AustLII Home | Databases | WorldLII | Search | Feedback

Melbourne University Law Review

Melbourne Law School
You are here:  AustLII >> Databases >> Melbourne University Law Review >> 2022 >> [2022] MelbULawRw 2

Database Search | Name Search | Recent Articles | Noteup | LawCite | Author Info | Download | Help

Blore, Kent --- "Proportionality under the Human Rights Act 2019 (Qld): When Are the Factors in s 13(2) Necessary and Su!cient, and When Are They Not?" [2022] MelbULawRw 2; (2022) 45(2) Melbourne University Law Review 419


PROPORTIONALITY UNDER THE HUMAN RIGHTS ACT 2019 (QLD): WHEN ARE THE FACTORS IN S 13(2) NECESSARY AND SUFFICIENT, AND WHEN ARE THEY NOT?

Kent Blore[1]*

With the introduction of the Human Rights Act 2019 (Qld), limits on human rights in Queensland must now be justified. Section 13(1) sets out the test for justification and s 13(2) lists a number of factors that ‘may’ be relevan)t. At first blush, the word ‘may’ makes the factors seem like a smorgasbord, to be picked from as one fancies. This article contends that the factors in s 13(2) align closely with ‘structured proportionality’, such that the factors are prerequisites for a conclusion that a limit is justified, rather than optional considerations. However, there are a number of possible exceptions to that general rule which give the word ‘may’ work to do. These include ‘internal limitations’ which modify the test of justification, and the need for flexibility in how public entities go about giving proper consideration to human rights.

Contents

The Human Rights Act 2019 (Qld) (‘Human Rights Act’) seeks to introduce a ‘culture of justification’.[1] Now, whenever an act, decision or statutory provision in Queensland limits a human right, that limit must generally be justified according to the test set out in s 13 of the Human Rights Act. The overall test of justification set out in s 13(1) is whether the limit on a human right is ‘demonstrably justified in a free and democratic society’.[2] While that language may be ‘lofty’,[3] it was borrowed from the equivalent provision in Canada,[4] which has long been held to incorporate a structured proportionality analysis.[5] That is an analysis used in many human rights systems worldwide and involves four sequential steps: (i) legitimate aim or proper purpose, (ii) suitability or rational connection, (iii) necessity or the minimal impairment principle, and (iv) fair balance.[6] Were there any doubt that s 13 was intended to import structured proportionality, s 13(2) then sets out a number of factors to assist in answering the overall question of justification in s 13(1). Those factors spell out the four sequential steps of structured proportionality.

However, the factors are prefaced with a statement that they ‘may be relevant’.[7] On a cursory reading of s 13, the word ‘may’ might be interpreted as meaning that the factors are optional considerations and that other considerations which are not listed in s 13(2) might be relevant. Part II of this article canvasses the limited case law to date on s 13 of the Human Rights Act in order to show that the question of whether the factors are mandatory and exhaustive is a live one, calling for an answer.

This article contends that the word ‘may’ does not mean that the justification analysis can be undertaken by randomly picking and choosing which factors in s 13(2) to apply, or even by applying some entirely extraneous factor not mentioned in s 13 at all. This is for three reasons. First, the text of s 13 and its historical background reveal that it is intended to import structured proportionality. Part III of this article seeks to demonstrate this by tracing the origins of the text used in s 13(1) to the general limitations clause in Canada (via the general limitations clauses in the Australian Capital Territory (‘ACT’) and Victoria), and the embrace of structured proportionality by the Supreme Court of Canada. The article then looks to the South African origins of the factors in s 13(2) before setting out the important ways in which those factors depart from the South African general limitations clause in order to align even more closely with structured proportionality.

Secondly, quite apart from the text of s 13, as a matter of logic, each of the steps of structured proportionality is relevant to a conclusion that a limit on human rights is justified. Each step tests the reasonableness of the limit on human rights from a slightly different angle, such that skipping a step means failing to test the reasonableness of the measure from that angle. If a person only applies some of the factors in s 13(2) they might reach a conclusion that the limit is justified when that conclusion is not truly open — which would be obvious to the person if they only stopped to apply each of the factors in s 13(2). Alternatives to structured proportionality serve to entrench the risk of such an oversight. In particular, a global judgment about justification would reduce the analysis under s 13 to a matter of impression. Such an approach would lack transparency and undermine the culture of justification which the Human Rights Act seeks to introduce. Likewise, the alternative of applying a categorisation approach — that is, only applying some of the factors in s 13(2) in certain categories of cases — would mean skipping factors which might reveal that the limit is not justified. As will be seen, a categorisation approach may have its place, but it should only be applied where selecting the right category would always produce the same outcome as applying all the factors in s 13(2) in any event.

In Part IV, this article seeks to show why structured proportionality is superior to any alternative approach to justification, if not logically required for any justification analysis. Part IV does this by reference to insights from the High Court over the last seven years in the context of the implied freedom of political communication, and more recently the freedom of interstate trade, commerce and intercourse under s 92 of the Constitution. Without the guidance of an express general limitations clause like s 13 of the Human Rights Act, the High Court has had to work from first principles to arrive at the conclusion that structured proportionality is the only viable option for justifying limits on the implied freedom and s 92 of the Constitution. The lessons learned in the context of the constitutional freedoms apply equally to other rights and freedoms that are relative in nature, including the human rights enshrined in the Human Rights Act.

The third reason why the word ‘may’ does not allow a smorgasbord approach is that ‘may’ has other work to do. Having established that the factors in s 13(2) are all necessary as a general rule, in Part V, the article explores a number of possible exceptions to that rule, which would give the word ‘may’ in s 13(2) some work to do:

1 where an internal limitation in a human right modifies the test of justification;

2 where an absolute right such as the freedom against torture is limited;

3 where the courts develop a categorisation approach for a particular right or a particular kind of limit;

4 where a public entity gives proper consideration to human rights notwithstanding that they fail to engage in a structured proportionality analysis; and,

5 where a statement of compatibility accompanying a Bill adequately justifies a proposed limit on human rights notwithstanding that the statement does not apply structured proportionality.

Importantly, each of these exceptions is itself proportionate, and thus coheres with the logic of proportionality underlying s 13. Because the word ‘may’ has other work to do, we are not forced to read s 13(2) as though it offered a smorgasbord of factors to be considered or dismissed at will according to one’s idiosyncrasies or arbitrary whim.

