University of New South Wales Law Journal
|
|
Home
| Databases
| WorldLII
| Search
| Feedback
University of New South Wales Law Journal |
|
DAN MEAGHER[*]
The main aim of this article is to develop an argument as to how the test for constitutionality outlined in Lange v Australian Broadcasting Corporation (‘Lange’) ought to be applied when assessing the compatibility of a law with the implied freedom of political communication.[1] It is an important issue, for in the post-Lange case law the test has been applied in two different ways and my analysis will demonstrate that the validity of a law may well depend on which approach is taken.
In Part II it is argued that the proper application of the Lange test requires that one approach (the two-tier approach) be abandoned. On this approach, if a law regulates the content of a political communication (as opposed to the mode of its delivery), more rigorous judicial scrutiny will follow. Instead, I argue for a single test for constitutionality where its application is through the proportionality framework and informed by the rationale of the implied freedom. Under this proposed review model, what counts is the isolation and evaluation of the communication interest at stake. So, for example, if the effect or practical operation of a law is to significantly restrict political communication, then more rigorous judicial scrutiny is appropriate. To then pass constitutional muster, the law must have a ‘compelling justification’ and provide benefits that outweigh its detrimental impact on political communication.
In Part III this conception of the Lange test and the proposed review model are applied to Australian racial vilification laws in order to assess their compatibility with the implied freedom.[2] This analysis is important in its own right for the issue has not been considered by the High Court nor been subject to detailed treatment in the lower courts or in academic literature.[3] It also provides an instructive case study to outline, explain and justify the principles that underpin my argument for this conception of the Lange test. In this regard, it should be of interest to constitutional lawyers more generally. It is concluded that under the proposed review model the current Commonwealth, State and Territory racial vilification laws are compatible with the implied freedom. First, though the laws may in some instances restrict the freedom to communicate on government or political matters, that burden is not significant. Second, and in any event, they are effective, appropriate or rational measures to secure a compelling constitutional end, provide benefits that significantly outweigh any detriment to ‘political communication’ and represent reasonable legislative efforts to do so in a manner that minimises the infringement of the implied freedom.
In Lange the High Court unanimously endorsed a test for assessing the compatibility of a law with the implied freedom.
First, does the law effectively burden freedom of communication about government or political matters either in its terms, operation or effect? Second, if the law effectively burdens that freedom, is the law reasonably appropriate and adapted to serve a legitimate end the fulfilment of which is compatible with the maintenance of the constitutionally prescribed system of representative and responsible government and the procedure prescribed by s 128 for submitting a proposed amendment of the Constitution to the informed decision of the people. If the first question is answered ‘yes’ and the second is answered ‘no’, the law is invalid.[4]
However, notwithstanding this unanimity, the cases that immediately followed saw the emergence of two distinct applications of the Lange test. One approach employs a single test for constitutionality regardless of whether the law regulates the content or mode of a political communication and in a manner deferential to the Parliament.[5] This involves extending a ‘margin of appreciation’ to the Parliament, recognising the Court’s limited institutional capacity to evaluate the substantive effect a law may have on the implied freedom.[6] The other is a two-tier approach that varies the level of judicial scrutiny depending on the nature of the law. This entails more rigorous scrutiny of laws that regulate the content rather than the mode of a ‘political communication’.[7] The reason is that content-restrictive laws are thought to necessarily impose a more significant burden on the implied freedom. The upshot is that there must be a ‘compelling justification’[8] or ‘overriding public purpose’[9] for the State to regulate content but a law that only targets the mode of a communication need not meet such an exacting standard.[10] The two-tier approach recently found favour with Heydon J in Coleman v Power[11] (‘Coleman’) and Gleeson CJ in Mulholland v Australian Electoral Commission[12] (‘Mulholland’). Some judges have in addition, or maybe as an adjunct,[13] to the two-tier approach considered ‘whether less drastic measures are available’.[14] In this way the two-tier approach seems to embody (at least to some extent) the American constitutional law principles of ‘strict scrutiny’ and ‘overbreadth’.[15] The former originates from the famous footnote four in United States v Carolene Products.[16] The latter is a component of stricter scrutiny of content laws in American First Amendment law. Such laws will only be upheld if ‘necessary to serve a compelling state interest and … narrowly drawn to achieve that end’.[17]
In any event, the argument that assessing a law’s restrictive impact on (political) communication ought not to proceed by drawing a rigid distinction between laws that regulate content rather than mode is not original. In the First Amendment context, John Hart Ely provides the following example to illustrate the point:
[T]he regulation of certain forms of communication for reasons other than their content may discriminate de facto (or even intentionally, though in a way that may not be provable) against certain clusters of messages. Sound-trucks, for example, are more frequently resorted to by those whose access to more expensive and less annoying media is limited. That surely is something that belongs in the calculation: a more serious threat should be required when there is doubt that the speaker has other effective means of reaching the same audience.[18]
Similarly, in Canada, a contextual rather than rigid categorical approach is favoured in the application of their constitutional guarantee of freedom of expression. What is critical when considering the restrictive impact of a law is ‘the importance of the freedom of expression interest at stake on the facts of the case’.[19] In Australia, Jeremy Kirk has also forcefully made the argument that a two-tier approach should not be used in the application of the implied freedom:
The utility or accuracy of the suggested two-tier approach is limited. If a law on the means or mode (rather than the content) of communication restricted the freedom significantly then such a law would and should be required to have a weighty justification. A law rationing the number of television or radio licences is clearly the sort of law envisaged as falling into the more easily justified category. Yet it is doubtful that a law restricting the number of newspapers would be viewed with the same equanimity. The difference between these cases is not the nature of the law but the presence of a sufficient justifying purpose. The restrictive efect on the freedom and the weight of the justifying ends are what is important, not the form of the law.[20]
Kirk’s argument provides a strong normative justification for rejecting an approach where the level of judicial scrutiny undertaken is determined by a preliminary assessment as to whether law regulates the content or mode of a political communication. It may well be the case that a greater political communication interest is usually at stake with laws that target the content rather than the mode of a communication. But, as Ely, the Canadian Supreme Court and Kirk have shown, it can become an inflexible and sometimes misleading threshold question with the consequence that the ‘freedom of expression interest at stake’[21] might be missed, undervalued or, in the worst case, even avoided. If so, it then becomes impossible to properly apply the Lange test.[22] For this requires a court to answer whether, notwithstanding its burden on the implied freedom, the law is reasonably and appropriately adapted to secure an end consistent with our system of constitutional government.
It is my argument that the two-tier approach ought, therefore, be abandoned. A single test should be employed with the isolation and evaluation of the ‘freedom of expression at stake’ the critical first step required in the review analysis. This evaluation might include, amongst other things, whether the restriction curtails the ability of a person or group from ‘participation in the political process’[23] (Ely’s sound-truck example), the importance of the communication in the particular political climate (an address, for example, regarding immigration policy in the wake of the so-called ‘war on terror’) and the amount and source of the expression restricted (the political advertising law invalidated in Australian Capital Television Pty Ltd v Commonwealth[24] (‘Australian Capital Television’)). Once this is done, it enables a judge to more confidently assess whether a law has ‘a sufficient justifying purpose’[25] to constitutionally warrant the (political) communication infringement. This is not to deny that the level of judicial scrutiny will depend on the nature and operation of a law. But the relevant level of scrutiny employed ought not be determined by a preliminary assessment as to whether the law regulates the content (more) or mode (less) of a political communication. For, as noted, that kind of threshold judgment may result in the actual freedom of expression interest at stake being missed, undervalued or avoided. In the context of my argument, more rigorous judicial scrutiny is appropriate when the efect or practical operation of the impugned law is to significantly infringe communication on political or government matters. It is then incumbent on the State to convince the court that the law has a ‘compelling justification’ and that it provides benefits that outweigh its restrictive effect on ‘political communication’.[26] It is worth noting here that Kirby J, an earlier proponent of the two-tier approach, appeared to abandon it recently in Mulholland.[27] Instead, he advocated a more general approach to judicial review suggesting that ‘in certain circumstances, courts have a heightened vigilance towards the potential abuse of the lawmaking power inimical to the rule of law’:
Such vigilance may be specially needed when the power is directed against unpopular minorities. In those cases, or in circumstances where current lawmakers pursue their own partisan advantage, courts may subject the legislative vehicles of such advantage to close attention.[28]
In this approach, there are close parallels with the theory of judicial review propounded by Ely. He said the courts should act as a referee for representative democracy, intervening only when the ‘ins are choking off the channels of political change to ensure they will stay in and the outs will stay out’ or ‘an effective majority are systematically disadvantaging some minority out of simple hostility or a prejudiced refusal to recognise commonalities of interest’.[29] For Kirby J, the Lange test now embodies a single constitutional principle whose application is to be informed by the theory of judicial review just outlined, not the preliminary assessment at the heart of the two-tier approach.
It should first be noted that the language of the Lange test is somewhat misleading regarding the level of judicial scrutiny it entails.
Despite the High Court’s use of the formulation ‘reasonably appropriate and adapted to’, it is clear that in the context of the freedom of communication, the Court does not use it to mean the minimal kind of review seen in other contexts … [T]he High Court has been quite explicit that, in this context, the formulation is synonymous with proportionality.[30]
This point was explicitly acknowledged by Kirby J in both Coleman[31] and Mulholland,[32] though he personally favours the use of proportionality for representing a more accurate and useful description of and explanation for the characterisation process involved in the application of the implied freedom. And in Mulholland, Gleeson CJ said he had no objection to the use of either the reasonably appropriate and adapted test or one of proportionality.[33] It was, however, ‘important to remember, and allow for the fact, that [proportionality] has been developed and applied in a significantly different constitutional context.[34] Further, the slight re-wording of the second limb of the Lange test in Coleman by McHugh J to make ‘clear that the Court did intend [it] to be read in a way that requires that both the end and the manner of its achievement be compatible with the system of representative and responsible government’[35] and the Court’s rejection in the same case of the submission made by the Attorneys-General of the Commonwealth and New South Wales that the test ‘should be weakened by requiring only that the law in question be “reasonably capable of being seen as appropriate and adapted”’[36] acknowledged that a more rigorous, proportionality-style review was appropriate and in fact undertaken by the Court.[37]
In any event, proportionality is generally understood to have three levels of inquiry involving graduating degrees of judicial scrutiny.[38] First, whether a law is ‘suitable’ in the sense of being ‘an effective, appropriate or rational means of achieving the claimed end’.[39] Second, whether a law is ‘necessary’, ‘in the sense that there are no alternative means available to achieve the same end which are less restrictive of the protected interest’.[40] Third, whether the legitimate end of a law outweighs the restriction it imposes.[41] For a law to be proportional it must satisfy each of these levels of inquiry. It is clear that the High Court’s implied freedom jurisprudence has involved the different proportionality levels both before and after Lange.[42] In particular, the overbreadth analysis and balancing process involved in second (necessity) and third level (balancing) proportionality respectively have been present in the reasoning of those judges that have employed the two-tier approach. For example, in Australian Capital Television, Mason CJ said that a ‘compelling justification’ must exist for a law that targets the content of a ‘political communication’ and it ‘must be no more than is reasonably necessary to achieve the protection of the competing public interest’.[43] Similarly, in Levy v Victoria[44] (‘Levy’), Gaudron J said that ‘[i]f the direct purpose of the law is to restrict political communication, it is valid only if necessary for the attainment of some overriding public purpose’.[45] In addition, first level proportionality (suitability) is contemplated by that part of the Lange test that asks whether a law that burdens the implied freedom ‘serve[s] a legitimate end the fulfilment of which is compatible with the maintenance of the constitutionally prescribed system of representative and responsible government’.[46]
However, whilst the test of proportionality ‘provides an efficient framework for judging restrictions and specifying objections’,[47] the ‘test itself does not give any guidance as to, and consequently does not place any restriction on, how judges assign weight to the competing interests’.[48] It means that ‘[f]or each level of proportionality it is possible to assess the requirements rigorously or deferentially’.[49] This choice, which a court inevitably must make when it undertakes this review analysis, requires reference to a theory or view as to the purpose of the implied freedom. This is the crux of the criticism made by Adrienne Stone regarding the High Court’s anti-theoretical approach to the implied freedom. That is, relying exclusively on the text and structure of the Constitution in order to ascertain the scope of the implied freedom cannot sufficiently guide a judge as to how the Lange test ought to be applied in any given scenario.[50] In Coleman, McHugh J acknowledged the force of Stone’s criticism but said it was ill-founded:
In determining whether a law is invalid because it is inconsistent with freedom of political communication, it is not a question of giving special weight in particular circumstances to that freedom … Freedom of communication always trumps federal, State and Territorial powers when they conflict with the freedom. The question is not one of weight or balance but whether the federal, State or Territorial power is so framed that it impairs or tends to impair the effective operation of the constitutional system of representative and responsible government by impermissibly burdening communications on political or governmental matters.[51]
But in my view, this riposte simply restates the relevant constitutional principle. It says nothing more as to how the court ought to measure or assess that impairment. On the other hand and more promisingly, there were passages in the judgments of Gleeson CJ in Coleman and Mulholland and Kirby J in the latter which may suggest the nascent stages of a theory of the implied freedom for each judge. In Kirby J’s case, a theory of judicial review more generally was suggested, as earlier noted. In Coleman, for example, Gleeson CJ proposed a standard of judicial review that he said was ‘consistent with the respective roles of the legislature and the judiciary in a representative democracy’.[52] His Honour then expanded on this theme in Mulholland by outlining in some detail the constitutional context in which the implied freedom arises and is to be applied. In the main, he emphasised the central role of Parliament in our constitutional system and whilst ‘[c]oncepts such as representative democracy and responsible government no doubt have an irreducible minimum content’,[53] he noted ‘how little of the detail of that system is to be found in the Constitution, and how much is left to be filled in by Parliament’.[54] This, for example, ‘gives Parliament a wide range of choice’ in the form and regulation of the electoral process.[55] For Glee son CJ, this constitutional context explains why judicial review of legislative action against the implied freedom ‘does not involve the substitution of the opinions of judges for those of legislators upon contestable issues of policy’.[56] In the absence of some kind of theory, the application of the implied freedom (and the second limb of the Lange test, in particular) becomes ‘a ritual incantation, devoid of clear meaning’.[57] Indeed, it may be true that even if a theoretical position is not explicitly articulated or consciously avoided, it cannot be forestalled.[58] For how a judge in fact applies the test for constitutionality must say something as to their view regarding the reason for or purpose of the implied freedom.[59] In this way, employing the two-tier approach might evidence a commitment to constitutionally protect a wide range of political communication irrespective of its content in order to further the important speech/expression values of truth[60] in political discourse and more informed democracy and selfgovernment.[61] For example, the two-tier approach employed by Mason CJ in Australian Capital Television was grounded in the values of self-government and a modified argument from truth.[62] The latter evinced an express distrust of the government regulating political communication, a close approximation of the argument made by Frederick Schauer that doubts the ability of government to decide what speech ‘is true and what is false’.[63]
In the next part of the article I will sketch an argument that I have developed more fully elsewhere as to what I consider to be the primary rationale of the implied freedom.[64] This theoretical position, in turn, suggests how the Lange test ought to be applied and I propose a review model to that end.
