Are RDA race hatred law amendments needed?

bolt-still-a-dickhead1Troppo author and frequent commenter John Walker asks:


The Bolt case was just one case-  is there much information about how 18C has been applied, on a wider scale.

Its pretty hard to judge whether there is a problem needing changes to the law , or not, on the basis of just one case. Do you have any info re other cases and judgements?

Fortunately, the answer is yes.  Section 18C of the Racial Discrimination Act 1975 was introduced by the former Keating Labor government in 1994.  In 2004 Deakin University legal academic Dan Meagher wrote a reasonably comprehensive review of its first decade of operation in an article titled “So Far So Good?: A Critical Evaluation of Racial Vilification Laws in Australia“.  I almost completely agree with his analysis and opinion, and so rather than reinvent the wheel I will simply reproduce edited extracts from his article with some relatively brief comments of my own at the end (although the whole of Meagher’s article is worth reading if you have time).

In essence, Meagher argues (and I agree) that racial vilification laws are needed, but that the current proscription of words which would be “offensive”, insulting” or humiliating” to reasonable members of a targeted race is much too vague, and consequently potentially has an unacceptably chilling effect on democratically essential free speech.  Similar conclusions have been expressed by a range of commentators well before the Bolt case.  This is not just a confected controversy beaten up by Abbotistas and Murdoch minions.

Whilst there is a need for effective racial vilification laws in Australia, the current laws lack sufficient precision and clarity in key respects. Of particular concern are the amendments made by the Racial Hatred Act 1995 (Cth) (‘RHA’) to the Racial Discrimination Act 1975 (Cth) (‘RDA’) and the free speech/public interest exemptions found in the RDA and the racial vilification laws of New South Wales, South Australia, Australian Capital Territory, Queensland, Victoria and Tasmania. An incoherent body of case law has developed as a consequence, where too much is left open to the decision-maker in each individual case. Many judgments are often little more than a series of findings of fact rather than reasoned pronouncements of the law. It has left the law in a state of unprincipled fluidity, where the good faith but ad-hoc assessment by individual judges and administrators of subjective, value-laden concepts determines controversies not the application of reasonably precise and knowable legal standards. This is problematic for a number of reasons. They will be detailed shortly.

But first it should be noted that indeterminacy in the law is not unique nor is precision and clarity always a virtue. Timothy Endicott has persuasively argued that vagueness in the law is on occasion unavoidable and sometimes desirable. However, what remains centrally important is that ‘the law must be capable of guiding the behaviour of its subjects’. In the area of racial vilification however there are compelling reasons why enhancing the precision and clarity of legislation is desirable.

Firstly, and most importantly, is the capacity of indeterminate racial vilification laws to unreasonably interfere with or pre-emptively chill the legitimate speech and communication interests of others. Whilst broad-ranging defences (the norm in Australian law) may allay some of these speech and communication concerns, this species of indeterminacy in turn has the capacity to erode the efficacy of such laws by failing to provide a remedy or meaningful protection to victims of racial vilification. Consequently, the primary goal of racial vilification laws in Australia — to regulate racial vilification without curbing legitimate public communication — is compromised when the laws themselves lack sufficient precision and clarity. Improving their precision and clarity would make these laws more accessible and, in this instance, strengthen the rule of law. With a firmer understanding of their legal rights and obligations the citizenry can plan their communicative conduct accordingly. This has an added importance with citizens now increasingly willing and able to seek legal redress for racial vilification.

It has resulted in a body of judicial and quasi-judicial decisions that often lack a coherent, underpinning principle.