Finally, in Part VI the article addresses the converse question of whether the factors in s 13(2) are sufficient. Despite the deep history of thinking about structured proportionality around the world, with one exception, theorists and judges have not managed to think of any additional element of justification which has not already been included in the list of factors in s 13(2). The one exception is already spelled out in s 13(1). In addition to being proportionate, a limit on human rights must be ‘under law’[8] — that is, authorised by law. Unless one of the exceptions identified in this article applies, the factors in s 13(2) are logically required and sufficient to answer the question of whether the limit on human rights is justified.

II Early Case Law on s 13: Why the Question Arises

Since the Human Rights Act commenced on 1 January 2020, Queensland courts and tribunals have been grappling with the justification analysis called for by s 13. Generally, courts and tribunals have simply recited s 13 and then focused attention on only one or two factors in s 13(2),[9] or even ignored all of the factors altogether.[10] To date, only one case has delved into what is required by s 13 in any depth: Australian Institute for Progress Ltd v Electoral Commission (Qld) (‘AIP’).[11]

AIP concerned the proper construction of the prohibited donor provisions of the Electoral Act 1992 (Qld).[12] On any available interpretation of the provisions, prohibiting people such as property developers from making political donations would mean less money would be available to fund political campaigns, and therefore limit the freedom of expression enshrined in s 21 of the Human Rights Act as well as the right to take part in public life enshrined in s 23.[13] The limit on these human rights enlivened the obligation in s 48 of the Human Rights Act to interpret the statutory provisions, if possible, in a way that is compatible with human rights.

As Applegarth J noted in AIP, ‘[t]he words “compatible with human rights” require consideration of ss 8 and 13’ of the Human Rights Act.[14] Section 8 defines ‘compatible with human rights’ as meaning either that the measure does not limit a human right, or that the measure does limit a human right but in a way that is ‘reasonable and demonstrably justifiable’ under s 13.[15] Section 13 then ‘provides a framework for deciding when and how a human right may be limited in a way which does not result in incompatibility’.[16] Subsection (1) ‘sets out the basic test for how a human right may be limited’.[17] It provides that ‘[a] human right may be subject under law only to reasonable limits that can be demonstrably justified in a free and democratic society based on human dignity, equality and freedom’. Subsection (2) then sets out a number of factors which ‘may be relevant’ in answering the overall question of justification in sub-s (1), and which are ‘intended to align generally with the principle of proportionality’.[18]

The equivalent legislation in the ACT and Victoria — the Human Rights Act 2004 (ACT) (‘ACT Human Rights Act’) and the Charter of Human Rights and Responsibilities Act 2006 (Vic) (‘Victorian Charter’) — contains equivalents of ss 13 and 48, but not s 8.[19] In the absence of a definition of ‘compatible with human rights’,[20] the relationship between the interpretive clause (s 48) and the general limitations clause (s 13) has been a vexed issue in those jurisdictions. In particular, the role of proportionality in interpreting legislation compatibly with human rights remains unsettled.[21] Under the clear direction provided by s 8 of the Human Rights Act, Applegarth J engaged in a robust proportionality analysis under s 13.[22] We will return to his Honour’s application of s 13 in the next Part when we unpack the components of s 13.

What is important for present purposes is that his Honour bookended that proportionality analysis with hints that a full structured proportionality analysis that works through all the factors in s 13(2) may not always be necessary or sufficient. He began by saying that it was ‘convenient’ to address the factors in s 13(2) because they ‘may be relevant’ to deciding the test of justification set out in s 13(1).[23] So much reflects the language of s 13(2), which uses the permissive ‘may’. That reading is also reinforced by the second reading speech, in which the Attorney-General stated that ‘these factors are only a guide’.[24]

Then, having addressed each of the factors set out in s 13(2), his Honour returned to s 13(1) as though the cumulative answer to the factors in s 13(2) may not be enough to answer the overall test of justification set out in s 13(1).[25] Again, that reading is supported by the extrinsic material. The explanatory notes state that the s 13(2) factors are ‘not exhaustive’,[26] but do not offer any clue as to what other factors may be relevant and in what circumstances.

The case law on s 13 to date — and, in particular, AIP — raises the question of whether the factors in s 13(2) are necessary, or whether some or all of the factors may be entirely ignored. AIP also raises the question of whether the factors in s 13(2) are sufficient, or whether some extraneous factor may prove decisive in the justification analysis. The answers to those questions are to be found in the text of s 13, read in light of the background of its historical antecedents as well as its underlying logic.

III Section 13 Imports a Structured Proportionality Analysis

Section 13(1) sets out the overall test for whether a limit on human rights is justified. It provides:

A human right may be subject under law only to reasonable limits that can be demonstrably justified in a free and democratic society based on human dignity, equality and freedom.[27]

The subsection bears an obvious textual affinity with the chapeau of s 7(2) of the Victorian Charter, as well as s 28(1) of the ACT Human Rights Act. The progenitor of all three Australian provisions was s 1 of the Canadian Charter of Rights and Freedoms adopted in 1982.[28] It provides:

The Canadian Charter of Rights and Freedoms guarantees the rights and freedoms set out in it subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society.[29]

In 1986, in the seminal case of R v Oakes, the Supreme Court of Canada held that inhering in s 1 is the structured proportionality test.[30] That test had developed in Germany and had already spread to other European human rights systems.[31] Structured proportionality continued to spread, and is now applied so ubiquitously in human rights jurisprudence that it has been said to have assumed a global constitutional status (with the notable exception of the United States (‘US’)).[32] Most agree that structured proportionality involves four sequential steps:[33]

(i) does the measure have an [sic] legitimate aim sufficient to justify the limitation of a fundamental right; (ii) is the measure rationally connected to that aim; (iii) could a less intrusive measure have been used; and (iv) bearing in mind the severity of the consequences, the importance of the aim and the extent to which the measure will contribute to that aim, has a fair balance been struck between the rights of the individual and the interests of the community?[34]