It is my view that the rationale of the implied freedom is a minimalist model of judicially protected popular sovereignty. This conception acknowledges that the conditions for sovereignty of the people are ultimately determined and enforced by the judiciary, for it is they who define and apply the implied freedom. But when discharging this role the courts must keep firmly in mind that the purpose of the implied freedom is to secure the effective functioning of our constitutional system of representative and responsible government. That is, to guarantee the democratic framework through which ‘the people’s representatives in Parliament … [can] implement the wishes of the people as they see them’[65] subject to the Constitution.[66]
There are three reasons that support my view regarding the rationale of the implied freedom. First, it fits best with the history and logic of the Australian Constitution. This is underpinned by a trust and faith in the ability of parliamentary government and the common law to secure the liberty of the individual and the collective welfare of the citizenry. This found constitutional expression in a variety of ways. Most notably, in the conscious decision of the framers to reject, for the most part, constitutional rights in favour of common law and parliamentary protections and to establish only the architecture for representative and responsible government with key aspects of its content and evolution left to the wisdom of Parliament.[67]
Second, the freedom of communication tradition in Australia underlines why the classic trio of free speech/expression rationales – the search for truth, the promotion of individual autonomy and the argument from democracy or self-government – do not underpin the implied freedom and why a minimalist model of judicially protected popular sovereignty does. This tradition is founded on a Constitution that provides only for the residual protection of individual liberties, its primary concern being the effective functioning of parliamentary government to secure the sovereignty of the people and provide the conditions for its meaningful exercise. The classic trio of rationales are located within a range of common law and statutory rules and operate to protect and enhance freedom of communication in these varied, sub-constitutional contexts. For example, the importance of the value of truth is central to the defamation defence of justification whilst the purpose of the defences of qualified privilege for government or political matters and absolute privilege for communications made during parliamentary proceedings, operate to promote democracy and selfgovernment.[68] And third, it recognise s the limited institutional capacity of the judiciary to determine what is necessary for the effective operation of our system of constitutional government. For a question of this nature will often have as much to do with politics and sociology as the law. In addition, there may be legitimate separation of powers concerns if the judiciary ignores the political nature of this task and second-guesses the judgement of Parliament on a question that takes it beyond its field of expertise and experience.[69]
In sum, a minimalist model of judicially protected popular sovereignty translates to a more limited, supervisory role for the judiciary in the interpretation and application of the implied freedom without, of course, leaving the issue of a law’s validity to be determined by the Parliament or executive. Jeremy Kirk makes the further point that the implied nature of the freedom raises ‘arguments of democratic legitimacy’.[70] He concludes that in relation to the implied freedom, ‘[t]he appropriate response is caution’[71] and the level of deference to the Parliament ought to be more than for express constitutional guarantees but less than when employed for the purposes of characterisation.[72] That seems about right.[73] It now falls to consider in more detail the degree of deference that ought to be applied at the different proportionality levels when assessing the compatibility of laws with the implied freedom.
But first, it is worth remembering that the test of proportionality need only be considered if a law effectively burdens political communication. A number of scholars have highlighted the analytical challenge that the infringement question poses for a court and made criticisms of the High Court’s sometimes dogmatic treatment of this question in their implied freedom jurisprudence.[74] There was also a further, more significant development on this point in the judgments of McHugh, Gummow and Hayne, Callinan and Heydon JJ in Mulholland. The central issue in that case was whether a law offended the implied freedom for denying a political party the right to have their name appear on the ballot paper for failing to satisfy two of the registration requirements under the relevant legislation. The first was the ‘500 rule’ which said that a party must have 500 members to be entitled to registration unless they have at least one member in Parliament. The second was the ‘no overlap rule’ that ‘prohibit[ed] two or more parties from relying on the same person as a member in calculating the number of members’.[75] These judges held that the law did not burden freedom of political communication ‘[b]ecause the [party] ha[d] no right to make communications on political matters by means of the ballot-paper other than what the Act gives’.[76] This finding was grounded in the following proposition:
[The implied freedom] gives immunity from the operation of laws that inhibit a right or privilege to communicate political and governmental matters. But, as Lange shows, that right or privilege must exist under the general law.[77]
That is, as a threshold requirement, there must be a pre-existing right of communication under the common law or statute before an issue as to the implied freedom can logically arise. This was said to be the consequence of the negative nature of implied freedom – it operates to limit legislative and executive power, it does not create or confer personal rights upon individuals.[78] But in my view this reasoning may operate to distort its proper application in some instances. For example, as these judges have noted, the bedrock concern of the implied freedom is to protect freedom of communication to the extent ‘necessary for the effective operation of that system of representative and responsible government provided for by the Constitution’.[79] That is the essence of the implied freedom and represents the primary duty of the court in its interpretation and application. It ‘has an institutional rather than individual foundation in that it is designed to facilitate the operation of representative government and not, except incidentally, to promote the general welfare of the individual’.[80] So it is correct to emphasise the negative nature of the freedom and deny it as a source of individual rights. Lange makes clear that the court must consider whether a law burdens freedom of political communication in its terms, operation or efect.[81] Therefore, if the effect of a law is to burden a species of communication considered necessary for the effective operation of constitutional government then the implied freedom is prima facie infringed and legislative or executive power is limited to that extent.[82] Such a limitation may result in an individual obtaining some kind of benefit or privilege. This benefit or privilege is not sourced directly to or housed within the implied freedom but flows incidentally from the limitation upon legislative or executive power necessary to secure constitutional government in that instance.[83]
The proposition derived by Callinan and Heydon JJ in Mulholland from the concession made by the appellant ‘that a legislative prohibition on the appearance of any party affiliation on the ballot paper would not contravene the implied freedom’[84] highlights, in my view, the problematic nature of this reasoning:
It follows that to legislate for a mixture of permissions and prohibitions, so as to permit the party affiliations of some candidates but not others to appear on the ballot paper, cannot interfere with the implied freedom.[85]
It is probably true that a blanket prohibition would not offend the implied freedom. Whilst it may deprive many voters of useful information, it cannot be said that the effective operation of constitutional government is precluded as every citizen must then rely upon or seek alternative sources of information to inform their votes. However, on Callinan and Heydon JJ’s reasoning, a law, for example, that prohibits party names on a ballot paper with the exception of the Australian Labor Party and the Liberal Party could not, by definition, burden the implied freedom.[86] But if Gleeson CJ and Gummow and Hayne JJ were correct in considering this information important for many voters based on actual voting practice[87] and a significant advantage to those parties on the ballot paper[88] (and I think that they are), then there is an argument that a law of this nature – that discriminates heavily in favour of the two major parties – distorts the reality of political communication to an extent that may imperil the effective operation of constitutional government. If so, then the implied freedom ought to be applied and failure to do so, because an individual in that instance may incidentally derive a legal benefit or privilege, is to ignore the operation and effect of the law on actual political communication. I say may because the Constitution would not in this instance compel the Parliament to provide all political parties with a statutory right to have their names appear on the ballot paper. For as noted, a law with a blanket prohibition, though not ideal for voters, would not on that count offend the implied freedom.
In any event, it is not inconsistent with the negative nature of the freedom for an individual to benefit incidentally from its application, as explained above. Indeed in some cases, its proper application will require it. That is not to assert that the implied freedom is a font of free-standing individual rights. It simply recognises that in protecting the freedom of communication necessary for the effective operation of constitutional government, there may be situations where an individual will obtain incidentally a legal benefit or privilege. This, in my view, honours rather undermines the essence of the implied freedom.[89]
My review model will now be outlined. In doing so, I am not suggesting that the current High Court has or will endorse this model. However, central aspects of it received support in some of the recent judgments in Coleman and Mulholland as will be detailed below. This model is informed by my rationale for the implied freedom and embodied within the proportionality framework.
The requirement of suitability (first level proportionality) is not difficult to satisfy. A law is suitable if it is ‘an effective, appropriate or rational means of achieving the claimed end’.[90] As Kirk explains, this question ‘serves as an objective test of purpose … [It] assesses whether the measure can in fact be characterised as having been made for that claimed purpose’.[91] No assessment, however, is made as to whether the claimed end is legitimate, simply whether ‘the measure can in fact be characterised as having been made for that claimed purpose’.[92] So the efectiveness of a law at this stage ought not to be an inquiry as to whether it is the ‘most practically desirable or effective way to achieve the end’.[93]
It is, however, possible to be quite rigorous even with first level proportionality. An example is the judgment of McLachlin J in the Canadian Supreme Court case of R v Keegstra[94] (‘Keegstra’) regarding a Canadian criminal provision that made it an offence punishable by imprisonment for two years to wilfully promote hatred against any section of the public distinguished by colour, race, religion or ethic origin:[95]
Where … there is an indication that the measure may in fact detract from the objectives it is designed to promote, the absence of a rational connection between the measure and the objective is clear … In my view, s 319(2) of the Criminal Code falls in this class of case … [I]t is far from clear that it provides an effective way of curbing hate-mongers.[96]
This introduces a strict qualitative component into the suitability analysis. Such an approach is inconsistent with its treatment in Australian constitutional law and is best considered at the second and third levels where different types of balancing processes are undertaken.[97] Employing a non-qualitative approach at this level in the proportionality analysis is, moreover, consistent with the rationale of the implied freedom and the more limited, supervisory judicial role that it entails.
The most significant effect that my rationale of the implied freedom has on the proposed review model occurs at the level of necessity (level two proportionality). For, in relation to the implied freedom, it is at this level where the amount of deference shown to the Parliament is likely to have the greatest impact on the final determination of a law’s validity. In its strictest form, this proportionality level asks ‘whether the legislation impairs the right to the minimum extent possible’.[98] Or in American constitutional law parlance, whether a law is ‘necessary to serve a compelling state interest and … narrowly drawn to achieve that end’.[99] The presence of these constitutional notions in the High Court’s implied freedom jurisprudence has been noted.[100] It is, however, my argument that not only ought the two-tier approach be abandoned but that the validity of a law should not be imperilled by the fact that the means least restrictive of the protected communication interest were not employed. This proposition involves the rejection of the nexus between more rigorous judicial scrutiny and a strict necessity or overbreadth-style analysis that exists in American constitutional law, amongst some judges on the Canadian Supreme Court and in the two-tier approach.
In the context of my argument, as earlier noted, more rigorous judicial scrutiny is appropriate when the effect or practical operation of the impugned law is to significantly infringe communication on political or government matters. It is then incumbent on the State to convince the court that the law has a ‘compelling justification’ and that it provides benefits that outweigh its restrictive effect on ‘political communication’.[101] This cost-benefit analysis[102] is undertaken mostly at the third level of the proportionality inquiry.
In any event, level two proportionality ought not to involve a strict necessity or overbreadth-style analysis in the context of the implied freedom for two reasons. First, a minimalist model of judicially protected popular sovereignty recognises the limited institutional capacity of a court to determine what is necessary for the effective operation of representative and responsible government under the Australian Constitution. The difficulty is that such an inquiry will often have as much to do with politics and sociology as the law. The compelling justification/overbreadth nexus ignores or fails to sufficiently appreciate this judicial limitation. Whilst an appellate court can require lower courts to build a strong factual record before hearing a case, invite relevant amicus curiae briefs and utilise experienced court staff such as masters and registrars to assist in its deliberations, the fact is that analysis of this kind drags courts as institutions and judges in particular some distance from their traditional adjudicative function. As one American judge succinctly put it, ‘judges are trained in the law. They are not penologists, psychiatrists, public administrators, or educators’.[103]
There is, in my view, something amiss when a judge strikes down a racial vilification law based on little more than intuition, (dubious) logic and unsupported factual proposition in a system where constitutional rights are not absolute but may be qualified by reasonable laws that ‘can be demonstrably justified in a free and democratic society’.[104] This was the case with the dissenting judgment of McLachlin J in Keegstra. The assertions of her Honour included that ‘it is far from clear that [a criminal provision] provides an effective way of curbing hate-mongers. Indeed, many have suggested it may promote their cause.’[105] The evidence of the ‘many’ was a newspaper article that suggested that the criminal trial of a person for racist expression may have given the unrepentant accused ‘a million dollars worth of publicity’[106] and the status of a martyr. The support for the latter claim was a passage from a Franz Kafka novel.[107] The judge further queried the congruence of the purpose and efficacy of the relevant (criminal) racial vilification law in the following manner:
The argument that criminal provisions for this kind of expression will reduce racism and foster multiculturalism depends on the [dubious] assumption that some listeners are gullible enough to believe the expression if exposed to it.[108]
This is a sweeping psychological and sociological proposition for a judge to make, especially in the absence of any empirical support. The treatise of Canadian lawyer and academic A Alan Borovoy was, moreover, the sole support cited for the two propositions that were critical to McLachlin J’s ultimate finding of invalidity. The first was that ‘[h]istorical evidence … gives reason to be suspicious of the claim that hate propaganda laws contribute to the cause of multiculturalism and equality’.[109] The evidence provided by Borovoy was the existence of racial vilification laws in pre-Hitler Germany and the fact that ‘this type of legislation proved ineffectual on the one occasion when there was a real argument for it’.[110] On this logic, one ought to do away with the principle of the separation of powers, federalism, freedom of the press and association and even the rule of law (‘Rechtsstaat’), as these constitutional principles and institutional arrangements were also ineffectual when really put to the test during the rise of Nazism in the 1930s.[111] The second proposition doubted ‘whether criminalization of expression calculated to promote racial hatred is necessary’[112] based on the conclusion drawn by Borovoy that ‘[h]uman rights legislation, focusing on reparation rather than punishment, has had considerable success in discouraging such conduct.’[113] If these represent the best reasons and evidence for substituting the considered opinion of a democratic legislature on a matter of serious public concern with that of an appellate court judge, then the limited institutional capacity of a court to properly determine these sorts of difficult issues is manifest.