Secondly, laws which ‘leave too much to be decided by persons other than the people’s representatives’ can be rightly criticised as undemocratic. This is not to suggest that reserving a measure of discretion for decision-makers in this area is objectionable. Indeed it is both inevitable and desirable with racial vilification laws as explained below. It is problematic however when the putative legal standards contained in a law provide little interpretive guidance in most cases to the relevant decision-maker. It is undemocratic because judges and administrators are, in effect, exercising legislative power by determining the substantive content of the laws they are to apply. This argument suggests that courts should limit ‘themselves to the accurate application of general rules, rules which should be clear, precise and empirically applicable expressions of the political will of the people’s representatives.’ It ‘is democratic in that it affirms that the source of these authoritative rules is empirically identifiable institutional acts which are the outcome of democratic procedures’ not the subjective conceptions of justice of judges and administrators articulated on a case by case basis. My analysis will show that too often the application of racial vilification laws in Australia has exhibited this undemocratic quality. Consistent with democracy and the principle of popular sovereignty that underpins the Australian Constitution, legislative power ought to be exercised by our elected not unelected representatives. Moreover, Geoffrey de Q Walker has noted that when ‘law is simply a series of patternless exercises of state power … the outcome of any encounter with government can no longer be predicted and equality before the law is also lost.’

Thirdly, laws which lack sufficient precision and clarity obfuscate and complicate the role of those public officials charged with their interpretation and execution. For example, this may manifest as an unwillingness on behalf of prosecutors to mobilise indeterminate criminal racial vilification laws where the higher standard of proof required compounds the problem of legislative imprecision. In the long-term, citizens (including public officials) may accord less respect to such laws which can undermine their efficacy, enforceability and ultimately their legitimacy.

But the concept of racial vilification is hard to pin down. Not least because reasonable minds will differ as to what level of racist conduct ought to constitute vilification for legal purposes and how one can determine with some predictability when that harm threshold is reached. It is a concept with a subjective component meaning that some degree of indeterminacy will necessarily characterise racial vilification laws. Indeed, it is no bad thing that decision-makers in this area have a level of discretion, so long as sufficient criteria exist to guide the exercise of that discretion. This guards against arbitrariness. It is a complex, emotive and delicate area where free speech and other legitimate concerns may be trammelled if the laws are enforced in a mechanistic or unthinking manner. Moreover, laws are more likely to be respected and therefore effective when applied, so far as possible, in a just as well as principled manner. …

The critical problem with s 18C is that its key words and phrases are sufficiently imprecise in both their definition and application as to make the putative legal standards they embody largely devoid of any core and ascertainable content.

The critical problem with s 18C is that its key words and phrases are sufficiently imprecise in both their definition and application as to make the putative legal standards they embody largely devoid of any core and ascertainable content. Of most concern are the phrases ‘offend, insult, humiliate or intimidate’ and ‘the act is done because of the race … of the other person’. The former phrase, in particular the meaning of the words ‘offend’ and ‘insult’, is so open-ended as to make any practical assessment by judges and administrators as to when conduct crosses this harm threshold little more than an intuitive and necessarily subjective value judgement. The fact that an act must be ‘reasonably likely’ to cross this harm threshold, though importing an objective test of liability, does not cure the definitional indeterminacy of these words that a decision-maker must objectively apply. Moreover, these words and the harm threshold they establish may capture a range of conduct which was arguably never intended by the Parliament to be regulated; an important point to which I shall shortly return. …

Indeed, with the possible exception of cases involving extreme racist conduct, the indeterminacy of s 18C is such that too many determinations could comfortably and justifiably have been decided the other way. This should come as no surprise as the legal standards in s 18C are sufficiently malleable to allow a judge or administrator to employ them to facilitate a decision which accords with their intuitive conception of what ‘justice’ requires in that case. It has resulted in a body of judicial and quasi-judicial decisions that often lack a coherent, underpinning principle.

The indeterminacy of the harm threshold has become manifest in the case law in two ways. Firstly, in a series of s 18C determinations the judge or administrator has, ‘effectively elevated the threshold by emphasising the concept of “hatred”. Secondly, in a smaller but still significant number of cases there has been a finding that s 18C has been offended without any harm threshold analysis or reasoning whatsoever. …

But it is clear enough that one can racially insult or offend another without ever expressing or intending hatred for that person’s race or ethnicity. Consider a claim by a politician that ‘[h]ome invasions are ethnically based, Lebanese or Iranian, not Australian.’ Or when Australian cricketer Darren Lehmann called a Sri Lankan opponent ‘a black cunt’ upon dismissal. These racial epithets no doubt offended and insulted the relevant victims and, moreover, may well be reasonably likely to elicit the same response from most members of the relevant race or ethnic group if not the wider community. It is submitted, however, that in both cases the conduct of itself did not amount to an expression of racial hatred. These words do not suggest an intense dislike or detestation of that person(s) on account of their race or ethnicity. Arguably, this sort of low-end racist conduct does not constitute the kind of extreme racist behaviour that Parliament intended the RHA to regulate.