In 1996, South Africa effectively adopted Canada’s general limitations clause in s 36 of its new bill of rights.[35] However, importantly for the future development of human rights in Australia, s 36 also appended a list of relevant factors in an attempt to give more content to the proportionality test required.[36] From early on in its jurisprudence, the Constitutional Court of South Africa has held that the listed factors must feed into a ‘global judgment on proportionality’ and that the court should not use them ‘mechanically’ as a ‘sequential check-list’.[37] This ‘flexible approach’[38] now stands apart from the more structured approach found elsewhere, including in Canada,[39] the United Kingdom[40] and New Zealand.[41]

According to the more structured approach, all factors must be considered and in their proper order. As to the order of the analysis, Aharon Barak points out:

The structured nature of the discretion involved in proportionality has many advantages. For one, it allows the person conducting proportionality analyses to think analytically, not to skip over things which should be considered, and to consider them in their time and place.[42]

Each step has a ‘time and place’ in the analysis, because each builds on the last. As Ariel Bendor and Tal Sela explain:

[T]he sub-tests of proportionality are structured like a legal version of the Russian matryoshka doll, with each stage of the constitutional analysis containing all the preceding ones. The necessity test [in s 13(2)(d)] encompasses the rationality test [in s 13(2)(c)]. In any factual situation in which there is no rational connection between the purpose that the legislative authority is seeking to achieve and the infringement on the constitutional right, it is possible to accomplish the same purpose by means that infringe on the right to a lesser extent. Similarly, the proportionality sensu stricto sub-test [in ss 13(2)(e)–(g)] includes both the necessity and rational connection requirements. When a statute infringes on a right without rational connection between the infringement and the purpose of the statute, or when there is an alternative to the statute in question, which is likely to accomplish its purpose at a lower cost to rights, the harm caused by that statute clearly exceeds its benefits compared with its various alternatives.[43]

Because each factor builds upon the last, logically, if a measure falls down at one stage of the analysis, it cannot be saved by any of the factors that come afterwards.[44] The fatal flaw will follow the measure through the remainder of the analysis. As Edelman J said recently, ‘[d]uplication is avoided because if a case fails at one stage it is unnecessary to consider whether subsequent stages are satisfied’.[45]

Drafters in Victoria and the ACT followed the format of the South African general limitations clause, in 2006 and 2008 respectively.[46] The general limitations clauses in Victoria and the ACT each contain a list of relevant factors almost identical to those in South Africa. At the time, commentators noted that, at least in terms of drafting style, the

Canadian model of a general limitations clause, embellished by the South African proportionality inquiry, has become the dominant model among advanced constitutional democracies.[47]

The form of s 13 of the Human Rights Act follows in that tradition. Subsection (2) provides:

In deciding whether a limit on a human right is reasonable and justifiable as mentioned in subsection (1), the following factors may be relevant —

(a) the nature of the human right;

(b) the nature of the purpose of the limitation, including whether it is consistent with a free and democratic society based on human dignity, equality and freedom;

(c) the relationship between the limitation and its purpose, including whether the limitation helps to achieve the purpose;

(d) whether there are any less restrictive and reasonably available ways to achieve the purpose;

(e) the importance of the purpose of the limitation;

(f) the importance of preserving the human right, taking into account the nature and extent of the limitation on the human right;

(g) the balance between the matters mentioned in paragraphs (e) and (f).

The first point of departure is that s 13(2) states that the enumerated factors ‘may be relevant’. In the ACT, Victoria and South Africa, the general limitations clauses state that ‘all relevant factors’ are to be taken into account ‘including’ those that are enumerated.[48] We will return to the relevance of ‘may’ later in the article in Part V. More importantly, for present purposes, the factors in s 13(2) are not identical to those enumerated in the ACT, Victoria and South Africa. This recently led one commentator to observe that ‘[s]ome of these modifications are curious and the intention underlying them is unclear’.[49] It is worth interrogating what is different in order to explore the possible reasons for the modifications.

Paragraph (a) — ‘the nature of the human right’ — is common to each of the general limitations clauses in South Africa, Victoria, the ACT and Queensland.[50] It is said to be important to begin the analysis by considering the nature of the human right, because doing so focuses attention on what it is that is being limited, and informs the level of justification required as one moves through the later steps in the justification analysis.[51] Logically, consideration of the nature of the human right precedes, and stands outside of, the four steps of structured proportionality.

Paragraph (b) — ‘the nature of the purpose of the limitation’ — has no counterpart in South Africa, Victoria or the ACT. It asks whether the nature of the purpose is ‘consistent with a free and democratic society based on human dignity, equality and freedom’ — that is, whether the measure has a proper purpose or a legitimate aim.[52] Take, for example, the purpose of the prohibited donor provisions considered in AIP. Justice Applegarth identified their purpose as being the prevention of ‘corruption and undue influence in the government of the State’.[53] Clearly that purpose is consistent with the values of our society and therefore capable of justifying a limit on human rights.[54]

The identification of a proper purpose is the first step of structured proportionality and is conspicuously absent from the list of factors elsewhere. Arguably, the ‘importance of the purpose of the limitation’ or the ‘nature and extent of the limitation’ included in the factors elsewhere fulfil this function.[55] However, as Barak points out, the threshold enquiry is not how important is the measure’s purpose or how deep is the incursion into the human right, but simply whether the purpose is one that is capable of justifying a limit on human rights.[56] Some objectives are never capable of justifying limits on human rights, such as setting out to discriminate or to limit a human right as an end in itself. Questions of weight (that is ‘importance’ and ‘extent’) are better left to the final balancing exercise.[57]

Paragraph (c) — ‘the relationship between the limitation and its purpose’ — also exists in South Africa, Victoria and the ACT.[58] The relationship between a measure and its purpose is relevant because, if the measure does not actually help to achieve the proper purpose, then logically that purpose cannot be used as the reason why a human right is being limited.[59] In structured proportionality, this enquiry is called ‘suitability’ or ‘rational connection’.[60] A measure will be suitable or rationally connected to its purpose if the measure ‘furthers’ or ‘helps to achieve’ the proper purpose in some way,[61] or goes some way towards ‘reali[sing]’ or ‘advanc[ing]’ the proper purpose.[62] In this connection, s 13(2)(c) of the Human Rights Act adds ‘including whether the limitation helps to achieve the purpose’ to provide more guidance on how to assess suitability. To return to AIP, Applegarth J noted that preventing property developers from giving political donations would help to achieve the anti-corruption purpose.[63] The ban was not like the one considered in Unions NSW v New South Wales, which prevented all corporations from giving political donations.[64] The fact that an entity is a corporation says nothing about whether it presents a risk of corruption (and therefore preventing all corporations from giving political donations is not rationally connected to the anti-corruption purpose). By contrast, the fact that an entity is a property developer does say something about the risk of corruption, as experience in New South Wales and Queensland shows.[65] It follows that banning political donations from property developers is rationally connected to preventing corruption and undue influence.

