any evaluation of Part IIA must begin from a structural fact: Australia has no constitutional or national statutory bill of rights, and therefore no free-standing right to freedom of expression. What it has is the implied freedom of political communication, derived in Lange v Australian Broadcasting Corporation (1997) from the system of representative and responsible government established by sections 7 and 24 of the Constitution. The implication confers no personal rights; it restrains legislative and executive power, invalidating laws that impermissibly burden communication about governmental and political matters. Under the structured proportionality analysis now favoured by the High Court, a burdensome law survives if it pursues a legitimate purpose by suitable, necessary and adequately balanced means (McCloy v New South Wales (2015); Brown v Tasmania (2017)).
The High Court has never squarely ruled on the validity of section 18C, but Federal Court authority has upheld Part IIA against implied freedom challenges (Bropho v Human Rights and Equal Opportunity Commission (2004); Toben v Jones (2003)), and most commentators consider it would survive, given its civil character, racial confinement and the breadth of section 18D (Chesterman 2000; Stone 2015). The divided judgments in Monis v The Queen (2013), concerning offensive postal communications, indicate that laws targeting offensiveness sit close to the constitutional line, but Part IIA does not prohibit political viewpoints as such. The scheme also partially implements article 4 of the International Convention on the Elimination of All Forms of Racial Discrimination. More importantly, the argument over section 18C is legislative rather than constitutional: Parliament may recalibrate the provision at any time, so the live question is not what a court must strike down but what a well-ordered Australian democracy ought to enact.
Dignity and liberty: the argument at the level of principle
The strongest defence of section 18C is dignitarian. Waldron (2012) argues that public hate speech undermines the assurance, owed to every member of society, that they hold equal standing and may go about their lives without being publicly branded as less than full members. Racial vilification attacks that assurance directly, and its costs fall disproportionately on minorities. Australian empirical work supports the account: Gelber and McNamara (2016) document withdrawal from public spaces, silencing and cumulative psychological injury among targets of hate speech, harms that counter-speech by victims is poorly placed to repair. The demographic context matters. More than half of Australians were born overseas or have a parent born overseas (Australian Bureau of Statistics 2022), and the Scanlon Foundation’s social cohesion surveys continue to report substantial rates of experienced discrimination among Australians from non-English speaking backgrounds (Markus 2023). In a polity of this composition, statutory assurance of equal standing is not symbolic excess but part of the infrastructure of social cohesion.
The liberty-based critique answers in the tradition of Mill (2015), for whom expression could be restricted only to prevent harm to others, never merely because it offends. On this account the statutory verbs ‘offend’ and ‘insult’ set the threshold below harm and invite the state to arbitrate civility; vague standards chill speakers who cannot predict liability; and the proper remedy for bad speech is more speech, not law. Forrester, Finlay and Zimmermann (2016) press these objections, contending that Part IIA is illiberal in principle and overbroad in text. The critique also has a democratic dimension: immigration, multiculturalism and Indigenous policy are contested political questions, and a provision touching speech in this territory risks narrowing debate that a self-governing people must be free to conduct. The position was captured in then Attorney-General George Brandis’s 2014 remark to the Senate that people have ‘a right to be bigots’.
Framed at this level, the dispute is narrower than it first appears. The liberty critique targets the statutory text, but the operative law, after Creek v Cairns Post Pty Ltd (2001) and Eatock v Bolt (2011), reaches only serious public denigration, conduct approximating the dignitary harm on which Waldron’s argument rests. A Millian can accept that assaults on equal standing are harms, not mere offence; Waldron’s point is that such speech damages interests as real as those defamation law protects. What survives of the liberty objection is a rule-of-law complaint rather than an absolutist one: a statute whose real reach depends on judicial gloss misleads citizens about their obligations and invites opportunistic complaints (Meagher 2004). That is a genuine defect, but a defect of drafting, not of principle.
The American contrast
The comparison most often invoked against section 18C is the United States. Under First Amendment doctrine, government may not regulate speech on the basis of its viewpoint; racist advocacy is constitutionally protected unless it amounts to incitement to imminent lawless action (Brandenburg v Ohio (1969)), a true threat, or another narrow categorical exception. The Supreme Court struck down a narrowly framed cross-burning ordinance as viewpoint discrimination (R.A.V. v City of St Paul (1992)) and protected the picketing of a soldier’s funeral with placards of startling cruelty (Snyder v Phelps (2011)). The doctrine rests on deep distrust of governmental power over opinion and confidence in the marketplace of ideas.
The contrast is instructive in both directions, but it is not an argument for repeal. Comparatively, the United States is the outlier: Canada, the United Kingdom, New Zealand and most European democracies maintain racial vilification laws, typically criminal and therefore more intrusive than Australia’s civil scheme (Gelber 2011). First Amendment absolutism is also embedded in an institutional context Australia lacks: an entrenched right, a rights-saturated political culture and a judiciary empowered to police the boundary. The American example is best read as a standing warning against state overreach, and as clarifying the choice Australia has made: a dignity-protective, civil, conciliation-first regime rather than criminal repression or constitutionalised laissez-faire.
Objection and reply: is the balance defensible?
The cumulative case for the current settlement is strong. The prohibition is confined to serious public racial denigration; section 18D shields good faith art, science, journalism and comment across the whole field of public affairs; the machinery is conciliation-first, inexpensive and non-punitive; the scheme aligns Australia with its international obligations; and it delivers, to communities that now constitute a majority of the population by migration background, a public assurance of equal standing whose value the empirical literature confirms (Australian Bureau of Statistics 2022; Gelber & McNamara 2016; Waldron 2012).