Moreover, judicial and administrative attempts to define words like ‘insult’ and ‘offend’ with a degree of precision become a circular and question-begging exercise. The best that can usually be done is to outline the Macquarie and/or Oxford English Dictionary definitions of the terms as Hely J did in Jones v Scully. But these dictionaries define the words using synonyms, which is of little use when the task of the decision-maker is to elucidate and then apply a reasonably precise legal standard. It simply raises the same definitional question for the synonyms used, and so on. For example, the difficulty in ascribing a clear meaning to the word ‘insult’ was illustrated by the 1972 House of Lords decision in Brutus v Cozens. The case concerned the interpretation of s 5 of the Public Order Act 1936 (UK). Lord Reid wrote:

We were referred to a number of dictionary meanings of ‘insult’ such as treating with insolence or contempt or indignity or derision or dishonour or offensive disrespect. Many things otherwise unobjectionable may be said or done in an insulting way. There can be no definition. But an ordinary sensible man knows an insult when he sees or hears it… Insulting means insulting and nothing else.


The amendments proposed by the Abbott government address the defects identified and discussed by Dan Meagher, in that the indeterminate words “offend”, “insult” and “humiliate” are removed and replaced by the word “vilify”, which is defined as meaning “to incite hatred against a person or a group of persons”.

This strikes me as an appropriate amendment which will improve the certainty and effective operation of the Act and arguably more accurately reflect the enactment intention of the Keating government which, as Meagher argues, was only ever to prohibit words which did indeed incite race hatred rather than to sanction lower level offensive or insulting language.

[T]he problem does not lie in the scope of proscribed language but rather in the breathtaking scope of the exemptions currently proposed.

However, that is not to suggest that the Brandis amendments are anything like acceptable in their present form.  Nevertheless, the problem does not lie in the scope of proscribed language but rather in the breathtaking scope of the exemptions currently proposed. The proposed new subsection 4 reads as follows:

(4) This section does not apply to words, sounds, images or writing spoken, broadcast, published or otherwise communicated in the course of participating in the public discussion of any political, social, cultural, religious, artistic, academic or scientific matter.

As you can see, this provision would effectively exempt just about any public expression of race hatred in just about any field of discourse. In other words, it would make the hate speech provisions of the Racial Discrimination Act effectively meaningless.

What should occur is that the Parliament should preserve the existing exemption provision found in section 18D. It exempts words in a much narrower range of public discourse, and only where they meet the test of being “done reasonably and in good faith”.  However, those words “reasonably and in good faith” are themselves quite indeterminate and in need of clarification, as Meagher also discusses.  Nevertheless, if the Coalition ends up with an exemption provision more or less like the current section 18D (albeit with criteria specified for assessing reasonableness and good faith), then in my view these amendments would not only be appropriate but a significant improvement on the present regime.

About Ken Parish

Ken Parish is a legal academic at Charles Darwin University, with research areas in public law (constitutional and administrative law) and teaching & learning theory and practice. He has been a legal academic for almost 12 years. Before that he ran a legal practice in Darwin for 15 years and was a Member of the NT Legislative Assembly for almost 4 years in he early 1990s.
This entry was posted in Law, Uncategorized and tagged . Bookmark the permalink.

38 Responses to Are RDA race hatred law amendments needed?

  1. Good post, Ken. You don’t post often enough here.

    While the arguments about the vagueness of the present section make sense, I would be much more persuaded of a need to amend it if there were examples of its practical application where I could feel some sympathy for the target of the complaint. Andrew Bolt certainly doesn’t count – as far as I am concerned, he has taken unworthy faux martyrdom to new heights. He should have apologised for the errors and put up a corrective to the articles and been done with. (Or, even better, the complainants sued for defamation.) I have always assumed that News must have encouraged him to fight the complaint for its mutual benefit, but I don’t know that I have ever read any commentary about the role they played in protracting a legal action which (I feel sure, once they knew of the errors) any “normal” lawyer would have told a “normal” client – “mate, you stuffed up a bit here, just apologise at the mediation conference and publish some corrections. Why spend $200,000 and the angst of being in court?”