Paragraph (d) — ‘whether there are any less restrictive and reasonably available ways to achieve the purpose’ — also exists in South Africa, Victoria and the ACT.[66] In structured proportionality, this stage is known as necessity testing.[67] It involves looking at alternative ways of achieving the proper purpose. Is there another way of achieving the proper purpose just as effectively, but in a way that limits human rights to a lesser degree? If such an alternative exists, then it cannot be said that the means selected are necessary. The same result can be obtained without harming human rights (or by harming them to a lesser extent).

As an example of necessity testing, the applicant in AIP put forward alternative interpretations of the prohibited donor provisions that would have allowed ‘think tanks’ like the Australian Institute for Progress to spend funds from property developers in an election because they themselves are not property developers.[68] Those alternative readings would impose a lesser burden on freedom of expression, but as Applegarth J pointed out, they would also not be as effective in preventing corruption.[69] In fact, such an interpretation would allow property developers to entirely thwart the ban by funneling money through third parties. It followed that there was no less restrictive alternative which would achieve the purpose.[70]

Paragraphs (e)–(g) go to whether the measure strikes a fair balance between the human right which has been impacted and the countervailing societal interests.[71] At this point,

[i]t may be questioned how a law which has passed the rigours of the first three stages of the proportionality analysis — pressing goal [in s 13(2)(b)], rational connection [in s 13(2)(c)], and minimum impairment [in s 13(2)(d)] — could fail at the final inquiry of proportionality of effects.[72]

The answer, according to the Supreme Court of Canada, ‘lies in the fact that the first three stages ... are anchored in an assessment of the law’s purpose’.[73] Only the fourth stage takes full account of the importance of the human right at stake and whether it outweighs the importance of the law’s purpose.

While the lists of factors in South Africa, Victoria and the ACT direct attention to the importance of the proper purpose,[74] that is only one side of the scales. The other side of the scales — ‘the importance of preserving the human right’ — is entirely absent in South Africa, Victoria and the ACT, as is any direction to engage in any kind of balancing between the two sides of the scales.[75] The failure to explicitly include the final balancing test would explain why the lists of factors in their general limitations clauses are inclusive (‘all relevant factors, including’), whereas the list of factors in Queensland appears to be exhaustive, or at least less explicitly inclusive (‘the following factors may be relevant’). There is no need to be explicitly inclusive if all the steps of structured proportionality have been included.

There is Victorian and ACT authority that balancing inheres in the general limitations clause.[76] However, in the absence of any explicit mention of fair balance, courts in those jurisdictions have been known to work through the listed factors and — having exhausted them — come to a conclusion that the limit is justified.[77] That means that limits are being justified without courts ever consciously addressing how important the human right is or whether the human rights are outweighed by countervailing societal interests. The final three paragraphs in s 13(2) of the Human Rights Act reinstate balancing as the crux of the test of justification. That this is where most of the work of justification is done[78] can be seen in the weighing analysis carried out by Applegarth J in AIP. His Honour grappled with the competing importance of free speech (especially in the context of election campaigns) as against the importance of reducing corruption to enhance our system of democracy.[79] Ultimately, his Honour found that the balance tipped in favour of preventing corruption.[80]

It can be seen that s 13 of the Human Rights Act follows in the tradition of general limitations clauses in Canada, South Africa, Victoria and the ACT. However, the list of relevant factors in s 13(2) diverges in important ways from the South African format adopted in the ACT and Victoria. Those differences reflect a closer alignment with the structured approach to proportionality. The ‘factors’ are more definitive, the balancing is more explicit, and overall, s 13(2) more clearly suggests the path of reasoning to follow.

The text, history and underlying logic of s 13 lead to the conclusion that it imports a structured proportionality analysis. Accordingly, as a general rule, a limit on a human right will only be justified under s 13 if: the limit has a proper purpose; there is a rational connection between the limit and that purpose; the limit is necessary to achieve its purpose, in the sense that the purpose cannot be achieved without limiting the human right or by limiting it to a lesser extent; and, the limit strikes a fair balance between the need to achieve its proper purpose and the human right at stake.[81]

IV The High Court on Why Structured Proportionality Is Required

Not only does structured proportionality arise from the text of s 13, it arises more fundamentally from the relative nature of human rights. Because human rights are generally not absolute, there will always be a need to determine when countervailing interests will trump them.[82] Structured proportionality offers the only viable way of doing that. So much emerges from the intellectual labours of the High Court over the last seven years in the context of the implied freedom of political communication, and more recently in the context of the freedom of interstate trade, commerce and intercourse under s 92 of the Constitution. Because the constitutional freedoms are not absolute, the High Court has also had to grapple with the question of when countervailing societal interests will prevail over each freedom. In 2015, in the case of McCloy v New South Wales (‘McCloy’), a narrow majority of the High Court decided that structured proportionality is the best way to determine whether a burden on the implied freedom of political communication is justified.[83] More recently, in the case of Palmer v Western Australia (‘Palmer’), another narrow majority of the High Court applied structured proportionality in the context of the freedoms guaranteed by s 92 of the Constitution.[84] In a series of cases between McCloy and Palmer, the Commonwealth and a number of states ventured to submit that structured proportionality is not the correct test, or that not all of its steps will be required in all cases.[85] Justice Gageler and Gordon J have also consistently disclaimed the utility of structured proportionality,[86] leading to the kind of ‘abstracted debate about methodology’ one might expect to see in ‘the pages of a law review’.[87] Given that the majority in McCloy borrowed structured proportionality from a human rights context,[88] the majority’s response to these criticisms of structured proportionality may in turn ‘inform the development of the proportionality test’ under s 13 of the Human Rights Act.[89] Further, given that the High Court worked from first principles, without the guidance of a general limitations clause like s 13, the Court’s reasoning says something more fundamental about what is involved in any justification analysis involving rights or freedoms that are relative in nature.[90]