Secondly, according to a minimalist model of judicially protected popular sovereignty, the implied freedom exists to guarantee the democratic framework mandated by the Constitution so ‘the people’s representatives in Parliament … [can] implement the wishes of the people as they see them’.[114] This is consistent with both the history and logic of the Constitution (which, for better or worse, is one of trust not distrust of government)[115] and our freedom of communication tradition, where the Constitution is concerned with the efficacy of parliamentary government to secure the sovereignty of the people and provide the conditions for its meaningful exercise.[116] On the other hand, the constitutional protection accorded to the individual to communicate freely stems incidentally from the institutional fortification performed by the implied freedom.[117] So the relevant constitutional duty of the High Court when measuring a law against the implied freedom is to protect political communication to the extent necessary to secure our system of constitutional government.[118] This is a different inquiry with a different focus to that which characterises a strict necessity or overbreadth-style analysis. For it is quite possible for a court to uphold the validity of a law (which entails a finding that the political communication necessary to secure our system of constitutional government is preserved) without employing the means least restrictive of the protected communication interest.[119]
It is my argument, that in the context of the implied freedom, the Canadian approach to necessity has much to commend it. In practice, the Canadian Supreme Court has been more deferential to the Parliament than the phrase ‘minimal impairment’ (their necessity/level two proportionality synonym) would suggest. This is particularly so when reviewing laws whose subject matters fall outside traditional areas of judicial expertise.[120] The very strict language and approach employed in R v Oakes[121] (the law ‘should impair “as little as possible” the right or freedom in question’[122] ) was quickly modified in the cases that immediately followed. The nature and complexity of these constitutional disputes made the Court realise that some level of deference to the Parliament was not only desirable but necessary.[123] Peter Hogg noted two conundrums the Court would face if Oakes-style strictness at this proportionality level were maintained. The first concerns the application of such a test in a federation, a point relevant to the Australian context:
A strict application of the least-drastic-means requirement would allow only one legislative response to an objective that involved the limiting of a Charter right. The law that least impaired the Charter right would be acceptable; all alternatives would fail. In a federal country like Canada, there ought to be some room for distinctive provincial responses to similar social objectives.[124]
The second, more important one is that
[i]n view of the ease with which a less drastic alternative to virtually any law could be imagined, the process of s 1 justification looked like the camel passing through the eye of the needle.[125]
The Court, therefore, recognised that
[a] reasonable limit is one which, having regard to the principles enunciated in Oakes, it was reasonable for the legislature to impose. The courts are not called upon to substitute judicial opinions for legislative ones as to the place at which to draw a precise line.[126]
In the later case of Attorney-General of Quebec v Irwin Toy Ltd[127](‘Irwin Toy’), the Court provided some important, additional comments as to how its ‘minimal impairment’ analysis (level two proportionality) may vary depending on the subject matter of the law:
whenever the government’s purpose relates to maintaining the authority and impartiality of the judicial system, the courts can assess with some certainty whether the ‘least drastic means’ for achieving the purpose have been chosen, especially given their accumulated experience in dealing with such questions … The same degree of certainty may not be achievable in cases involving the reconciliation of claims of competing individuals or groups or the distribution of scarce government resources.[128]
Importantly, the more deferential approach was appropriate when the Court was ‘called upon to assess competing social science evidence respecting the appropriate means for addressing [a] problem’.[129]
This Court will not, in the name of minimal impairment, take a restrictive approach to social science evidence and require legislatures to choose the least ambitious means to protect vulnerable groups. There must nevertheless be a sound evidentiary basis for the government’s conclusions.[130]
For these reasons a similar approach to necessity is appropriate and necessary in the application of the Lange test.[131] It was employed by a majority of the Canadian Supreme Court in Keegstra when considering whether there was minimal impairment of the freedom of expression guarantee by the Canadian Criminal Code race hatred provision.[132] It recognises the limited institutional capacity of a court to answer with certainty the complex, multi-layered social science questions that inevitably arise in this review analysis. When measuring laws in terms of necessity, Australian courts ought, therefore, to be ‘looking for a reasonable legislative effort to minimize the infringement of the [implied freedom], rather than insisting that only the least possible infringement could survive’.[133] This is consistent with the rationale of the implied freedom and the more limited, supervisory judicial role it entails. It is worth noting here that a number of judges in Coleman and Mulholland expressed support for the proposition that the application of the Lange test ‘does not call for nice judgments as to whether one course is slightly preferable to another’[134] or ‘a judicial conclusion that the law is the sole or best means of achieving that [legitimate] end’.[135] As Gleeson CJ noted:
the word ‘necessary’ has different shades of meaning. It does not always mean ‘essential’ or ‘unavoidable’, especially in a context where a court is evaluating a decision made by someone else who has the primary responsibility for setting policy.[136]
Importantly, it may reasonably be argued that this approach was central to the reasoning of McHugh J in Coleman and Finn J of the Federal Court in Bennett v President, Human Rights and Equal Opportunity Commission.[137] In both cases the impugned laws were held to offend the implied freedom and were invalidated. These cases demonstrate that the review model approach I propose still provides meaningful scrutiny of legislative and executive action even though the judicial role in this area is a more limited and supervisory one.
One final point regarding laws that deal with a subject matter as sensitive and controversial as racial vilification, is that they are usually the product of a long gestation period, are subject to rigorous parliamentary debate and committee scrutiny and often represent the final response to a range of expert reports and inquiries.[138] In my view and on the review model proposed, this parliamentary history may be relevant to the question of necessity. I do not wish to suggest that the level of parliamentary scrutiny can determine the validity of a racial vilification or that the end legislative product of this process will always display optimum sensitivity to the relevant constitutional right. But, the process itself and its considerable depth may provide material support for an argument that a reasonable legislative effort was made to frame a law that minimises the infringement of the implied freedom. It should at least give an appellate court judge reason for pause before striking down a law in the name of constitutional necessity if there is ‘a sound evidentiary basis for the government’s conclusions’.[139]
This level of the proportionality inquiry considers whether ‘the restrictions or detriments caused [by the law] outweigh the importance of the end of the beneficial result achieved’.[140] The amount of work the balancing stage (level three proportionality) has to do in the application of the Lange test will depend on the extent to which a law restricts or infringes the implied freedom. For example, if a law significantly restricts political communication then it will require a ‘compelling justification’ or ‘overriding public purpose’ to be valid. But even a law with a compelling justification may infringe the freedom so seriously that its benefit will not outweigh its detriment. On the other hand, it will be relatively easy for the State to justify a law where its restrictive impact is minimal.
It is worth noting that if a law is intact when it reaches the balancing stage, a court has already found that the law is an effective, appropriate or rational means to secure its claimed end and represents a reasonable legislative effort to minimise the infringement of the implied freedom. It may follow that in many cases an ultimate finding of validity is likely, particularly so when the purpose of or justification for a law is compelling.[141] But it is certainly possible under my review model for a law to be ‘necessary’ but fail the balancing test. Consider, for example, a law enacted during wartime that restricts the publication of material that may seriously undermine the prosecution of the war. This would be a rational end to help secure the State’s goal of a national and unified war effort. If it were the only workable option open to the legislature, then it would be a reasonable effort to frame a law that minimises its infringement of the implied freedom. However, this measure may so seriously infringe political communication or the protected interest may be considered so important in the circumstances that any benefit the law achieves is outweighed by the detriment it causes.[142]
So in the application of the Lange test, this level of the proportionality inquiry may have important work to do if the law seriously infringes the implied freedom or in the circumstances the protection of political communication is considered paramount.
The High Court is yet to consider the compatibility or otherwise of any Australian racial vilification law with the implied freedom. There have, however, been four judicial and administrative decisions that have done so.[143] They are Kazak v John Fairfax Publications,[144] Deen v Lamb,[145] Jones v Scully[146] and Islamic Council of Victoria v Catch the Fire Ministries.[147] The most important was the decision of the Federal Court in Jones v Scully.[148]
In Jones v Scully, the respondent argued that the racial vilification provisions in Part IIA of the Racial Discrimination Act 1975 (Cth) (‘RDA’) were invalid. Curiously, Hely J did not expressly consider the important threshold question of whether racial vilification even amounted to political communication. The Lange test was in fact applied, so one can assume that it was so characterised.[149] If it was not considered to be political communication, then no issue as to the validity of these racial vilification provisions on account of the implied freedom could logically arise. The respondent had published and distributed a pamphlet to the residents of a suburb in Launceston, Tasmania, claiming amongst other things that Jews were anti-democratic, immoral, sexually deviant and tyrannical.[150] It is difficult, however, to see the possibility of this communication affecting federal voting choices. This, according to Lange, is the nexus required before a communication is considered ‘political’ and accorded constitutional protection.[151] It seems clear enough, however, that on this formula racial vilification can amount to political communication.[152] For example, a vicious and hateful anti-Arab immigration diatribe at a political rally that called on Australians to vote for Jane Doe (a federal candidate with this policy) would prima facie be protected political communication even though it constitutes unlawful racial vilification.[153]
In any event, the relevant focus of the constitutional inquiry is whether the impugned law in its terms, operation or effect is compatible with the implied freedom.[154] So notwithstanding the doubt as to whether the relevant communications in Jones v Scully amounted to political communication and the absence of consideration to that end, the constitutional analysis undertaken by Hely J was still required and relevant. In this regard, after consideration of Australia’s international obligations under the International Convention on the Elimination of All Forms of Racial Discrimination, Hely J had little trouble in finding that the ‘legitimate end sought to be obtained by the RDA is the elimination of racial discrimination’.[155] He then said that ‘bearing in mind the exemptions available under s 1 8D, Part IIA of the RDA is reasonably appropriate and adapted to serve the legitimate end of eliminating racial discrimination’.[156]
The Lange test requires, however, consideration of the logically prior and more complex issue of whether the impugned law in fact burdens the implied freedom.[157] In this instance, it was at least arguable that the implied freedom was not infringed, because conduct that amounts to racial vilification is still lawful if a free speech/public interest defence is available. That is, if the conduct was done reasonably and in good faith for an academic, artistic, scientific, research or public interest purpose, it is lawful. Though instances of unlawful racial vilification amounting to constitutionally protected political communication may be few, they are possible as the Jane Doe hypothetical illustrates. So Hely J was probably correct to hold that ‘[i]t is conceivable that the restrictions imposed by … the RDA might in certain circumstances effectively burden freedom of communication about government and political matters.’[158] What is of concern is the absence of justificatory reasoning on a question of some complexity before this conclusion was reached. The failure to properly address this difficult question, the cornerstone of the relevant constitutional review analysis, occurs in another racial vilification case where the same issue was considered and the implied freedom jurisprudence considered more generally.[159] It is perfectly understandable for a judge to be ill at ease with such an inquiry. The determination of the substantive impact of an impugned law on the freedom of political communication ‘must be a highly speculative matter of political science and political philosophy which is very dependent on what particular conception of representative government is involved’.[160] The limited institutional capacity of a court to confidently answer a question of this nature is not therefore surprising. But this fact ought to be acknowledged rather than glossed over or, worse still, ignored. Notwithstanding the lack of detailed constitutional analysis, in each of the four cases the impugned Australian racial vilification law was found to be compatible with the implied freedom.[161] Two further points of possible import emerged from these cases. First, in two decisions the view was expressed that the relevant racial vilification provisions would be invalid but for the presence of the broadly defined free speech/public interest defences.[162] Second, none of the judges expressly employed the two-tier approach. This may simply reflect the fact that three of the four decisions were made in an administrative (quasi-judicial) setting where detailed constitutional analysis is considered inappropriate.[163] But at least in the case of Jones v Scully it may be a conscious decision to employ a single test for constitutionality (as endorsed in Lange) irrespective of whether the impugned law regulates the content or mode of a communication. I will return to this important point and elaborate on its possible relevance in the next section of the article.
Even though the issue of the constitutionality of Australian racial vilification laws has received little judicial scrutiny, one might argue that they ought not to be in doubt considering the logic of the Lange test. After all, proscribing racial vilification in order to tackle racism and the serious harms and inequalities that it engenders are goals ‘compatible with the maintenance of the constitutionally prescribed system of representative and responsible government’.[164] This argument is further buttressed by the fact that in most Australian jurisdictions conduct that amounts to racial vilification is still lawful if done reasonably and in good faith for an academic, artistic, scientific, research or public interest purpose.[165] Indeed, notwithstanding the rigorous scrutiny employed by the majority judges in Australian Capital Television,[166] it was noted that the implied freedom may be infringed ‘for the protection or vindication of the legitimate claims of individuals to live peacefully and with dignity within such a society’.[167] Such a qualification seems tailor-made to cover laws proscribing racial vilification. Even in comparative terms, it seems incongruous that Australian racial vilification laws could be vulnerable when, for example, the widely drawn Canadian criminal provision earlier noted survives constitutional challenge within a legal framework that contains a comprehensive and free-standing right to freedom of expression in a constitutional Charter of Rights.[168]
Whilst it is my argument that Australian racial vilification laws are constitutionally sound,[169] it is by no means certain that the current or a future High Court would agree when one considers the open-ended nature of the Lange test and the fact that the precise scope of the implied freedom remains far from settled. It is well to remember that at the time of the passage of the Racial Hatred Bill 1995 (Cth)[170] through the House of Representatives, former Commonwealth Solicitor-General Sir Maurice Byers thought it was clearly unconstitutional due to the fact that ‘freedom of communication on public affairs and political discussion is, in truth, no different from freedom of speech’.[171] Indeed, one of the two central questions regarding the scope of the implied freedom that remain presently open is the definition of political communication. This was not considered in Lange[172] and has received only minimal further elucidation in later cases.[173] But as noted, at least for the purposes of this article it seems clear enough that racial vilification can amount to ‘political communication’.[174]
In any event, what is important for the purposes of this part of the article is the re-emergence of the two-tier approach in the implied freedom cases after Lange.[175] As earlier noted, this approach embodies to some extent the American constitutional law principles of ‘strict scrutiny’ and ‘overbreadth’.[176] Should a majority of the High Court endorse and then take seriously the logic of its two principles then Australian racial vilification laws are constitutionally vulnerable.[177] For the primary concern of these laws is to restrict certain political ideas and information not the modes of their communication. So on the two-tier approach, if a particular racial vilification law were challenged, strict or more rigorous scrutiny must be applied and its constitutionality may turn on whether the measures it employed were least restrictive of the implied freedom.[178] And if a judge is so disposed, it takes no great feat of judicial ingenuity to devise an alternative racial vilification law that may place a lesser burden on the implied freedom.