    I think that the brilliant analogy by Michelle Grattan about it being like an asbestos shed summed it up very well – sure, you wouldn’t word it the same way if starting from scratch, but now that its there and doesn’t seem to be causing much problem, renovations on it now may be more trouble than its worth.

    • Tel says:

      If you are unwilling to defend a scoundrel, then you don’t really believe in free speech.

      The common Australian approach of broad legislation, followed by selective enforcement is a recipe for third world despotism.

      • desipis says:

        Or maybe you just believe in a more civilized and less racial definition of “free speech”.

        Also, the law in question is a law that gives rise to private action, not public prosecution, so selective enforcement isn’t really an issue in this situation.

    • desipis says:

      Why spend $200,000 and the angst of being in court?

      Seems like it was an effective marketing strategy. He’s put himself forward as the martyr of free speech.

  2. Jim Rose says:

    British Columbia has an extremely broad hate speech law that prohibits the publication of any statement that “indicates” discrimination or is “likely” to expose a person or group or class of persons to hatred or contempt.

    Professor Sunera Thobani of the University of British Columbia faced a hate crimes investigation after her vicious diatribe against American foreign policy. She said that Americans are “bloodthirsty, vengeful and calling for blood.”

    The Canadian hate-crimes law was created to protect minority groups from hate speech. But in this case, it was invoked to protect Americans.

  3. David Walker says:

    A side note: Any graphic calling for someone to be “sodomised with a calculator” is a fair way beyond the pale. (Test: would you even think about putting this up if the subject was a woman?) Can we ditch the graphic please?

    • Ken Parish says:

      It’s no doubt a bit gross but clearly satire and clearly not sexist. The ABC Chaser mob are currently being sued for defamation for showing an image of Murdoch attack journo Chris Kenny having sex with a dog. Maybe that was a bit beyond the pale, although again it was clearly satire. This image doesn’t even arguably impute aberrant sexual conduct to Bolt, so it clearly isn’t defamatory. Finally, it’s lifted from Independent Australia, which isn’t my favourite publication but does suggest it’s an image within the domain of mainstream political discourse.

      • David Walker says:

        Ken, I don’t think it’s sexist. I think it’s nasty. Again the test: if the subject was a woman, would “it’s clearly satire” excuse us?

    • Ken Parish says:

      In fact, with a bit more Googling I just discovered that the reference is to an article on Crikey by Scott “Possum” Steel titled Why Andrew Bolt should be Sodomised with a Calculator – Part 142.

      • Ken Parish says:

        Mind you, at the risk of arguing with myself I should note that Possum’s reference to sodomising Andrew Bolt with a calculator was an attempted scathing criticism of Bolt’s alleged innumeracy for suggesting that Australia’s increased asylum seeker arrival rate of 19% in early 2009, versus a global rate of increase of 12%, proved that the Rudd Labor government’s relaxation of the Howard government’s policies (rather than the global increase) was a major cause of the greater number of boat people arriving.

        Unfortunately for Steel, the arrival numbers kept increasing radically after his “sodomising” remark, to the point where boat-borne asylum seekers were arriving at the rate of 30,000 per year just before Gillard and then Rudd re-embraced Howard-esque punitive policies. That would appear to be an increase of around 6000% above the 2008 figures. No-one argues these days that Rudd’s 2008 policy relaxation was not a significant causative factor. Perhaps it is Possum who should be sodomised with a calculator?

        • Patrick says:

          No, just whacked with the reality stick. As should anyone who seriously believed in kinder gentler boat people policy.

        • paul walter says:

          If he is to be sodomised with a calculator, may it be that it is mainframe.