In Brown v Tasmania (‘Brown’) — one of the first implied freedom cases to follow McCloy — Tasmania and Queensland both submitted that the correct test is whether the burden on political communication is ‘reasonably appropriate and adapted’.[91] While acknowledging that this criterion[92] may involve applying some or all of the elements of structured proportionality, Queensland argued that it should be applied in an unstructured way, ‘in the nature of a synthesis of all relevant considerations as to justification’.[93] In that sense, the test proposed by Queensland was indistinguishable from the flexible approach applied in South Africa, which calls for a ‘global judgment’.[94] Chief Justice Kiefel, Bell and Keane JJ held that ‘[s]uch an approach would invite little more from the Court than an impression’ and would ‘not address the need for transparency in reasoning’.[95] Whereas a global judgment is ‘pronounced as a conclusion, absent reasoning’,[96] structured proportionality ‘help[s] control intuitive assessments, [and] make[s] value judgments explicit’.[97] A ‘more clearly structured approach’ also accords with ‘the analytical approach to legal reasoning characteristic of the common law’.[98]

More recently, in Palmer, Gageler J defended a less structured approach on the basis that the evaluative judgment would then be more sensitive to the constitutional value at stake. According to his Honour, the rigid formula of structured proportionality results in some factors receiving ‘unwarranted analytical prominence’.[99] Conversely, factors that do not readily fit in any of the steps ‘get ignored or suppressed or downplayed’, or worse still, ‘pushed down to be swept up in the residual inquiry into “adequacy of balance”’.[100] Justice Gageler’s complaint appears to be that structured proportionality lulls judges into the false comfort that they are avoiding intuitive assessments, which only serves to downplay what is at stake in the value judgment that remains in the final balancing exercise.[101] His Honour’s real concern appears to be that insufficient weight is accorded to the value of the constitutional freedom in the final balancing exercise. That might be said to be a complaint about how structured proportionality is applied, not with whether it should be applied. Translated to a human rights context, some of his Honour’s concerns may be addressed by the requirement in s 13(2)(a) of the Human Rights Act to identify at the outset the nature and underlying values of the human right at stake. If done well, carrying out the first step in s 13(2)(a) may help to ensure that the entirety of the justification analysis ‘cleaves to the reasons’ for the human right.[102] In any event, Gageler J’s defence of a less structured approach stands at odds with the prevailing view of a majority of the High Court since McCloy: that impression and intuition are to be avoided as much as possible when justifying burdens on constitutional freedoms.

The Commonwealth took a slightly different tack in Brown. It submitted that not all of the sub-tests of structured proportionality would be required in all cases:

In summary [the Commonwealth] accept[ed] that the first enquiry, that of ‘suitability’, which is as to the connection of a measure adopted by a statute to its purpose, is relevant to all cases involving the freedom. The second test, that of ‘necessity’, should be taken as sometimes, but not always, decisive. The last assessment, that of strict proportionality or ‘balancing’, should only be undertaken where the burden on the freedom is ‘direct and substantial’, it submit[ted].[103]

Two years later, in Clubb v Edwards (‘Clubb’), Victoria joined the Commonwealth in submitting that ‘it is not necessary in [all] case[s] to undertake all of the proportionality testing’.[104] How much proportionality testing is required would depend on which category the burden on political communication fell into. Burdens which merely regulate the time, manner and place of political communication would need to pass only some of the hoops of structured proportionality, whereas burdens which are discriminatory or directed to the content of speech may need to pass all four hurdles.[105] The proposals put forward by the Commonwealth and Victoria were versions of what is known as a categorisation approach (or what Gageler J terms a ‘calibrated’[106] approach). Categorisation involves developing different tests with different levels of scrutiny, depending on the right engaged and the circumstances in which it is engaged. Categorisation is widely seen as the most promising alternative to structured proportionality.[107] Even Barak, who is perhaps the world’s most influential proponent of structured proportionality,[108] recognises that ‘[t]he most important alternative offered to proportionality, by far, is that of categorization’.[109] The advantage of categorisation is that the test will be tailored to the concrete scenario of the particular category, and therefore less abstract than applying structured proportionality from scratch.[110]

The paradigm example of categorisation — though by no means the only example — is provided by US constitutional law. Broadly, limits on rights under the United States Constitution are subject to strict scrutiny, intermediate scrutiny or minimal scrutiny. At one end of the spectrum, to be justified, certain kinds of limits on fundamental rights such as freedom of political speech must be necessary and ‘narrowly tailored’ to achieve a compelling governmental interest.[111] At the other end of the spectrum, a limit on equality which is not based on suspect or quasi-suspect classifications need only be rationally connected to a legitimate aim.[112] Although the US has had a few centuries’ head start on incrementally building up their categories, other human rights systems have also begun to develop more specific tests for particular rights or burdens. For example, the South African Constitutional Court has developed the Grootboom test[113] — akin to a strong form of the Wednesbury unreasonableness standard[114] — specifically for socio-economic rights.[115] As another example, in R v Jordan (‘Jordan’), the Supreme Court of Canada recently introduced a categorisation approach to the right to a speedy trial.[116] Following Jordan, if a trial is delayed by 18 months in a provincial court or 30 months in a superior court, the delay will be ‘presumptively unreasonable’ and only capable of being justified if the Crown can establish ‘exceptional circumstances’.[117] Under the Jordan test, once a delay is appropriately categorised, the outcome of the justification analysis is close to automatic. Some suggest that categorisation may even be inevitable in any sufficiently mature legal system.[118] As courts apply structured proportionality and precedents accumulate, categories tend to emerge naturally over time.