In Western Australia, for example, only criminal sanctions exist for the regulation of racial vilification. It would be quite reasonable for a judge in an overbreadth analysis to ‘suggest that the very fact of criminalization itself may … represent an excessive response to the problem of hate propagation’.[179] ‘Criminalization’ may not, moreover, be ‘necessary’ if ‘[o]ther remedies are perhaps more appropriate and more effective’ considering that the ‘sanction of the criminal law may pose little deterrent to a convinced hate-monger who may welcome the publicity it brings’.[180] These were some of the arguments made by the three minority judges of the Canadian Supreme Court in Keegstra to strike down as overbroad the provision of the Canadian Criminal Code earlier noted.[181] It follows that the proper application of the two-tier approach and its principles of strict scrutiny and overbreadth may well invalidate the Western Australian criminal provisions and a range of other Australian racial vilification laws besides.[182]
It was argued earlier that the two-tier approach ought to be abandoned for it is inconsistent with the Lange test properly understood. It may operate to distort the application of the test by establishing an analytical framework grounded in a sometimes misleading threshold distinction between laws that regulate the content as opposed to the mode of a political communication. Instead, I advocated the single test for constitutionality unanimously endorsed by the High Court in Lange and proposed a review model informed by the rationale of the implied freedom to this end. That review model will now be applied to current Australian racial vilification laws to assess their compatibility or otherwise with the implied freedom.
The Jane Doe hypothetical amounts to political communication for the subject matter of the communication is such that it may reasonably be considered relevant to the federal voting choices of its likely audience.[184] If the Commonwealth law (the RDA) proscribes this communication (and I think that it does), it necessarily infringes the implied freedom because its primary purpose is to restrict that species of communication. The RDA proscribes this communication for it is ‘reasonably likely, in all the circumstances, to offend, insult, or humiliate another person or a group of people’ and race and ethnic origin is clearly one reason for the communication.[185] The communication also falls outside the free speech/public interest defences for not being made ‘reasonably and in good faith’.[186] Whilst the case law is unsettled on this point, I have argued elsewhere that the better view is that ‘reasonably’ refers to the method by which the communication is made not its content.[187] Making this communication at a public rally is clearly ‘reasonable’ in this sense. But the case law suggests that the language and behaviour used by the communicator must also be reasonable.[188] So making this political communication in a vicious and hateful manner would, therefore, be considered unreasonable. Moreover, the fact that a communication is made in this way will not only make it unreasonable but may evidence an absence of ‘good faith’. In other words, the extreme nature of a racist communication may suggest that its underlying purpose is vilification rather than a bona fide contribution to political discourse.[189] As French J recently noted in Bropho v Human Rights and Equal Opportunity Commission:[190]
good faith requires more than subjective honesty and legitimate purposes. It requires, under the aegis of fidelity or loyalty to the relevant principles in the Act, a conscientious approach to the task of honouring the values asserted by the Act.[191]
It can, therefore, be concluded that the RDA may ‘effectively burden freedom of communication about government or political matters either in [their] terms, operation or effect’.[192]
If in the Jane Doe hypothetical the communication would ‘incite hatred towards, serious contempt for, or severe ridicule of, a person or group of persons on the ground of the race of the person or members of the group’, then a similar conclusion can be made regarding the racial vilification laws of New South Wales, South Australia, the Australian Capital Territory, Queensland, Victoria and Tasmania. These laws employ this same, more elevated harm threshold[193] and their civil provisions incorporate free speech/public interest defences equivalent to those found in the RDA.[194] This communication will not attract a free speech/public interest defence as noted above. Moreover, for the communication to attract civil liability, it need not be shown that the speaker subjectively intended to incite hatred, serious contempt for, or serious ridicule of another or that any incitement in fact occurred.[195] It is enough if it is likely to do so in an objective sense. So it seems clear enough that a racist communication of this nature, made in a vicious and hateful manner at a political rally, must objectively incite hatred, serious contempt for, or serious ridicule of another on the ground of their race.[196] Therefore, the racial vilification laws of New South Wales, South Australia, the Australian Capital Territory, Queensland, Victoria and Tasmania also ‘effectively burden freedom of communication about government or political matters either in [their] terms, operation or effect’.[197]
The only racial vilification law that differs significantly from this format comes from Western Australia. This law creates eight criminal offences.[198] It was originally enacted and recently amended in direct response to the racist activities of a local white supremacist group and was tailored to combat that specific menace.[199] Considering the breadth of the offences available under Western Australian law, it seems clear enough that the racist communication described in the Jane Doe hypothetical would attract criminal liability.[200] The important point is that it too will burden freedom of political communication in its terms, operation or effect.
The crucial step in the proposed review model is to make a clear assessment as to the significance of the detriment to political communication effected by a racial vilification law. In other words, once the ‘freedom of [political] expression interest at stake’[201] is ascertained, this informs the proper application of the Lange test. If, for example, the RDA provisions operate to restrict an important species of communication or the ability of a group to participate meaningfully in the political process, then the State will need to establish that, notwithstanding the significant detriment to political communication, the law has a ‘compelling justification’ and is reasonably tailored to this end. So, determining the ‘freedom of [political] expression interest at stake’ is the base-point from which the necessity and balance of a racial vilification law can be meaningfully assessed.[202]
In one sense, any political communication infringement is significant, for it denies to the citizenry that which the Constitution has singled out for protection. But that protection is not absolute. So if the benefit outweighs the political communication detriment, then such a law will buttress rather than undercut constitutional government. It means that not every political communication is equally valued, constitutionally. In review terms, the lower the value of the political communication infringed, the easier it is for the government to justify its law. It is my argument that very little political communication of value is burdened by Australian racial vilification laws. Whilst ascribing ‘value’ is mostly a subjective assessment, in the context of my argument it refers to the fact that no instance of racial vilification that amounts to political communication falls foul of Australian civil law (and is thereby punished or chilled) on the basis of its content unless it was not made ‘reasonably and in good faith’. So long as a person employs a reasonable method for making their political communication and is not acting for an improper purpose, they can do or say whatever they want, even if it entails inciting ‘hatred towards, serious contempt for, or severe ridicule of, a person or group of persons on the ground of the race of the person or members of the group’.
The extent of this freedom of political communication sensitivity can be illustrated by the Jane Doe hypothetical with slightly changed facts. Let us say that the person makes the anti-Arab immigration speech at the political rally in a manner that is impassioned and robust rather than vicious and hateful. It would then be made reasonably and in good faith, in the relevant legal sense. During the speech the speaker cites, again in good faith, a spate of recent violent crimes in Sydney whose perpetrators were all recently migrated young Arab men as support for his/her view, though the racial profile of most Sydney criminals is, in fact, very different. This person will likely offend the harm threshold for, in the current political climate and in the context of the ongoing (so-called) ‘war on terror’, this communication is likely in an objective sense to ‘incite hatred towards, serious contempt for, or even severe ridicule’ of Arabs on the ground of their race. Even though this communication may be factually inaccurate and racially prejudiced, it communicates a potentially common political viewpoint. And, so long as it is made reasonably and in good faith, it is a ‘political communication’ of ‘value’ in our constitutional system of parliamentary government and one that is properly protected under the laws of New South Wales, South Australia, the Australian Capital Territory, Queensland, Victoria and Tasmania. In this way, the ‘value’ of a political communication is preserved both constitutionally and legislatively, even in circumstances where it amounts to racial vilification.
In those State and Territory jurisdictions (New South Wales, South Australia, Western Australia, the Australian Capital Territory, Queensland and Victoria) where serious instances of racial vilification attract criminal liability, the ‘freedom of [political] expression interest at stake’ is, in most cases, even lower. For example, in these jurisdictions the only kind of political communication infringed is that by which ‘a public act, incite[s] hatred towards, serious contempt for, or severe ridicule of, a person or group of persons on the ground of the[ir] race … by means which include threatening, or inciting others to threaten, physical harm towards, or towards any property of, the person or group of persons’.[203] In Western Australia, the only kind of political communication infringed is when the communication (which can include the possession of threatening or abusive written or pictorial material whether or not publication is intended) is made with the intent to harass or create, promote or increase hatred, serious contempt or harassment of a racial group or a member of that group.[204] That such (extreme) conduct is unnecessary to effectively communicate an idea or viewpoint that may reasonably be relevant to the federal voting choices of its likely audience seems clear enough, at least under Australian political conditions. It is also worth noting that these criminal provisions require proof beyond a reasonable doubt of the mens rea to incite or intend the relevant proscribed harm or consequence.[205] It further lessens the ‘value’ of any political communication that is so characterised. Indeed, to sanction racist harassment, incitement and (the threat of) violence as political communication invites the destruction of constitutional government. For if any means may be employed to make a political communication, we can be sure that persons or groups with an extreme racist agenda will do so. This serves to undermine the very reason why political communication is accorded constitutional status in the first place – to secure the conditions that provide for the meaningful exercise of the sovereignty of the people through our system of constitutional government.
There are, however, four offences in Western Australia that do not require proof of a mens rea for criminal liability to arise.[206] So it is a crime, for example, for any person to engage in conduct that is likely to harass a racial group or member of a racial group.[207] If a political communication can attract strict criminal liability, the freedom of expression interest at stake would seem considerable. But for each of these strict liability offences the same kind of broadly defined, free speech/public interest defences discussed above are available.[208] So as long as a person in Western Australia chooses a reasonable method for making their political communication and is not acting for an improper purpose, they can do or say whatever they like, even if it is likely to harass or create, promote or increase hatred of or serious contempt for a racial group or member of a racial group.
This analysis demonstrates that the ‘freedom of [political] expression at stake’ in each of the Australian racial vilification laws is not significant. But at least some political communication is restricted, however marginal it may be. So the answer to the first limb of the Lange test must be ‘yes’: the laws do effectively burden freedom of political communication. It now falls to consider the application of the second limb of the Lange test under the proposed review model.
First level proportionality is a relatively easy standard for Australian racial vilification laws to meet. For they need only be ‘an effective, appropriate or rational means of achieving the claimed end’.[210] No assessment is made at this level as to the legitimacy of that end or desirability of the measures contained in the law. Whilst the Commonwealth, State and Territory racial vilification laws employ a range of divergent regulatory mechanisms, they share a common purpose – to address the issue of racism and the serious harms and inequalities that it engenders. The Commonwealth and Tasmanian Parliaments have chosen to rely exclusively on civil law and, to this end, have established a two-step process. The first step involves an attempt at confidential conciliation between the parties facilitated by the Commonwealth Human Rights and Equal Opportunity Commission (‘HREOC’) and the Tasmanian Anti-Discrimination Commissioner, respectively. In the event that the conciliation process is unsuccessful, there is the option, at the Commonwealth level, for the matter to proceed to judicial determination[211] and in Tasmania, to the Anti-Discrimination Tribunal, a division of its Magistrates Court.[212]
The laws of New South Wales, the Australian Capital Territory, Queensland and Victoria contain similar and sometimes identical civil provisions but have, in addition, created criminal offences for conduct amounting to serious or aggravated racial vilification.[213] South Australia has also developed an innovative, multi-pronged legislative regime involving a criminal offence for serious racial vilification,[214] the availability of up to $40 000 civil damages as a remedy for the criminal offence[215] and a statutory tort for racial victimisation.[216] As noted, Western Australia relies exclusively on criminal sanctions to address both its specific racist menace and proscribe only the more serious instances of racial vilification.
The specific details of these regulatory mechanisms demonstrate considered and often context-sensitive legislative responses to racial vilification. This is no surprise, for the laws were invariably preceded by a combination of parliamentary reports, commissions of inquiry, detailed scrutiny and robust debate when passing through the respective Legislatures.[217] The important point is that they are effective, appropriate or rational measures to secure their claimed end.
Even if an Australian racial vilification law is an appropriate or rational measure to secure its claimed end, it still ought to be struck down if it is not ‘a reasonable legislative effort to minimize the infringement of the [implied freedom]’.[218] At first blush, there may be some doubt as to whether this is the case with the RDA. The imposition of a relatively low and objective harm threshold (‘offend, insult, humiliate or intimidate’) may result in the proscription of much valued political communication.[219] But the broad sweep of the free speech/public interest defences available and the fact that they were expressly included to protect and promote freedom of communication and shore up the constitutionality of the provisions in relation to the implied freedom,[220] is strong evidence that the RDA is a reasonable legislative effort by the Commonwealth to minimise the political communication infringement. It might, however, be argued that the law burdens political communication more than is strictly necessary, for an otherwise available defence is lost if the communication is not made reasonably and in good faith. But on the review model proposed, necessity need not entail the employment of legislative means least restrictive of political communication. So long as a court is satisfied that the Commonwealth is cognisant of and has made a reasonable legislative response to the ‘freedom of [political] expression interest at stake’ in its racial vilification law (and it ought to be), then it satisfies this proportionality level. The same conclusion can be made regarding the civil provisions in the racial vilification laws of New South Wales, the Australian Capital Territory, Queensland, Victoria and Tasmania. They provide equivalent free speech/public interest defences but have a higher harm threshold (‘hatred, serious contempt or severe ridicule’) than the RDA. So these State and Territory laws will necessarily meet the test for necessity for they operate to proscribe less political communication[221] than the RDA, which has been found to be proportional in the relevant sense.