      • I’ve just re-read Possum’s essay (thanks for the link, Ken), and I still think itss fair comment. I have just enough mathematics to understand most of what he said, and it makes a lot of sense.

        • Patrick says:

          Sure, just unfortunately completely wrong on the broader point.

          Bolt’s argument is as predictable as it is convenient – Rudd is being soft on border protection and as a result, increasing hordes of refugees are arriving on our fair shores.

          Convenient, and with hindsight merely stating the obvious. Whereas with hindsight Possum resembles, at best, climate change deniers pointing to the last ten years of data as a reason to throw the whole crock out.

        • I take it you’ve done a similar statistical analysis on more recent data, Patrick, which is why you’re so confident that Bolt is right, and Possum (an actual statistician) is wrong.

          I’d like to see your workings.

    • My recollection is that Possum (Scott Steele) used this in an irritated response to Bolt’s wilful and continual innumeracy and ignorance. At the time, and in its context, it struck me as being entirely reasonable.

  4. David Walker says:

    The words in the past are great, though: indeed, they get to the central flaw of 18C better than anything I’ve seen:

    “… sufficiently imprecise in both their definition and application as to make the putative legal standards they embody largely devoid of any core and ascertainable content”

  5. David Walker says:

    Oops. The words in the post are great.

  6. Ha! I see that in a post today about Antony Loewenstein’s support for amending the Act, Bolt grumbles that:

    … Loewenstein never mentions how it was once used against him – and not, as in my case, because he was fighting racial stereotypes, but because he was airing the very cruelest:

    Sorry, I know the post is about how the issue is not just limited to what happened to Bolt (although, let’s face it, the Coalition wasn’t showing any concern until it affected Andrew), but I couldn’t resist noting Andrew’s noble sentiment.

  7. desipis says:

    A good post on the topic. The barn-door of a loophole in subsection (4) is by far the most problematic aspect on the proposed changes.

    However, what remains centrally important is that ‘the law must be capable of guiding the behaviour of its subjects’.

    This point is pretty much why I disagree with Waleed Aly’s criticism on the proposed subsection (3). If, as according to Waleed Aly, ordinary reasonable members of the Australian community are insufficiently qualified to determine what is problematic (whether the standard is “vilification” or “offense”), then how fair is it to punish them for breaching that standard?

    I do feel that “hatred” is perhaps narrowing the law a bit too far. I agree offense is too broad, but I think inciting “contempt” or even inciting “discrimination” would be a better standard to measure against. In a sense, rather than considering emotional harm as an “offense” standard does, look to defamation and it’s concepts of damaging reputation, albeit through using or encouraging racial prejudice instead of making false statements.

  8. John walker says:

    Ken thank you .

  9. I used to be not trampis says:

    I agree and have given this article some publicity. I hope it helps.

  10. Paul Frijters says:

    Thanks for the post, Ken.

  11. Gummo Trotsky says:

    Good post Ken. And a welcome relief to see it. I’ve found this debate over s18C of the RDA a frustrating distraction from real life for a bit too long – good to see something sensible written about it from the legal viewpoint.

    I find the repeal of s18E a bit puzzling. It makes a nice gift to Rupert Murdoch and Newscorp – they no longer have to answer for the actions of their employees and agents – but I’d say it’s as ill-considered as my post on investor state dispute resolution ;)

  12. Tim Macknay says:

    Thanks for this post Ken.

    There’s one thing about the proposed amendments that’s been bugging me which I’d be interested to hear your views on: As the proposed subsection (4) would essentially authorise racial vilification if it is “spoken, broadcast or published” etc, does this have implications for the various State-based anti-vilification laws?

    I haven’t tried to analyse it in any depth, but on the face of it, a Federal law that authorises racial vilification would seem to be inconsistent with a State law that prohibits it. This seems to raise the possibility that the proposed amendments could neutralise State laws prohibiting racial vilification through the operation of section 109 of the Constitution. I’ve not seen this discussed elsewhere and I’d be interested to hear your thoughts.