Among the judges of the High Court of Australia, only Gageler J has taken up the invitation to apply a categorisation approach to the implied freedom of political communication.[119] A majority continues to eschew categorisation in favour of structured proportionality. Their Honours’ reasons for doing so reflect the two schools of thought in the literature. At times the majority has stated that structured proportionality is no more than a useful ‘tool of analysis’,[120] and that conceivably a different approach might be applied. In Palmer, Kiefel CJ and Keane J frankly admitted:

It cannot be suggested that structured proportionality is a perfect method. None is, but some method is necessary if lawyers and legislators are to know how the question of justification is to be approached in a given case. Structured proportionality certainly seems preferable to its main competitors.[121]

The choice of the majority to apply structured proportionality therefore echoes Barak’s pragmatic view that ‘of all the options available to ensure human rights in a pluralistic, democratic society, proportionality is the best available option’.[122]

On the other hand, the majority has also treated structured proportionality as logically required. This reflects the view that justification analysis necessarily involves an assessment of the ‘rationality and reasonableness’ of the limit on the right or freedom at stake.[123] Influential theorists such as Robert Alexy and David Beatty posit that structured proportionality is uniquely placed to assess the rationality and reasonableness of limits on rights and freedoms which are relative. Their point about rationality goes beyond the ‘rational connection’ component of structured proportionality (in s 13(2)(c)). Their point, as Alexy puts it, is that ‘there is no other rational way in which the reason for the limitation can be put in relation to the constitutional right’.[124] According to Beatty, justification analysis without structured proportionality is ‘a logical impossibility’.[125]

Following that school of thought, a majority of the High Court has embraced the view that structured proportionality ‘direct[s] attention to different aspects of what is implied in any rational assessment of the reasonableness of a restriction’.[126] Because each step of structured proportionality tests the rationality of the measure in a slightly different way, skipping a step risks missing an error in reasoning.[127] A step in the analysis can be assumed if the measure obviously falls down later in the analysis.[128] But otherwise, each step in the analysis is logically indispensable to figuring out whether a limit is justified. As to whether you can skip necessity testing, Kiefel CJ, Bell and Keane JJ pointed out in Clubb: ‘It could hardly be said that a measure which is more restrictive of the freedom than is necessary can rationally justify the burden.’[129] If the legislature can achieve its purpose without burdening free political communication, its choice to go out of its way to do so would be ‘inexplicable’ and not ‘justified’.[130] Thus, necessity testing is logically indispensable. As to whether you can skip adequacy of balance, the plurality in Clubb noted that a failure to consider adequacy of balance necessarily involves a failure to consider whether a measure is ‘a rational response to the mischief at which it is directed’.[131] In Palmer, Kiefel CJ and Keane J said that balancing is one of the ‘rational means’ by which burdens on freedoms are justified.[132] That is, if a court skips the final balancing stage, an irrational response may be inadvertently justified. Ultimately, for a majority of the High Court, a categorisation approach should not be applied because it skips questions which must be answered if one is to conclude that the limitation is justified. As Nettle J said in Clubb, a categorisation approach would risk substituting ‘principles of analysis capable of general application’ for categories ‘which in some contexts may but in others should not lead to the conclusion’ that the limit is justified.[133]

Perhaps the strongest argument for applying structured proportionality under s 13 of the Human Rights Act comes from its fiercest critics in the High Court. One of the key reasons why Justices Gageler and Gordon have rejected structured proportionality for constitutional freedoms is that constitutional freedoms are not like human rights. Constitutional freedoms are generally structural limits ‘addressed to legislative power, not rights’.[134] The focus is on what the constitutional system requires, not on optimising an individual’s rights as far as possible. For example, when it comes to the implied freedom of political communication, the freedom is confined to ‘what is necessary for the effective operation of that system of representative and responsible government provided for by the Constitution’.[135] It does not seek to maximise freedom of expression beyond what is necessary for the system.

By contrast, because human rights are intrinsically valuable to the individual, they are generally seen as optimisation requirements. This means that they must be realised ‘to the greatest extent possible given the legal and factual possibilities’.[136] As Gordon J said in Clubb, by reference to the work of Alexy:

[T]he three sub-principles of proportionality express this idea of optimisation — the first and second, suitability and necessity, refer to optimisation relative to the factual possibilities and the third, the law of balancing, concerns optimisation of the legal possibilities.[137]

To engage in an analysis of that kind

might be all well and good in a constitutional system in which a function of the judiciary is understood to be the enhancement of political outcomes in order to achieve some notion of Pareto-optimality.[138]

But, as Gageler J pointed out in Murphy v Electoral Commissioner, ‘[t]hat is not our system’, at least not when it comes to constitutional freedoms.[139] Justices Gageler and Gordon may well have a point. But even if they do, their reasons for rejecting structured proportionality in the context of constitutional freedoms are also powerful reasons for accepting structured proportionality in the context of human rights.

Ultimately, the lessons learned over the last seven years in the context of constitutional freedoms tell us something about the analysis required by s 13 of the Human Rights Act. Constitutionally entrenched freedoms and unentrenched human rights under the Human Rights Act are both relative in nature.[140] Section 13(1) calls for any limit on a human right to be ‘demonstrably justified’. Structured proportionality offers the only viable way of ‘ascertaining the rationality and reasonableness’ of a limit on a right or freedom which is relative in nature, and therefore whether the limit is justified.[141] Structured proportionality also offers the only way to optimise human rights — that is, to realise the full potential of human rights by protecting them to the greatest extent possible given the factual and legal possibilities. Although, on one reading, the word ‘may’ in s 13(2) leaves open the possibility of adopting a flexible South African approach, such an approach would reduce justification analysis to a matter of impression and allow courts to conceal their process of reasoning. Justification under s 13 should not be approached as a matter of intuition. While categorisation is not unknown in human rights systems around the world, the creation of any category requires a leap of faith that the category itself will lead to accurate conclusions about whether the limit is justified. By contrast, structured proportionality is designed to always answer that question accurately. Moreover, as a majority of the High Court has pointed out, the factors of structured proportionality remain logically indispensable to a conclusion that the limit is justified, even if an attempt is made to speed up the analysis by using categories. In Part V, this article will consider the possibility that categorisation may have a future role to play under s 13 of the Human Rights Act. For now, the discussion in the High Court over the last seven years tells us that any category that might be adopted would still need to be tested against structured proportionality. Ultimately, just as all of the factors of structured proportionality are logically indispensable in the context of the implied freedom, they are logically indispensable in the context of human rights.

V The Exceptions That Prove the Rule: When the s 13(2) Factors Are Not Mandatory

So far, this article has not confronted the word ‘may’. Section 13(2) introduces the factors by stating that they ‘may be relevant’. According to the logic of structured proportionality, the factors ‘are’ — not ‘may be’ — relevant. Yet, as a matter of statutory construction, the word ‘may’ in the chapeau of s 13(2) should be given work to do.[142] This article contends that exceptions to the general rule (that all factors must be considered) can be accommodated by the structure of s 13, and the word ‘may’ can be given work to do, provided two criteria are met.