On the other hand, it might be argued that, in those jurisdictions where both civil and criminal sanctions are employed (New South Wales, the Australian Capital Territory, South Australia, Queensland and Victoria), the additional use of the criminal law is excessive and makes that which is initially proportionate (the civil provisions) no longer so. In Keegstra, McLachlin J made a number of arguments along these lines. They included: that criminal provisions will not deter hate-mongers; that such provisions make free speech martyrs of the accused; that civil sanctions are more effective; and that criminal law procedures and sanctions are too severe and operate to significantly chill legitimate expression.[222] I have earlier argued that many of these propositions are dubious and lack empirical support.[223] But even if accepted as true, the State is not required to rebut these claims nor to demonstrate that the legislative means least restrictive of the implied freedom were, in fact, employed to meet the relevant test for necessity. If the criminal provisions are a ‘reasonable legislative effort to minimize the infringement of the [implied freedom]’, they pass constitutional muster. To this end, and as shown in my earlier analysis on the significance of the detriment caused by these provisions, little political communication of value attracts criminal sanctions. The only kind of political communication infringed is that by which ‘a public act, incite[s] hatred towards, serious contempt for, or severe ridicule of, a person or group of persons on the ground of the[ir] race … by means which include threatening, or inciting others to threaten, physical harm towards, or towards any property of, the person or group of persons’. Or when the political communication (which can include the possession of threatening or abusive written or pictorial material whether or not publication is intended) is made with the intent to harass or create, promote or increase hatred, serious contempt or harassment of a racial group or a member of that group, or would be likely to do so.[224]
The extent of the political communication sensitivity becomes apparent when it is understood that under most of the laws criminal liability is attracted only with the presence of an aggravating factor (for example, threat to do violence to person or property or an intent to create, promote or increase hatred of or serious contempt for, or harassment towards a racial group or member of a racial group). In other words, the idea or viewpoint contained in the ‘political communication’ is of no concern to these criminal laws. The need to prove mens rea also ensures that no political communication that unintentionally or even negligently incites the relevant, very serious harm will have criminal consequences.[225] For the strict liability offences in Western Australia, the availability of the broadly defined, free speech/public interest defences ensures that a political communication cannot attract criminal sanctions if it is made in good faith and in a manner which is reasonable. So it can be seen that the respective Parliaments have made reasonable efforts to minimise the infringement of the implied freedom when framing criminal laws to punish the most serious instances of racial vilification. Moreover, as these civil and criminal provisions individually satisfy the test for necessity, their combined use in these State and Territory laws can brook no constitutional objection. The comments of Dickson CJ in Keegstra are relevant in this regard. That case also involved a legislative scheme where both criminal and civil racial vilification provisions were utilised:
In my view, having both avenues of redress at the state’s disposal is justified in a free and democratic society. I see no reason to assume that the state will always utilize the most severe tool at hand, namely, the criminal law, to prevent the dissemination of hate propaganda. Where the use of the [criminal] sanction … is imprudent, employing human rights legislation may be the more attractive route to take, but there may equally be circumstances in which the more confrontational response of criminal prosecution is best suited to punish a recalcitrant hate-monger. To send out a strong message of condemnation, both reinforcing the values underlying [the criminal provision] and deterring the few individuals who would harm target group members and the larger community by intentional communication of hate propaganda, will occasionally require use of the criminal law.[226]
But does this passage contain the implicit suggestion (or at least possibility) that a legislative regime that uses criminal provisions only to regulate racial hatred might not satisfy the relevant test for necessity? If so, the Western Australian criminal provisions may be constitutionally vulnerable. It must, however, be kept in mind that the only relevant question for a court at this proportionality level, for any kind of racial vilification provision, is whether the law itself represents a reasonable legislative effort to minimise the infringement of the implied freedom. But as earlier noted, the Western Australian provisions law effect minimal political communication infringement. Criminal liability is imposed only when an aggravating factor is present and upon proof of mens rea or when a political communication was not made reasonably and in good faith and is likely to harass or create, promote or increase hatred of or serious contempt for a racial group or member of a racial group. Indeed, one might reasonably argue that using only criminal provisions to regulate racial hatred displays more sensitivity to the implied freedom by preserving a greater space for ‘political’ and other forms of communication without attracting the possibility of legal sanction. This point alone would suggest that it is proportional in this sense for a Parliament to use a criminal provision as the sole means for regulating racial vilification.[227] In any event, whilst the Western Australian law provides for criminal sanctions only, it is still open to the citizens of that State to pursue a civil remedy through the RDA.[228] So, legal coverage in Western Australia in fact extends to both civil and criminal sanctions in the same way as the racial vilification laws of New South Wales, the Australian Capital Territory, Queensland and Victoria. For these reasons, the Western Australian criminal provisions also satisfy the test for necessity.
As earlier noted, if in the application of the Lange test on the review model proposed a racial vilification law is proportional when it reaches the balancing stage, an ultimate finding of validity is likely.[229] My analysis has shown that in each instance, the relevant racial vilification law is an effective, appropriate or rational or measure to secure its claimed end and a reasonable legislative effort has been made to minimise the infringement of the implied freedom. These laws operate to proscribe political communication of little value to our constitutional system of parliamentary government. A person may lawfully communicate any political idea or viewpoint, even when it amounts to racial vilification, so long as they make it reasonably and in good faith. These laws, moreover, do have a ‘compelling justification’ (to address the issue of racism and the serious harms and inequalities that it engenders) and provide a range of significant benefits to the victims of racial vilification and to the citizenry more generally. These include the provision of a civil and/or criminal remedy to victims of racial vilification who have often suffered serious psychological and physical damage. This, in turn, operates to protect their personal liberty and freedom of speech and promotes substantive legal equality as a consequence. These conditions serve to advance personal development, meaningful democracy and a tolerant citizenry. In sum, these racial vilification laws satisfy the balancing test for providing benefits that clearly outweigh the negligible detriment they have on ‘political communication’.
There is disagreement as to how the Lange test should be applied when assessing the compatibility of a law with the implied freedom. My analysis has shown that the validity of a law may depend on which review approach a judge chooses to employ. One aim of this article was to demonstrate that the proper application of the Lange test requires that one approach (the two-tier approach) be abandoned and to argue for a conception of that test which is informed by and further serves what I consider to be the rationale of the implied freedom.
In Part II of the article a review model was proposed, to this end, that incorporates a particular application of the test for proportionality. This model recognises that the constitutional duty of the court is to guarantee the necessary rather than optimal level of political communication for our constitutional system of government to effectively function in order to secure the meaningful exercise of the sovereignty of the people. It translates to a more limited supervisory judicial role in the application of the implied freedom. More specifically, it is argued that if a law is an effective, appropriate or rational means to secure a legitimate constitutional end and its benefits outweigh its detriment to political communication then it ought to be valid even when those means are not the least restrictive of the implied freedom.
In Part III this conception of the Lange test and the review model proposed were applied to Australian racial vilification laws to assess their compatibility with the implied freedom. Each law was found to be proportional in the relevant constitutional sense and therefore compatible with the implied freedom. In the language of Lange, they are reasonably appropriate and adapted measures to secure a legitimate, indeed compelling, constitutional purpose. If the citizenry still rejects the efficacy of these racial vilification laws the remedy is political not constitutional. This is what the rationale of a minimalist model of judicially protected popular sovereignty necessarily entails.
[*] School of Law, Deakin University. My thanks are due to Professor George Williams and the anonymous referees for providing valuable comments and suggestions on the earlier drafts of this article and the Faculty of Law at the University of Cape Town for providing such friendly and beautiful surrounds in which to write it. Thanks also to those who participated in the New Researchers, New Research conference held at the University of Adelaide on September 30 – October 2, 2004 where an earlier version of this paper was presented.
[1] [1997] HCA 25; (1997) 189 CLR 520. For the remainder of the article referred to as the ‘implied freedom’.
[2] The following Australian jurisdictions have racial vilification laws: Racial Discrimination Act 1975 (Cth) pt IIA; Anti-Discrimination Act 1977 (NSW) ss 20(C), (D). See also Criminal Code 1913 (WA) ss 76– 80H; Racial Vilification Act 1996 (SA) ss 3, 6; Wrongs Act 1936 (SA) s 37; Discrimination Act 1991 (ACT) ss 65–7; Anti-Discrimination Act 1991 (Qld) ss 124A, 131A; Racial and Religious Tolerance Act 2001 (Vic) ss 7–14, 24–5; Anti-Discrimination Act 1998 (Tas) ss 17, 19, 55.
[3] The constitutionality of Australian racial vilification laws has, however, been examined in four lower court judicial and quasi-judicial decisions. For a detailed analysis of these cases, see below Part III(A). For relevant academic commentary on the issue see Michael Chesterman, Freedom of Speech in Australian Law: A Delicate Plant (2000) 237–43; Luke McNamara and Tamsin Solomon, ‘The Commonwealth Racial Hatred Act 1995: Achievement or Disappointment’ [1996] AdelLawRw 9; (1996) 18 Adelaide Law Review 259, 278–83; Saku Akmeemana and Melinda Jones, ‘Fighting Racial Hatred’ in Commonwealth of Australia, Race Discrimination Commissioner, The Racial Discrimination Act: A Review (1995) 156– 62. It is worth noting that in the last Parliament – ie, before the 9 October 2004 election – the Australian Labor Party introduced the Racial and Religious Hatred Bill 2003, a Private Members’ Bill that was originally moved by Robert McClelland MP. The Bill provided for the following criminal sanctions in serious cases of racial vilification: it would be a crime to threaten property damage or physical harm to another person or group because of their race, colour, religion or national or ethnic origin; and engaging in public acts that have the intention and likely effect of inciting racial or religious hatred against a person or group would also be a crime. The criminal provisions are nearly identical to those that formed part of the original Racial Hatred Bill 1994 (Cth) but were deleted during its passage through the Parliament. See further Luke McNamara, Regulating Racism: Racial Vilification Laws in Australia (2002) 40–2.
[4] [1997] HCA 25; (1997) 189 CLR 520, 567–8.
[5] Levy v Victoria [1997] HCA 31; (1997) 189 CLR 579, 596–7 (Brennan CJ).
[6] Brennan J first endorsed this approach in Australian Capital Television Pty Ltd v Commonwealth [1992] HCA 45; (1992) 177 CLR 106, 159 and then again in Levy v Victoria [1997] HCA 31; (1997) 189 CLR 579, 598. This approach was endorsed by the Full Court of the Federal Court in Mulholland v Australian Electoral Commission [2003] FCAFC 91; (2003) 198 ALR 278, 289 (Black CJ, Weinberg and Selway JJ). More recently, McHugh J said ‘the reasonably appropriate and adapted test gives legislatures within the federation a margin of choice as to how a legitimate end may be achieved’: Coleman v Power [2004] HCA 39; (2004) 209 ALR 182, 209. On the other hand, Kirby J found the concepts of ‘margin of appreciation’ and ‘deference’ unhelpful in Mulholland v Australian Electoral Commission [2004] HCA 41; (2004) 209 ALR 582, 646. This appeared to reflect a view that such notions may ‘distract courts from their duty to uphold the law’.
[7] See Adrienne Stone, ‘Lange, Levy and the Direction of the Freedom of Political Communication Under the Australian Constitution’ [1998] UNSWLawJl 38; (1998) 21 University of New South Wales Law Journal 117, 131–4.
[8] Levy v Victoria [1997] HCA 31; (1997) 189 CLR 579, 647 (Kirby J).
[9] Ibid 619 (Gaudron J).
[10] Wojciech Sadurski also advocates a strict scrutiny test for the review of racial vilification laws primarily ‘due to the proximity of [racist] speech to a public debate on political issues’: Wojciech Sadurski, ‘Offending with Impunity: Racial Vilification and Freedom of Speech’ [1992] SydLawRw 14; (1992) 14 Sydney Law Review 163, 193.
[11] [2004] HCA 39; (2004) 209 ALR 182, 266–7.
[12] [2004] HCA 41; (2004) 209 ALR 582, 595.
[13] But see below Part II(C)(2)(b) for an argument that rejects the nexus between strict scrutiny and an overbreadth-style analysis in relation to the implied freedom.
[14] Kruger v Commonwealth [1997] HCA 27; (1997) 190 CLR 1, 128 (Gaudron J). A similar approach is apparent in two judgments in Levy v Victoria [1997] HCA 31; (1997) 189 CLR 579, 614–5 (Toohey and Gummow JJ), 619 (Gaudron J). But see Levy v Victoria [1997] HCA 31; (1997) 189 CLR 579, 598 where Brennan CJ explicitly rejects the notion that ‘overbreadth’ has any place in Australian constitutional law. It is worth mentioning here that in the more recent case of Roberts v Bass [2002] HCA 57; (2002) 212 CLR 1 both Gaudron and Kirby JJ applied the Lange test to assess whether the common law defence of qualified privilege was compatible with the implied freedom without employing the two-tier approach which they advocated in the earlier post-Lange decisions noted above. It was unclear whether this represented a complete retreat from this approach by both judges or that they considered it inappropriate when measuring the common law (as opposed to statute law) against the constitutional freedom. See 26–30 (Gaudron, McHugh and Gummow JJ), 58–63 (Kirby J). But in my view a bifurcated review approach is unsustainable as both statute and common law have the capacity to regulate the content and/or mode of a political communication. It is, however, worth noting that in Mulholland, Kirby J appeared to abandon the two-tier approach in favour of an approach that echoes the theory of judicial review propounded by John Hart Ely in Democracy and Distrust (1980): [2004] HCA 41; (2004) 209 ALR 582, 647–8. On this point, see below text accompanying n 27–9.
[15] On strict scrutiny and overbreadth in relation to the regulation of speech in American constitutional jurisprudence see Lawrence Tribe, American Constitutional Law (2nd ed, 1988) 789–804 and 1058–61 respectively; see also Sadurski, above n 10, 178–9 for a discussion of strict scrutiny in the context of Australian racial vilification laws.
[16] [1938] USSC 104; 304 US 144, 152 fn 4 (1938). It suggests that legislation restricting ‘those political processes which can ordinarily be expected to bring about repeal of undesirable legislation, is to be subjected to more exacting judicial scrutiny’.
[17] Perry Education Association v Perry Local Educators Association, [1983] USSC 34; 460 US 37, 45 (1983).
[18] Ely, above n 14, 111 (footnotes omitted and emphasis added).