    • Ken Parish says:

      Thanks Tim.Yes that is a very interesting question. As you know, s 6A of the RDA expressly provides for state anti-discrimination laws to operate concurrently with the Commonwealth RDA,and s 18F provides similarly in relation to the hate speech provisions specifically. However those provisions can only negate the finding of an implied intention that the Commonwealth Act was intended to cover the field for Constitution s 109 purposes. They can’t cure a direct inconsistency, which is what the proposed subsection 4 exemption/s may well create (for precisely the reason you state).

      However, an equally interesting argument that might be run against that conclusion is that the new subsection 4 would not be a valid Commonwealth law, because it could not be regarded as a bona fide attempt to implement an international treaty (cf Horta v Commonwealth). The only constitutional head of power supporting the RDA is the s51(xxix) external affairs power, implementing the International Convention on the elimination of all Forms of Racial Discrimination. I haven’t thought this through, but on first blush it seems to me that you could make a plausible argument that an exemption provision which renders speech inciting racial hatred lawful in almost all public situations is not a bona fide attempt to implement the treaty.

  13. john Walker says:

    “the comment is an expression of a genuine belief held by the person making the comment.” surely exempts just about anything?

  14. Tim Macknay says:

    That is a plausible counter-argument Ken, and certainly does nothing to dispel my impression that the exposure draft was not well thought through.

  15. Tim Macknay says:

    John Walker, the thing to remember about the “genuine belief” clause is that it’s just one part of a larger and more onerous provision. The comment also has to be said “reasonably and in good faith” and must be “fair”.

    • john Walker says:

      Tim thanks.
      Is the provision a sequence of tests , each one of which must be passed, before going on to the next one?

      • Ken Parish says:

        John, here is a link to s 18D. As you will see, it defines a number of areas of exemption, but all of them are conditioned by the requirement that the communication must have been “said or done reasonably and in good faith”. Moreover, as Tim noted, genuine belief is a condition of the “fair comment” exemption, which requires that the comment to be fair in all the circumstances. Genuine belief is just one of the determinants of fairness of a comment.

        The existing exemptions in s 18D are much tighter than what Brandis is proposing, which is effectively that speech inciting racial hatred would be exempt from action under the RDA in most areas of public discourse and with no requirement that the speech must be reasonable, in good faith or the subject of genuine belief on the part of the speaker. It would effectively make the purported prohibition of race hatred speech a meaningless charade.

        • john Walker says:

          ” It would effectively make the purported prohibition of race hatred speech a meaningless charade.”
          obviously agree.
          The wording comes from a ‘exposure Draft”, how final is that sort of document?

        • Tim Macknay says:

          John, exposure drafts are not at all final. They’re generally released for the purpose of identifying issues with the draft that can be rectified before the Bill is introduced to Parliament. Given some of the issues with this draft (such as the Constitutional validity issue identified by Ken above), and the fact that the general response has been much more ambivalent than the government expected, I’d be surprised if the draft got to Parliament without substantial changes.

        • john Walker says:

          Tim glad to hear that , on reading it , subsection 4 feels like a very first draft.

      • Gummo Trotsky says:

        The other point on ther ‘genuine belief’ clause is that some of Bromberg J’s comments on Bolt – in the long catalogue of findings of fact – had a somewhat sceptical tone when it came to the question of whether Bolt was expressing a genuine belief.

  16. Mel says:

    Great article, Ken. I agree that the bar in 18C should be much higher than “offence” and “insult”.

    In the Eatock v Bolt case the judge said:

    I have not been satisfied that the offensive conduct that I have found occurred,is exempted from unlawfulness by section 18D. The reasons for that conclusion have to do with the manner in which the articles were written, including that they contained errors of fact, distortions of the truth and inflammatory and provocative language.

    As someone with a sociology degree and who still vaguely follows the antics of far-left sociological scholarship, it is plain to me that much of this scholarship on racial issues, and in particular the more extreme variants of white privilege theory, fall foul of 18C and the tone, factual errors and hyperventilating language rule out 18D as a defence if the Eatock precedent is applied.

    I would experience tremendous schadenfreude if a conservative activist used the RDA as it currently stands to drag some of nuttier left wing scholars before the courts for white-baiting ;)

Comments are closed.