First, the exception must not be arbitrary. Picking and choosing which factors to apply would offend the right to equality before the law in s 15(3) of the Human Rights Act. That right guards against arbitrary application of the law.[143] Equality before the law would clearly be undermined if the court were to apply the final balancing stage of structured proportionality in one case (potentially resulting in a finding of incompatibility with human rights), but inexplicably decide not to apply the final balancing stage in the next case (potentially resulting in a finding of compatibility). One only needs to look at the many cases decided at the necessity stage or the final balancing stage to see that the result would have been different had the court declined to consider all the factors. Consistent application of the factors in s 13(2) ‘reflects the certainty to which the law aspires’.[144] Public entities and ‘members of the legislative branch want to know, should know, and are entitled to know’,[145] the standard by which their acts, decisions and legislation will be judged. They cannot know in advance whether they will be acting compatibly with human rights if the answer to that question turns on the court’s random or arbitrary selection of factors in s 13(2).[146] Accordingly, at the very least, there must be a principled reason for any departure from the general rule that all the factors in s 13(2) are relevant.

Secondly, the exception must itself be proportionate. Barak recognises a space for exceptions to the general rule in the form of what he calls ‘principled balancing formulas’.[147] Rather than engage in ad hoc proportionality for every limit on human rights, principled balancing formulas provide a more concrete test for determining compatibility in a particular category of case. Importantly, for Barak, principled balancing formulas must themselves satisfy the elements of structured proportionality. In Barak’s view, this means that every application of the formula would pass structured proportionality, were one to stop and engage in the analysis. In this way, the exception to the general rule will itself ensure compliance with the general rule.

This article contends that there are a number of potential exceptions to the general rule which would meet these two criteria and give the word ‘may’ work to do.

A Internal Limitations

The most obvious way that the test of justification set out in s 13(2) may be altered is where the text of a human right sets out a particular way of justifying a limit on the right. Components of human rights that do this are called ‘internal limitations’ or ‘specific limitations’.[148] A clear example is s 30(2) of the Human Rights Act, which provides that unconvicted detainees have a right to be segregated from convicted detainees, ‘unless reasonably necessary’.[149] That is, the provision itself sets out particular criteria for determining whether a limitation on the right can be justified.

Internal limitations exist because most of the rights protected by the Human Rights Act are drawn from the International Covenant on Civil and Political Rights (‘ICCPR’),[150] which does not have a general limitations clause like s 13 of the Human Rights Act.[151] Accordingly, the drafters of the ICCPR needed to set out the content of the right as well as the way that a limit on the right could be justified. It may have been ‘inapt’ for the Human Rights Act to combine the internal limitations of the rights derived from the ICCPR with a general limitations clause.[152] This overlay effectively results in a duplication of limitations clauses. Be that as it may, the ‘hybrid model’ adopted by the Human Rights Act must be made to work.[153]

In Re Kracke and Mental Health Review Board, Bell J set out one way of reconciling internal limitations with the general limitations clause in the Victorian Charter. His Honour said:

Some rights are expressed in terms that contain specific limitations. As relevant in the present case, the possibility of imposing lawful and non arbitrary limitations on the right to privacy in s 13(a) of the [Victorian] Charter is one example. The right to be free of arbitrary detention in s 21(2) is another. Where rights are expressed in terms that contain a specific limitation, the nature and content of the rights in their plain state are not seen to be reduced by the specific limitation. Rather, the specific limitation is seen as an indication of what might be considered in determining whether any limitations are reasonable and justified under the general limitations provision in [the equivalent of s 13 of the Human Rights Act].
Thus, when identifying the scope of the right at the engagement stage, this is done broadly and purposively, even where the right contains a specific limitation. Such a limitation becomes subsumed in the overall justification analysis which is undertaken in the next stage. That is why the international jurisprudence shows there is very considerable interplay between the application of specific limitations provisions on the one hand and general limitations provisions on the other.[154]

If the internal limitation in the segregation right in s 30(2) of the Human Rights Act is ‘subsumed in the overall justification analysis’ in s 13, then not all of the factors in s 13(2) will apply. In the logic of structured proportionality, non-segregation between convicted and unconvicted detainees will be ‘necessary’ if there is no less restrictive and reasonably available way to achieve a proper purpose. That is, the internal limitation means that the analysis stops at s 13(2)(d), without needing to go on to consider fair balance under ss 13(2)(e)–(g).[155]

The principled basis for departing from the general rule lies in the text of the human right itself. The internal limitation also reflects a legislative judgment that applying the more specific test will produce a proportionate outcome. For example, the internal limitation of ‘reasonable necessity’ for the right to segregation under s 30(2) reflects a judgment call that any failure to segregate convicted and unconvicted detainees will always strike a fair balance (ss 13(2)(e)–(g)), in circumstances where the limit has a proper purpose, is rationally connected to that purpose, and goes no further than necessary to achieve the purpose (ss 13(2)(b)–(d)). Because fair balance will always be satisfied in those circumstances, there is no need to progress beyond necessity. In this way, internal limitations offer a principled and proportionate exception to the general rule that all of the factors in s 13(2) are relevant, thereby giving ‘may’ in the chapeau of s 13(2) work to do.

B Absolute Rights

The word ‘may’ would also be given work to do if some of the human rights protected by the Human Rights Act were absolute. Although the extrinsic material to the Human Rights Act states that human rights are ‘not absolute’,[156] international law recognises that some rights are absolute.[157] Limits on absolute rights are incapable of justification.[158] International law also recognises a related category of rights known as non-derogable rights.[159] Often these rights are also absolute, though not always.[160] Non-derogable rights are rights of such ‘great importance’ that their operation ‘may not be suspended, even in times of national emergency’.[161] The equivalent at the domestic level would be a category of rights that are so important that they may not be subject to an override declaration under s 43 of the Human Rights Act. For example, the freedom against torture in art 7 of the ICCPR is both absolute and non-derogable.[162] Section 17 of the Human Rights Act protects the same freedom. At international law, no matter how dire circumstances may be, the state can never justify torture.