[19] R v Keegstra [1990] INSC 224; [1990] 3 SCR 697, 761 (Dickson CJ) (emphasis added) endorsing the contextual approach outlined by McLachlin J in the case of Rocket v Royal College of Dental Surgeons of Ontario [1990] 2 SCR 232. But for a criticism of this contextual approach see Jamie Cameron, ‘The Past, Present, and Future of Expressive Freedom under the Charter’ (1997) 35 Osgoode Hall Law Journal 1.
[20] Jeremy Kirk, ‘Constitutional Guarantees, Characterisation and the Concept of Proportionality’ [1997] MelbULawRw 1; (1997) 21 Melbourne University Law Review 1, 17 (footnotes omitted and emphasis added). But for an argument in favour of a more rule-based approach to the application of the implied freedom ‘of which strict scrutiny is one example’ see Adrienne Stone, ‘The Limits of Constitutional Text and Structure: Standards of Review and the Freedom of Political Communication’ [1999] MelbULawRw 26; (1999) 23 Melbourne University Law Review 668, 702–8.
[21] R v Keegstra [1990] INSC 224; [1990] 3 SCR 697, 761 (Dickson CJ).
[22] Tony Blackshield has argued that, in the application of the implied freedom, the distinction made by some judges between laws that regulate content as opposed to the mode of a political communication is not significant. For those judges in practice use ‘substantially the same test’ as those who advocate a single test for constitutionality. ‘The Implied Freedom of Communication’ in Geoffrey Lindell (ed), Future Directions in Australian Constitutional Law (1994) 253–4. But see Stone, ‘The Limits of Constitutional Text and Structure’ above n 20, 685–6 where the author argues that with the two-tiered approach ‘the scales [are] already weighted in favour of the freedom of political communication … [which] … reduce[s] the flexibility accorded by the proportionality test’.
[23] R v Keegstra [1990] INSC 224; [1990] 3 SCR 697, 761 (Dickson CJ).
[24] [1992] HCA 45; (1992) 177 CLR 106.
[25] Kirk, above n 20, 17.
[26] On this point, see below Part II(C)(2) and Part III(C)(2).
[27] [2004] HCA 41; (2004) 209 ALR 582, 647.
[28] Ibid.
[29] Ely, above n 14, 103.
[30] Stone, ‘The Limits of Constitutional Text and Structure’, above n 20, 678 citing Lange [1997] HCA 25; (1997) 189 CLR 520, 567 where the Court said that ‘[i]n this context, there is little difference between the test of “reasonably appropriate and adapted” and the test of proportionality’.
[31] [2004] HCA 39; (2004) 209 ALR 182, 240.
[32] [2004] HCA 41; (2004) 209 ALR 582, 648–50.
[33] Ibid 594.
[34] Ibid (Gleeson CJ). On this point, see below text accompanying n 116.
[35] [2004] HCA 39; (2004) 209 ALR 182, 207. That part of the test would now read: ‘is the law reasonably appropriate and adapted to serve a legitimate end [in a manner] which is compatible with the maintenance of the constitutionally prescribed system of representative and responsible government?’ This rewording was endorsed by Kirby J: 233 and Gummow and Hayne JJ: 229.
[36] Ibid 205 (McHugh J), 230 (Gummow and Hayne JJ), 233 (Kirby J).
[37] It is worth noting here that McHugh J in Coleman made it clear that by inserting the phrase ‘in a manner’ into the second limb of the Lange test the Court requires that both the end and the means of an impugned law be compatible with constitutional government. However, in the same judgment at 207 he denies that this part of the Lange test involves the weighing or balancing of interests which is the essence of one component of the test for proportionality. If in fact McHugh J would reject the view that he applies something close to a proportionality test, then it is reasonable to ask how he would assess and apply the ‘in a manner’ part of the Lange test.
[38] See Kirk, above n 20, 2–5.
[39] Ibid 6.
[40] Ibid 7.
[41] Ibid 8–9 where Kirk outlines the ‘balancing’ process involved in third level proportionality.
[42] On this point, see above text accompanying n 31–3; Stone, ‘The Limits of Constitutional Text and Structure’, above n 20, 681–4 and Kirk, above n 20, 16–19.
[43] [1992] HCA 45; (1992) 177 CLR 106, 143; see also 235 (McHugh J).
[44] [1997] HCA 31; (1997) 189 CLR 579.
[45] Ibid 619; see also 614–5 (Toohey and Gummow JJ) and Kruger v Commonwealth [1997] HCA 27; (1997) 190 CLR 1, 128 (Gaudron J).
[46] Lange [1997] HCA 25; (1997) 189 CLR 520, 567.
[47] Kirk, above n 20, 63.
[48] Stone, ‘The Limits of Constitutional Text and Structure’, above n 20, 686.
[49] Kirk, above n 20, 5.
[50] For her important argument that the development of such a theory requires considerations external to the text and structure of the Constitution and is for this reason incompatible with the interpretive methodology outlined in Lange, see Stone, ‘The Limits of Constitutional Text and Structure’, above n 20, 696–9.
[51] [2004] HCA 39; (2004) 209 ALR 182, 207 (McHugh J).
[52] Ibid 192.
[53] [2004] HCA 41; (2004) 209 ALR 582, 586.
[54] Ibid 585.
[55] Ibid 590.
[56] Ibid 592. See also Coleman [2004] HCA 39; (2004) 209 ALR 182, 192 (Gleeson CJ).
[57] [2004] HCA 39; (2004) 209 ALR 182, 240 (Kirby J). See also 256 where Callinan J said he considered the second limb of the Lange test to be ‘somewhat inscrutable’ and that ‘an appreciation of what is reasonably appropriate and adapted to achieving a legitimate end may be very much a matter of opinion’. However, unlike Kirby J, he favours a more relaxed test, though continuing to express strong reservations as to the place of the implied freedom in Australian constitutional law.
[58] See Stone, ‘The Limits of Constitutional Text and Structure’, above n 20, 696 where Stone suggests that the High Court in Lange deliberately avoided articulating a theoretical basis for the implied freedom, although points out that strict adherence to the constitutional text and structure is itself a ‘philosophical commitment’: fn 166.
[59] There is, however, an argument that a judge ought to expressly articulate the why and let that inform the how. It would still allow the incremental development of the implied freedom but in a manner that is more likely to be coherent and principled. But see Adrienne Stone, ‘Freedom of Political Communication, the Constitution and the Common Law’ (1998) 26 Federal Law Review 219, 235, 238.
[60] The argument from truth is generally associated with the political theory of John Stuart Mill, On Liberty (People’s ed, 1913). It entered American constitutional jurisprudence through the famous dissent of Justice Holmes in Abrams v United States, [1919] USSC 206; 250 US 616, 630 (1919) (joined by Brandeis J dissenting): ‘[T]he best test of truth is the power of the thought to get itself accepted in the competition of the market, and that truth is the only ground upon which their wishes safely can be carried out. That at any rate is the theory of our Constitution.’
[61] The argument from self-government originates from the work of Alexander Meikeljohn, Political Freedom: The Constitutional Powers of the People (1965) and ‘The First Amendment is an Absolute’ (1961) Supreme Court Review 245.
[62] [1992] HCA 45; (1992) 177 CLR 106, 138–9, 145.
[63] Frederick Schauer, Free Speech: A Philosophical Inquiry (1982) 33. In relation to racial vilification laws, Sadurski considers that ‘a suspicion that politicians and legislators will overstate the degree of harm produced by a given activity’ is at the heart of why strict judicial scrutiny is appropriate: see above n 10, 178.
[64] See Dan Meagher, ‘What is “Political” Communication? The Rationale of the Implied Freedom of Political Communication’ [2004] MelbULawRw 14; (2004) 28 Melbourne University Law Review 438, 451; Part II(C).
[65] George Williams, Human Rights Under the Australian Constitution (1999) 230.
[66] It is worth noting here that Jeremy Waldron has developed a strong theoretical basis for the protection of rights through the ordinary democratic/legislative process. He argues ‘that our respect for such democratic rights is called seriously into question when proposals are made to shift decisions about the conception and revision of basic rights from the legislature to the courtroom, from the people and their admittedly imperfect representative institutions to a handful of men and women, supposedly of wisdom, learning, virtue and high principle’. ‘A Right-Based Critique of Constitutional Rights’ (1993) 13 Oxford Journal of Legal Studies 18, 20. My argument regarding the rationale of the implied freedom outlined in this part of the article similarly emphasises the centrality of the democratic process to the realisation of this particular constitutional right. But on my conception the judiciary still plays an important (though more supervisory) role in guaranteeing the democratic framework necessary to secure the sovereignty of the people and provide the conditions for its meaningful exercise. On the other hand, Waldron’s theory of constitutional rights highlights ‘the difficulty, complexity, and controversy attending the idea of rights’ (19) and the ongoing and inevitable disagreement amongst the citizenry as to their basic content. In these circumstances, he rejects the wisdom, democratic legitimacy and therefore moral authority of the courts to perform this constitutional role.
[67] On this point see John La Nauze, The Making of the Australian Constitution (1972) 227–32; Sir Harrison Moore, The Constitution of the Commonwealth of Australia (2nd ed, 1910) 78; Sir Owen Dixon, Jesting Pilate (2nd ed, 1997) 10 1–2; Sir Robert Menzies, Central Power in the Australian Commonwealth (1967) 54; Helen Irving, To Constitute a Nation (1997) 162–70; Mulholland [2004] HCA 41; (2004) 209 ALR 582, 586–8 (Gleeson CJ); Meagher, ‘What is “Political” Communication?’, above n 64; Part II(C)(1).
[68] See further Meagher, ‘What is “Political” Communication?’, above n 64; Part II(C)(2).
[69] On this point, see Part II(C)(3).
[70] Kirk, above n 20, 61. See also 55 and 61–3 where Kirk also notes the relevance of the limited institutional capacity of courts to determine issues of this nature to the level of judicial deference to the Parliament in the application of the implied freedom.
[71] Ibid 56.
[72] Ibid 62–3.
[73] There is considerable strength in the Kirk view that ‘[t]he court generally should be more deferential in the characterisation context than for constitutional guarantees, although an intermediate category of deference may be justified for implied guarantees’: ibid 63. As he notes, in the context of the implied freedom, its implied nature raises a democratic concern that does not exist for the characterisation of express legislative powers, so the complex social and political science evidence and issues that inevitably arise support an intermediate category of deference.
[74] See, eg, Tom Campbell, ‘Democracy, Human Rights and Positive Law’ [1994] SydLawRw 16; (1994) 16 Sydney Law Review 195, 201–4; Gerald Rosenberg and John Williams, ‘Do Not Go Gently into that Good Right: The First Amendment in the High Court of Australia’ (1997) Supreme Court Review 439, 458–64; Deborah Cass, ‘Through the Looking Glass: The High Court and the Right to Speech’ in Tom Campbell and Wojciech Sadurski (eds), Freedom of Communication (1994) 184–91. The failure to properly address the infringement question has also been a feature of those lower court decisions where racial vilification laws have been challenged, as my analysis below will demonstrate: see Part III(A).
[75] Mulholland [2004] HCA 41; (2004) 209 ALR 582, 583 (Gleeson CJ).
[76] Ibid 614 (McHugh J), 631–4 (Gummow and Hayne JJ), 673–4 (Callinan J), 678–9 (Heydon J).
[77] Ibid 614 (McHugh J) citing a passage in his judgment in Levy [1997] HCA 31; (1997) 189 CLR 579, 622. This passage was endorsed at 632–3 (Gummow and Hayne JJ) and 678–9 (Heydon J).
[78] Ibid.
[79] See Coleman [2004] HCA 39; (2004) 209 ALR 182, 206 (McHugh J), 229 (Gummow and Hayne JJ), 256 (Callinan J), 270 (Heydon J).
[80] Williams, above n 65, 168.
[81] [1997] HCA 25; (1997) 189 CLR 520, 567.
[82] A law of this kind can, of course, still pass constitutional muster if it meets the second limb of the Lange test.
[83] There is a parallel here with the argument made by Adrienne Stone that the High Court was incorrect in Lange to hold that, due to the negative nature of the implied freedom, it did not apply to the common law, except indirectly. She argues, persuasively in my view, that the direct application of the Constitution to the common law is perfectly consistent with the negative nature of the implied freedom: Adrienne Stone, ‘Rights, Personal Rights and Freedoms’ [2001] MelbULawRw 13; (2001) 25 Melbourne University Law Review 374, 400–17.
[84] (2004) 206 ALR 582, 678 (Heydon J), 673–4 (Callinan J).
[85] Ibid 678–9 (Heydon J).
[86] Indeed, Heydon J does not even consider what appears on a ballot paper to be political communication, only a communication ‘between the executive government and the electors’: ibid 679. No other judge in Mulholland came to this conclusion. On the contrary, Gleeson CJ, correctly in my view, said that ‘[i]t is a communication about a matter that is central to the competitive process involved in an election’: ibid 592. Kirby J agreed: ibid 651. McHugh J considered it political communication but only in the limited sense of the ballot paper being a record of a voter’s electoral preference: ibid 610–1.
[87] Ibid 591 (Gleeson CJ), 620 (Gummow and Hayne JJ).
[88] Ibid 591–2 (Gleeson CJ), 619 (Gummow and Hayne JJ).
[89] See ibid 650–1 where Kirby J, for this reason, rejects the analysis of McHugh, Gummow, Hayne, Callinan and Heydon JJ on this point.
[90] Kirk, above n 20, 6 (emphasis added).
[91] Ibid.
[92] Ibid.
[93] Ibid 5.
[94] [1990] INSC 224; [1990] 3 SCR 697.
[95] Criminal Code, RSC 1985, c C–46, ss 3 19(2), (3).
[96] [1990] INSC 224; [1990] 3 SCR 697, 852.
[97] As Kirk notes, above n 20, 8 ‘in an indirect fashion, necessity may presuppose some weighing of interests’.
[98] R v Keegstra [1990] INSC 224; [1990] 3 SCR 697, 854 (McLachlin J).
[99] Perry [1983] USSC 34; 460 US 37, 45 (1983).
[100] See above text accompanying n 7–17.
[101] See further below Part III(C)(2).
[102] This is the phrase used by Kirk to describe the balancing process undertaken at the third proportionality level: Kirk, above n 20, 8.