The most serious objection does not deny any of this; it says the process is the punishment. In Prior v Queensland University of Technology (2016), claims against students over Facebook posts about an Indigenous-designated computer laboratory were dismissed as lacking reasonable prospects, but only after a process that had hung over the respondents for years. A complaint against the cartoonist Bill Leak was withdrawn, yet not before substantial cost, publicity and distress. On this view Part IIA chills through uncertainty and procedure rather than through adverse judgments: respondents settle or self-censor to avoid the ordeal, whatever the law’s true threshold. The Parliamentary Joint Committee on Human Rights (2017) documented these concerns, and the Turnbull government’s proposal to replace ‘offend, insult, humiliate’ with ‘harass’ failed in the Senate. The objection is reinforced by the textual point already conceded: words that overstate the law’s reach invite the very marginal complaints that impose these costs.
The rebuttal is that every element of this objection identifies a procedural or drafting defect, none a foundational one. The procedural criticisms were substantially met by the Human Rights Legislation Amendment Act 2017 (Cth), which empowered the Commission’s President to terminate unmeritorious complaints and required the court’s leave before terminated complaints proceed; the QUT and Leak episodes would be unlikely to unfold the same way today. The drafting criticism supports codification, not repeal. The Australian Law Reform Commission (2016) concluded that the wording of section 18C merited review, and writing the Creek threshold into the statute, for example as ‘degrade, intimidate or seriously humiliate’, would align text with operation at no cost to its protective work. Replacing the standard with ‘harass’ would excise conduct at the core of the mischief, since sustained public denigration of a group can devastate the assurance of equal standing without harassing any identifiable individual. Repeal would go further still: not a restoration of neutrality but an expressive message of withdrawal, read by targeted communities as a statement about whose standing the state will protect (Gelber & McNamara 2016).
Judged in the round, the balance is defensible. Its principled core, a civil guarantee against serious public racial denigration, survives the strongest liberty objections once construction, exemptions and machinery are considered; what remains is best answered by refinement within the architecture, not its demolition.
Conclusion
Section 18C occupies a distinctive position in Australian public law: a statutory limit on expression in a polity that protects political communication only through a structural implication, and a recurring symbol in a wider argument about what multicultural citizenship requires. As construed and administered, the provision represents a defensible settlement. Its harm threshold is high, its section 18D exemptions are broad, its processes are civil and conciliation-based, and its justification, the protection of equal standing on which Waldron’s dignity account and the Australian evidence converge, answers rather than evades the Millian challenge. The American alternative illuminates the stakes but presupposes a constitutional order Australia has deliberately not adopted. What legitimately remains of the critique concerns words and procedure: language that overstates the law’s reach, and complaint burdens the 2017 amendments reduced but did not eliminate. The path indicated is incremental, aligning the text with its judicially settled meaning and tightening the procedural gate, while retaining a scheme that has for three decades asked Australians to conduct their fiercest public arguments without denying each other’s standing to take part.
References
Australian Bureau of Statistics 2022, Cultural diversity of Australia, ABS, Canberra.
Australian Human Rights Commission 2022, 2021-22 complaint statistics, AHRC, Sydney.
Australian Law Reform Commission 2016, Traditional rights and freedoms: encroachments by Commonwealth laws, ALRC report no. 129, ALRC, Sydney.
Chesterman, M 2000, Freedom of speech in Australian law: a delicate plant, Ashgate, Aldershot.
Forrester, J, Finlay, L & Zimmermann, A 2016, No offence intended: why 18C is wrong, Connor Court Publishing, Redland Bay.
Gelber, K 2011, Speech matters: getting free speech right, University of Queensland Press, St Lucia.
Gelber, K & McNamara, L 2016, ‘Evidencing the harms of hate speech’, Social Identities, vol. 22, no. 3, pp. 324-341.
Markus, A 2023, Mapping social cohesion, Scanlon Foundation Research Institute, Melbourne.
Meagher, D 2004, ‘So far so good? A critical evaluation of racial vilification laws in Australia’, Federal Law Review, vol. 32, no. 2, pp. 225-253.
Mill, JS 2015, On liberty, utilitarianism and other essays, Oxford University Press, Oxford.
Parliamentary Joint Committee on Human Rights 2017, Freedom of speech in Australia, Parliament of Australia, Canberra.
Stone, A 2015, ‘The ironic aftermath of Eatock v Bolt’, Melbourne University Law Review, vol. 38, no. 3, pp. 926-958.
Waldron, J 2012, The harm in hate speech, Harvard University Press, Cambridge, MA.
Legislation and international instruments
Australian Human Rights Commission Act 1986 (Cth).
Commonwealth Constitution, ss 7 and 24.
Human Rights Legislation Amendment Act 2017 (Cth).
International Convention on the Elimination of All Forms of Racial Discrimination 1965.
Racial Discrimination Act 1975 (Cth).
Racial Hatred Act 1995 (Cth).
Cases
Brandenburg v Ohio (1969) 395 US 444.
Bropho v Human Rights and Equal Opportunity Commission (2004) 135 FCR 105.
Brown v Tasmania (2017) 261 CLR 328.
Creek v Cairns Post Pty Ltd (2001) 112 FCR 352.
Eatock v Bolt (2011) 197 FCR 261.
Lange v Australian Broadcasting Corporation (1997) 189 CLR 520.
McCloy v New South Wales (2015) 257 CLR 178.
Monis v The Queen (2013) 249 CLR 92.
Prior v Queensland University of Technology [2016] FCCA 2853.
R.A.V. v City of St Paul (1992) 505 US 377.
Snyder v Phelps (2011) 562 US 443.
Toben v Jones (2003) 129 FCR 515.