[103] Frank M Johnson Jr, ‘The Role of the Federal Courts in Institutional Litigation’ (1981) 32 Alabama Law Review 271, 274.
[104] Canadian Charter of Rights and Freedoms, s1, pt I of the Constitution Act 1982, being Schedule B to the Canada Act 1982 (UK), 1982, c–11.
[105] Keegstra [1990] INSC 224; [1990] 3 SCR 697, 852 (McLachlin J) (emphasis added).
[106] Ibid 853 citing an article published in the Globe and Mail (Toronto), March 1, 1985, 1.
[107] Ibid citing The Trial (1976) 203.
[108] Ibid.
[109] Ibid 854.
[110] Ibid citing A Alan Borovoy, When Freedoms Collide: The Case for our Civil Liberties (1988) 50.
[111] See Michael Stolleis, The Law under the Swastika (1998) 1–22.
[112] Keegstra [1990] INSC 224; [1990] 3 SCR 697, 861.
[113] Ibid citing Borovoy, above n 112, 221–5.
[114] Williams, above n 65, 230.
[115] See above Part II(C)(1) and for a more detailed treatment see Meagher, ‘What is “Political” Communication?’, above n 64.
[116] But see Sadurski, above n 10, where the author, though concerned primarily with the ‘philosophy of free speech’ (165), situates his argument for strict scrutiny within the context of racial vilification laws existing in Australia and in other European and North American jurisdictions. Importantly, he states that so long as a society has a right to free speech, whether constitutional or otherwise, ‘the underlying idea adopted … is that the [strict scrutiny] framework is of universal application’(168). However, this claim of ‘universality’ is problematic. For there are fundamental differences in the nature of the free speech ‘rights’ that exist, for example, in Australia, Canada and the United States. Each right has a unique text, history and position within its constitutional framework. This is central to and suggestive of the kind of judicial review that best fits that constitutional tradition. This point was made recently by Gleeson CJ in Mulholland [2004] HCA 41; (2004) 209 ALR 582, 594 where he noted that regarding the test for proportionality in the application of the implied freedom, it is ‘important to remember, and allow for the fact, that [the test] has been developed and applied in a significantly different constitutional context’. Sopinka J of the Canadian Supreme Court has also noted that Canadian ‘courts have tended to be more deferential to governmental restrictions on freedom of expression. This is due to a number of factors. Canada evolved in a tradition of parliamentary supremacy where legislative decisions are final. Moreover, s 1 of our Charter specifically permits the government to justify infringements’: Justice John Sopinka, ‘Should Speech that Causes Harm be Free’ in Jane Duncan (ed), Between Speech and Silence: Hate Speech, Pornography and the new South Africa (1996) 134–5. Australia too has this tradition of parliamentary supremacy and a qualified right to freedom of speech. But compare this with the review approach of some judges of the Constitutional Court of South Africa who urge rigorous scrutiny of laws that burden the right to freedom of expression especially ‘[having] regard to our recent past of thought control, censorship and enforced conformity to governmental theories’ – S v Mamabolo [2001] ZACC 17; 2001 (3) SA 409 (CC) [37] (Kriegler J) (Chaskalson P, Ackermann, Goldstone, Madala, Mokgoro, Ngcobo, Yacoob JJ, Madlanga AJ and Somyalo AJ concurring). This passage was cited with approval in the majority judgment of Langa DCJ in Islamic Unity Convention v Independent Broadcasting Authority and Others [2002] ZACC 3; 2002 (4) SA 294 (CC). So in this regard Sadurski’s argument and framework for strict scrutiny cannot be ‘universal’ save in the most limited, abstract sense. It is not appropriate in the Australian context. For as earlier argued, the two-tier approach is incompatible with the Lange test and the need for rigorous scrutiny of laws that significantly burden the implied freedom ought not to imply the strict scrutiny/overbreadth nexus that exists, most notably, in American First Amendment law; see above, Part II(A).
[117] On this point see Meagher, ‘What is “Political” Communication?’, above n 64, Part II(B)(2)(ii).
[118] See Coleman [2004] HCA 39; (2004) 209 ALR 182, 206 (McHugh J), 229 (Gummow and Hayne JJ), 256 (Callinan J), 270 (Heydon J).
[119] The decision of the Canadian Supreme Court in Keegstra is an example. So too is the decision of the Federal Court in Jones v Scully (2002) 120 FCR 243. For it is clearly possible to frame a law less restrictive of the implied freedom than ss 1 8C, 1 8D of the Racial Discrimination Act 1975 (Cth). This would be done if s 1 8C embodied the higher harm threshold (‘incite hatred towards, serious contempt for, or severe ridicule of’) present in the racial vilification laws of the State and Territory jurisdictions or if the free speech/public interest defences in s 1 8D were more narrowly drawn.
[120] See Irwin Toy v Quebec [1989] 1 SCR 927, 994.
[121] R v Oakes [1986] 1 SCR 103.
[122] Ibid 139 (Dickson CJ).
[123] On this point see Peter Hogg, Constitutional Law of Canada (2000) 750–2.
[124] Ibid 752.
[125] Ibid 753.
[126] R v Edwards Books [1986] 2 SCR 713, 781–2.
[128] Ibid 994. There is some evidence that in the Australian context the High Court is also willing to more strictly scrutinise laws whose subject matters relate to the operation of the judicial system (its independence, procedures, accessibility) and that a more adventurous streak emerges when interpreting the scope of the separation of judicial power effected by Chapter III of the Constitution. On judicial independence, see Kable v Director of Public Prosecutions (NSW) [1996] HCA 24; (1996) 189 CLR 51; legal representation in criminal trials: Dietrich v The Queen [1992] HCA 57; (1992) 177 CLR 292; separation of judicial power effected by Chapter III of the Constitution: Re: Wakim; Ex parte McNally (1999) 198 CLR 511; due process: Leeth v Commonwealth (1992) 174 CLR 455. For an excellent critique of this approach in comparison with the strict textualism of Lange and its implied freedom progeny, see Williams, above n 65, 240–3.
[129] Irwin Toy [1989] 1 SCR 927, 994.
[130] Ibid 999.
[131] In the Canadian context, Hogg has noted that regarding ‘hate propaganda laws’ (756, fn 184) the Canadian Supreme Court has accorded the Parliament a margin of appreciation for ‘it does not take a vivid imagination to devise a law that would be less intrusive of the … Charter right than the law that was enacted’: Hogg, above n 123, 756.
[132] ‘It … must be shown that [the criminal provision] is a measured and appropriate response to the phenomenon of hate propaganda, and that it does not overly circumscribe the … guarantee’: Keegstra [1990] INSC 224; [1990] 3 SCR 697, 771 (Dickson CJ) (emphasis added).
[133] Hogg, above n 123, 754.
[134] Coleman [2004] HCA 39; (2004) 209 ALR 182, 209 (McHugh J).
[135] Ibid 267 (Heydon J).
[136] Mulholland [2004] HCA 41; (2004) 209 ALR 582, 594.
[137] [2003] FCA 1433; (2004) 204 ALR 119. In that case, Finn J held that regulation 7(13) of the Public Service Regulations 1998 (Cth) was invalid for offending the implied freedom, stating at 14 1–2 that ‘[i]t is one thing [for the Commonwealth] to regulate the disclosure of particular information for legitimate reasons relating to that information and/or effects of its disclosure. It is another to adopt the catch-all approach of reg 7(13) which does not purport either to differentiate between species of information of the consequences of disclosure … [I]t unreasonably compromises the freedom by transforming the freedom into a dispensation. It is not an appropriate filtering device to protect the efficient workings of government in a way that is compatible with the freedom.’ (emphasis added).
[138] For example, in Keegstra [1990] INSC 224; [1990] 3 SCR 697, 724–5, Dickson CJ outlined the pre-legislative history of the impugned criminal provision. This included the formation of the Special Committee on Hate Propaganda in Canada and the publication of a unanimous report in 1966. Luke McNamara has documented in considerable detail the same for the racial vilification laws of the Commonwealth, NSW, WA and SA: see Regulating Racism, above n 3, 35–49, 12 1–30, 222–37 and 259–79, respectively.
[139] Irwin Toy [1989] 1 SCR 927, 999.
[140] Kirk, above n 20, 8.
[141] Hogg makes a similar, though stronger, point in relation to a law that is constitutionally intact upon reaching the balancing component of the Canadian Charter of Rights and Freedoms s 1 limitation analysis: ‘If a law is sufficiently important to justify overriding a Charter right (first step), and if the law is rationally connected to the objective (second step), and if the law impairs the Charter right no more than is necessary to accomplish the objective (third step), how could its effects then be judged to be too severe?’: Hogg, above n 123, 757.
[142] However, during World War I, the United States Supreme Court held that a law with a similar operation did not offend the First Amendment: Schenck v United States [1919] USSC 64; 249 US 47 (1919). The defendant printed and distributed to men who had been drafted for military service a circular that said conscription was ‘despotism in its worst form and a monstrous wrong against humanity in the interest of Wall Street’s chosen few’. It exhorted them to resist the conscription law and not ‘support an infamous conspiracy’.
[143] There has also been a challenge to the racial vilification provisions in the Racial Discrimination Act 1975 (Cth) in Toben v Jones [2003] FCAFC 137; (2003) 199 ALR 1 but this was on characterisation grounds rather than for infringing the implied freedom.
[144] [2000] NSWADT 77 (Unreported, Hennessy, Farmer, Jowett, 22 June 2000).
[145] [2001] QADT 20 (Unreported, President Walter Sofronoff, 8 November 2001).
[146] [2002] FCA 1080 (Unreported, Hely J, 2 September 2002).
[147] [2003] VCAT 1753 (Unreported, Higgins J, 21 October 2003).
[148] [2002] FCA 1080 (Unreported, Hely J, 2 September 2002).
[149] Ibid [235]–[240].
[150] Ibid [44].
[151] Lange [1997] HCA 25; (1997) 189 CLR 520, 560–1.
[152] See further Meagher, ‘What is “Political” Communication?’, above n 64, Part III(A).
[153] For the reason why, see below Part III(C)(1).
[154] On this point see Tony Blackshield and George Williams, Australian Constitutional Law and Theory (3rd ed, 2002) 1196.
[155] Jones v Scully [2002] FCA 1080 (Unreported, Hely J, 2 September 2002) [239]. See also International Convention on the Elimination of All Forms of Racial Discrimination, opened for signature 21 December 1965, 660 UNTS 195 (entered into force 4 January 1969).
[156] Ibid [240].
[157] On this point, see above Part II(C)(1).
[158] Jones v Scully [2002] FCA 1080 (Unreported, Hely J, 2 September 2002) [238].
[159] The issue was not considered in Kazak v John Fairfax Publications [2000] NSWADT 77 (Unreported, Hennessy, Farmer, Jowett, 22 June 2000) and was given a mostly superficial treatment by the majority judges in Australian Capital Television [1992] HCA 45; (1992) 177 CLR 106. On this point regarding Australian Capital Television, see Campbell, above n 74, 201–4. The recent notable exception, however, was the case of Mulholland where there was detailed discussion of whether the impugned law burdened political communication; see above Part II(C)(1).
[160] Campbell, above n 74, 203.
[161] Islamic Council of Victoria v Catch the Fire Ministries [2003] VCAT 1753 (Unreported, Higgins J, 21 October 2003) found the Racial and Religious Tolerance Act 2001 (Vic) to be valid; Deen v Lamb [2001] QADT 20 (Unreported, President Walter Sofronoff, 8 November 2001) found s 124A of the Anti-Discrimination Act 1991 (Qld) to be valid; Jones v Scully found pt IIA of the Racial Discrimination Act 1975 (Cth) to be valid; Kazak v John Fairfax Publications [2000] NSWADT 77 (Unreported, Hennessy, Farmer, Jowett, 22 June 2000) found s 20C of the Anti-Discrimination Act 1977 (NSW) to be valid. On the constitutionality of Australian racial vilification laws see further Chesterman, above n 3, 238–43.
[162] See Jones v Scully [2002] FCA 1080 (Unreported, Hely J, 2 September 2002) [238]; Deen v Lamb [2001] QADT 20 (Unreported, President Walter Sofronoff, 8 November 2001) 2.
[163] My thanks are due to one of the anonymous referees for this point.
[164] Lange [1997] HCA 25; (1997) 189 CLR 520, 567.
[165] See Anti-Discrimination Act 1977 (NSW) s 20C(2); Wrongs Act 1936 (SA) s 37; Discrimination Act 1991 (ACT) s 66(2); Anti-Discrimination Act 1991 (Qld) s 124A(2); Racial and Religious Tolerance Act 2001 (Vic) s 11; Anti-Discrimination Act 1998 (Tas) s55.
[166] [1992] HCA 45; (1992) 177 CLR 106, 143 (Mason CJ), 169–76 (Deane and Toohey JJ), 234–5 (McHugh J).
[167] Ibid 174 (Deane and Toohey JJ).
[168] Keegstra [1990] INSC 224; [1990] 3 SCR 697.
[169] See below Part III(C).
[170] This was the law that incorporated the racial vilification provisions (pt IIA) into the Racial Discrimination Act 1975 (Cth).
[171] Sir Maurice Byers, ‘Free Speech a Certain Casualty of Race Law’, The Australian (Sydney) 21 November 1994, 11. Sir Maurice Byers also appeared as counsel in Australian Capital Television.
[172] On this point see Meagher, ‘What is “Political” Communication?’, above n 64, Part III(A). But see Chesterman, above n 3, 44–9.
[173] The most important case is Levy [1997] HCA 31; (1997) 189 CLR 579 where the High Court made clear that non-verbal conduct or symbolic speech ‘which is capable of communicating an idea about the government or politics of the Commonwealth’ counts as ‘political communication’: 595 (Brennan CJ). In addition, there have been two State Supreme Court decisions where political communication was held not to encompass discussions regarding the conduct of the judiciary: John Fairfax Publications Pty Ltd v Attorney-General (NSW) [2000] NSWCA 198; (2001) 181 ALR 694; Herald & Weekly Times & Bolt v Popovic [2003] VSCA 161 (Unreported, Winneke ACJ, Gillard and Warren AJJA, 21 November 2003). More recently, the High Court confirmed in Coleman and Mulholland that political communication included ‘[c]ommunications between the executive government and public servants and the people’: Mulholland [2004] HCA 41; (2004) 209 ALR 582, 610 (McHugh J) and see above n 86.
[174] It is worth noting, for precedential value, the broad conception of ‘political communication’ favoured in Theophanous v Herald and Weekly Times Ltd [1994] HCA 46; (1994) 182 CLR 104, that it ‘refers to all speech relevant to the development of public opinion on the whole range of issues which an intelligent citizen should think about’: 124 (Mason CJ, Toohey and Gaudron JJ) quoting Eric Barendt, Freedom of Speech (1987) 152. Deane J said ‘political communication and discussion “extends to all political matters, including matters relating to other levels of government within the national system [of government] which exists under the Constitution”’: 164 (footnotes omitted). This may well be good law absent a clear statement to the contrary by the current, very differently constituted, High Court. It was this definitional breadth and elasticity that lay at the heart of Sir Maurice’s constitutional concerns.
[175] The two-tier approach has found favour most recently with Heydon J in Coleman [2004] HCA 39; (2004) 209 ALR 182, 266–7 and Gleeson CJ in Mulholland [2004] HCA 41; (2004) 209 ALR 582, 595.
[176] See above Part II(A).
[177] Adrienne Stone, eg, considers that the two-tier approach is likely to prevail: see Stone, ‘Lange, Levy and the Direction of the Freedom of Political Communication Under the Australian Constitution’, above n 7, 134. Whilst writing at a time when only NSW and Western Australian had racial vilification laws, Wojciech Sadurski considered it unlikely that these kinds of group vilification laws would survive strict scrutiny: Sadurski, above n 10, 190.
[178] It should, however, be noted that even though Gleeson CJ and Heydon J have recently endorsed the two-tier approach, both judges reject a conception and application of the Lange test that requires a law to employ means least restrictive of the implied freedom – see Coleman [2004] HCA 39; (2004) 209 ALR 182, 267 (Heydon J); Mulholland [2004] HCA 41; (2004) 209 ALR 582, 594–5 (Gleeson CJ), 680 (Heydon J).
[179] Keegstra [1990] INSC 224; [1990] 3 SCR 697, 860–1 (McLachlin J).
[180] Ibid 861.
[181] Ibid 852–62 (McLachlin J) (La Forest and Sopinka JJ concurring).
[182] The racial vilification laws of NSW, the ACT, SA, Qld and Vic also contain criminal provisions. It should, however, be noted that, unlike the Canadian criminal provision reviewed in Keegstra, these State and Territory laws only affix criminal liability where there is an aggravating factor (threat to the person or property) accompanying the act of racial vilification: see below text accompanying n 224–26.
[183] Lange [1997] HCA 25; (1997) 189 CLR 520, 567.
[184] The likely audience of this communication would include attendees at the political rally and those persons whom may reasonably be informed of it by the media or other communication conduits. The relevance of the communication to their federal voting choices may, of course, have either a positive or negative impact. For a more detailed treatment of this point see Meagher, ‘What is “Political” Communication?’, above n 64; Part III(D)(1).
[185] Racial Discrimination Act 1975 (Cth) s 1 8C.
[186] Racial Discrimination Act 1975 (Cth) s 18D: an instance of racial vilification will be lawful if done reasonably and in good faith for an academic, artistic, scientific or other public interest purpose.
[187] The following cases endorse this view – Commonwealth: Jones v Scully [2002] FCA 1080 (Unreported, Hely J, 2 September 2002)[159]; Warner v Kucera (Unreported, Commissioner Johnston, 10 November 2000) 2, 33–4; Toben v Jones [2003] FCAFC 137; (2003) 199 ALR 1, 13 (Carr J),19 (Kiefel J), 38 (Allsop J); NSW: Wagga Wagga Aboriginal Action Group v Eldridge (1995) EOC 92–70 1, 78,268 (Bartley, Farmer and Luger); Hellenic Council of NSW v Apoleski and Macedonian Youth Association (No 1) [1995] NSWEOT (Unreported, Biddulph, Alt, Mooney, 25 September 1997) 16; Hellenic Council of NSW v Apoleski (No 2) [1995] NSWEOT (Unreported, Biddulph, Alt, Mooney, 25 September 1997) 10. Qld: Deen v Lamb [2001] QADT 20 (Unreported, President Walter Sofronoff QC, 8 November 2001) 2. For a detailed discussion on this point see Dan Meagher, ‘So Far So Good ?: A Critical Evaluation of Racial Vilification Laws in Australia’ (2004) 32 Federal Law Review 225; Part III(C)(2)(i).
[188] For example, in Wagga Wagga Aboriginal Action Group v Eldridge (1995) EOC 92–701, 78,268 (Bartley, Farmer and Luger) the respondent at a local council meeting challenged the bona fides of Aboriginal land claims in the Wagga Wagga region and the reconciliation process. But he did so using vulgar and odious racial epithets. The NSW Equal Opportunity Tribunal held that the respondent’s actions were not done reasonably or in good faith pointing out that ‘proper procedures to oppose the claim were available to Mr Eldridge, and there was no need to act as he did’ (78,268).
[189] For example, in Jones v Scully [2002] FCA 1080 (Unreported, Hely J, 2 September 2002) Hely J found at [197]–[198] that the respondent acted neither reasonably or in good faith in distributing a leaflet that, amongst other things, suggested that Judiasm is worse than a Satanic Cult. On this point, see further Meagher ‘So Far So Good?’, above n 187; Part III(C).
[190] [2004] FCAFC 16; (2004) 204 ALR 761.
[191] [2004] FCAFC 16; (2004) 204 ALR 761, 786.
[192] Lange [1997] HCA 25; (1997) 189 CLR 520, 567.
[193] See Anti-Discrimination Act 1977 (NSW) s 20C(1); Racial Vilification Act 1996 (SA) s 4 and Wrongs Act 1936 (SA) s 37; Discrimination Act 1991 (ACT) s 66; Anti-Discrimination Act 1991 (Qld) s 124A ; Racial and Religious Tolerance Act 2001 (Vic) s 7(1); Anti-Discrimination Act 1998 (Tas) s 17(1). In addition, the laws of NSW, SA, the ACT, Qld and Vic make criminal conduct that breaches this harm threshold but has an aggravating factor, the threat to person and/or property: see below Part III(C)(3)(b) for an examination of these criminal provisions.
[194] See Anti-Discrimination Act 1977 (NSW) s 20C(2); Wrongs Act 1936 (SA) s 37; Discrimination Act 1991 (ACT) s 66(2); Anti-Discrimination Act 1991 (Qld) s 124A(2); Racial and Religious Tolerance Act 2001 (Vic) s 11; Anti-Discrimination Act 1998 (Tas) s 55. The only difference in the content of the defences is that the Victorian law, unlike its NSW, ACT, SA, Qld and Tasmanian counterparts, does not protect conduct that is absolutely privileged under defamation law. This is of no great practical import as it is unlikely that racist conduct that attracts absolute privilege could then be the subject of a complaint under the relevant racial vilification provisions in the absence of express legislative sanction.
[195] See Wagga Wagga Aboriginal Action Group v Eldridge (1995) EOC 92–701, 78,268 (Bartley, Farmer and Luger); Kazak v John Fairfax Publications [2000] NSWADT 77 (Unreported, Hennessy, Farmer, Jowett, 22 June 2000); Western Aboriginal Legal Service Limited v Jones [2000] NSWADT 102 (Unreported, Rees, Silva and Luger, 31 July 2000). See further McNamara, Regulating Racism, above n 3, 182–7.
[196] In Wagga Wagga Aboriginal Action Group v Eldridge (1995) EOC 92–701, 78,268 (Bartley, Farmer and Luger) a case with similar facts, it was held that the respondent had objectively incited hatred, serious contempt for, or serious ridicule of this group on grounds of their race: see above n 188.
[197] Lange [1997] HCA 25; (1997) 189 CLR 520, 567.
[198] The Western Australian racial vilification law contains no civil, only criminal, sanctions. On the significance of this point for the law’s validity, see below text accompanying n 221–3.
[199] On the background to the original Western Australian criminal provisions, see McNamara, Regulating Racism, above n 3, 222–5. The original provisions contained in the Criminal Code 1913 (WA) made it a crime to publish, distribute or display written or pictorial material that is threatening or abusive intending to create or promote hatred of any racial group (s 78) or to be in possession of such materials for the same purpose (s 77). It was also a crime to publish written or pictorial material intending to harass any racial group (s 80) or to be in possession of such materials for the same purpose (s 79). These criminal provisions were repealed by the Criminal Code Amendment (Racial Vilification) Act 2004 (WA) and replaced with the following eight offences: It is a crime to engage in conduct, otherwise than in private, by which a person intends to create, promote or increase animosity towards (defined in s 76 to mean hatred or serious contempt), or harassment of, a racial group, or a person as a member of a racial group (s 77) or whose conduct is likely to do so (s 78). It is a crime for any person to possess written or pictorial material that is threatening or abusive, intending the material to be published, distributed, or displayed by that person or another and either intends the publication, distribution or display of the material to create, promote or increase animosity towards, or harassment of, a racial group, or a person as a member of a racial group (s 79) or the publication, distribution or display of the material would be likely to do so (s 80). It is a crime for any person to engage in conduct, otherwise than in private, by which the person intends to harass a racial group, or a person as a member of a racial group (s 80A) or to engage in conduct that is likely to do so (s 80B). It is a crime for any person to possess written or pictorial material that is threatening or abusive, intending the material to be displayed by that person or another and the person either intends the display of the material to harass a racial group, or a person as a member of a racial group (s 80C) or the display of the material would be likely to do so (s 80D). It is, however, a defence to a charge under ss 78, 80, 80B and 80D to prove that the accused person engaged in conduct, or intended the material to be published, distributed or displayed, reasonably and in good faith: (a) in the performance, exhibition or distribution of an artistic work; (b) in the course of any statement, publication, discussion or debate made or held, or any other conduct engaged in, for any genuine academic, artistic, religious or scientific purpose or any purpose that is in the public interest; or (c) in making or publishing a fair and accurate report or analysis of any event or matter of public interest (s 80G).
[200] At the very least, the Jane Doe communication is likely to harass a racial group, or a person as a member of a racial group (s 80B) and the unreasonableness of the language used to make the communication would preclude the availability of a free speech/public interest defence under s 80G.
[201] Keegstra [1990] INSC 224; [1990] 3 SCR 697, 761 (Dickson CJ).
[202] This is different from the two-tier approach, as the effect or practical operation of the law on political communication is the central consideration for the purposes of judicial review, not whether it regulates its content or method of expression: see above Part II(A).
[203] See Anti-Discrimination Act 1977 (NSW) s 20D; Racial Vilification Act 1996 (SA) s 4; Wrongs Act 1936 (SA) s 37; Discrimination Act 1991 (ACT) s 67; Anti-Discrimination Act 1991 (Qld) s 131A; Racial and Religious Tolerance Act 2001 (Vic) s 24.
[204] Criminal Code 1913 (WA) ss 77, 79, 80A, 80C.
[205] Whilst there is no authority on point due to the fact that no person has been prosecuted under the criminal provisions of Australian racial vilification laws, a court will find that mens rea is an essential component of every criminal offence unless there is a clear parliamentary intention to the contrary: Sweet v Parsley [1969] UKHL 1; [1970] AC 132; He Kaw Teh v The Queen [1985] HCA 43; (1985) 157 CLR 523.
[206] Criminal Code 1913 (WA) ss 78, 80, 80B, 80D.
[207] Criminal Code 1913 (WA) s 80B.
[208] Criminal Code 1913 (WA) s 80G: see above n 199 where the defences are outlined.
[209] Lange [1997] HCA 25; (1997) 189 CLR 520, 567.
[210] Kirk, above n 20, 6.
[211] Human Rights and Equal Opportunity Act 1986 (Cth) s 46PO.
[212] Anti-Discrimination Act 1998 (Tas) ss 13, 78.
[213] See above text accompanying n 203.
[214] Racial Vilification Act 1996 (SA) s 4.
[215] Racial Vilification Act 1996 (SA) s 6.
[216] Wrongs Act 1936 (SA) s 37.
[217] See above text accompanying n 138.
[218] Hogg, above n 123, 754.
[219] On this point see Meagher, ‘So Far So Good ?’, above n 187, text accompanying n 3 1–3.
[220] On this point see McNamara, Regulating Racism, above n 3, 53–4.
[221] The higher harm threshold in the State and Territory laws means that more ‘political communication’ is lawful in these jurisdictions than under the Racial Discrimination Act 1975 (Cth). The amount of ‘political communication’ left unregulated by the State and Territory laws is the ‘gap’ between the two harm thresholds. In other words, that range of communications that may reasonably offend, insult, humiliate or intimidate on the grounds of race but do not amount to hatred, serious contempt or severe ridicule, will not attract civil liability.
[222] Keegstra [1990] INSC 224; [1990] 3 SCR 697, 852–855 and 860–862.
[223] See above text accompanying n 104–113.
[224] See above Part III(C)(2).
[225] Dickson CJ makes a similar point in the relation to the Canadian criminal provision in Keegstra [1990] INSC 224; [1990] 3 SCR 697, 773–6.
[226] Ibid 785.
[227] In Keegstra, Dickson CJ noted that, in relation to addressing racial vilification, ‘the government may legitimately employ a more restrictive measure, either alone or as part of a larger programme of action’: [1990] INSC 224; [1990] 3 SCR 697, 785 (emphasis added), though no reason independent of the justification offered for the combined (civil and criminal) legislative package was given for this conclusion.
[228] For example, the overarching federal (civil) coverage provided by the Racial Discrimination Act 1975 (Cth) was a reason for the shape of SA racial vilification laws: see McNamara, Regulating Racism, above n 3, 267. It cannot, however, have been a factor in WA choosing only criminal sanctions, because its law was enacted in 1990, before the passage of the Racial Hatred Act 1994 (Cth) which incorporated the civil racial vilification provisions in the Racial Discrimination Act 1975 (Cth).
[229] See above text accompanying n 141.
AustLII:
Copyright Policy
|
Disclaimers
|
Privacy Policy
|
Feedback
URL: http://www.austlii.edu.au/au/journals/UNSWLawJl/2005/4.html