International law also recognises that some rights have an absolute component. For example, the Committee on Economic, Social and Cultural Rights — the treaty monitoring body of the International Covenant on Economic, Social and Cultural Rights (‘ICESCR’)[163] — has long taken the view that socio-economic rights contain ‘minimum core obligations’.[164] These represent

a universal absolute bottom-line of obligations under each of the rights of the ICESCR, which has to be respected, promoted and fulfilled by all state parties to the ICESCR, regardless of their level of economic development.[165]

For example, in order to fulfil the right to education in art 13 of the ICESCR, the state must provide universal primary education as a bare minimum.[166] The concept of minimum core obligations may have particular significance for the socio-economic rights enshrined in the Human Rights Act, namely the rights to education and health in ss 36 and 37.[167]

The status of a human right as absolute or non-derogable may be taken into account when considering the ‘nature of the human right’ under s 13(2)(a). That may translate into requiring a very weighty proper purpose to outweigh the importance of the human right.[168] Because all human rights are subject to s 13, even a limit on an absolute or non-derogable right at international law may, in theory, be capable of justification under the Human Rights Act. That has been the approach of the Supreme Court of Canada. For example, at international law, judicial independence (as an aspect of the right to a fair hearing) ‘is an absolute right that is not subject to any exception’.[169] Yet, at the domestic level, the Supreme Court of Canada has allowed for the possibility that the state may deviate from judicial independence.[170] Nonetheless, any exception to judicial independence calls for ‘a high threshold of justification under [the Canadian equivalent of s 13 of the Human Rights Act]’,[171] and cannot be justified save where there are ‘dire and exceptional financial emergencies caused by extraordinary circumstances such as the outbreak of war or imminent bankruptcy’.[172]

Alternatively, that the human right is absolute or non-derogable at international law may be taken into account by treating the right as absolute under the Human Rights Act as well, thus recognising that no purpose can justify limiting the human right, regardless of whether it is proper, rationally connected, necessary or strikes a fair balance. If this approach were taken, no factor in s 13(2) would be relevant after the first. This approach may even be encouraged by s 48(3) of the Human Rights Act which allows courts to refer to ‘[i]nternational law and the judgments of domestic, foreign and international courts and tribunals’ when determining the scope of human rights.[173] However, international authorities will not be relevant under s 48(3) where they lie at odds with the structure of the Human Rights Act.[174] One of the key features of the Act’s structure is that all human rights are subject to s 13.[175] Nonetheless, other human rights systems with a general limitations clause recognise the existence of absolute rights. For example, the Supreme Court of New Zealand has held that ‘the right to a fair trial, affirmed by s 25(a) [of the New Zealand Bill of Rights Act 1990 (NZ)], is an absolute right’.[176] The word ‘may’ in s 13(2) accommodates the possibility that Queensland courts may likewise recognise that some human rights are absolute under the Human Rights Act.

Calibrating the test for limits on rights that are absolute or non-derogable at international law — whether by treating them as absolute or by requiring a weightier proper purpose — also involves the application of a principled balancing formula. To recognise that a human right is absolute is to make a judgment call that the human right is so important that no countervailing objective can ever outweigh it.[177] Because the limit will necessarily fall down at the balancing stage, there is no need to work through proper purpose, suitability and necessity in vain. The conclusion that the right is absolute is simply a shorthand way of expressing that proportionality analysis. If Queensland courts adopt this shorthand and find that some human rights under the Human Rights Act are absolute, the word ‘may’ will have work to do.

C Developing a Categorisation Approach

The word ‘may’ in s 13(2) might also accommodate the development of a categorisation approach. As outlined in Part IV, artificial categories should be avoided because they risk skipping questions that must be answered if one is to conclude that the limit on human rights is justified. However, categories may emerge organically over time. If they are drawn appropriately, categories can help to shortcut the analysis while still arriving at proportionate outcomes. In the event that Queensland courts one day go down the path of categorisation, it is worth thinking about how that would work, and how the word ‘may’ in s 13(2) would fit with a categorisation approach.

The various tests that apply under a categorisation approach — such as strict scrutiny, intermediate scrutiny and minimal scrutiny in US constitutional law — are also ‘principled balancing formulas’.[178] According to Barak, categorisation is permissible if the tests which demand less than full structured proportionality still produce proportionate outcomes.[179] Categorisation might be introduced to Queensland by transplanting categories that have already been developed in other human rights systems. It is unlikely that any wholesale importation of the US categories would ensure proportionate outcomes in Queensland, given the sheer difference between our rights and contexts.[180] For example, a discriminatory burden on free speech may avoid strict scrutiny in the US because the law is facially neutral, but in Queensland, the discriminatory effect would demand greater justification under a proportionality analysis.[181]

However, the US is not the only human rights system that has developed categories. Given that Queensland has enshrined the socio-economic rights to education and health in s 36 and s 37 of the Human Rights Act, Queensland courts might borrow the Grootboom test from South Africa to take account of the duty to progressively realise socio-economic rights.[182] Similarly, given that Queensland also protects the right to a speedy trial in s 32(2)(c) of the Human Rights Act, Queensland courts might adapt the rigid Jordan test from Canada, so that a delay beyond a certain ceiling becomes ‘presumptively unreasonable’.[183] Again, s 48(3) of the Human Rights Act encourages recourse to these international authorities. Whether Queensland courts should borrow categories from other jurisdictions will turn on whether the categories would produce proportionate outcomes in the different context of Queensland.

Queensland courts may also develop their own categorisation approach over time, as a large body of jurisprudence builds up and patterns can be identified. As noted above, the seeds of categorisation are already built into the Human Rights Act, with internal limitations and the possibility of absolute rights. The similarities between Queensland’s Human Rights Act and the equivalent legislation in the ACT and Victoria may also speed up the process of accumulating a body of case law.[184] Again, whether Queensland courts should develop their own categories will turn on whether those categories would produce a proportionate outcome.

While this course is open, the development of categories is not inevitable. The complexity of the US system of categorisation is unique among human rights jurisdictions. Most other human rights jurisdictions apply a form of structured proportionality,[185] even if they also apply a categorisation approach for a narrow range of rights. The impulse to treat all human rights as equally deserving of respect and protection also tells against categories based on hierarchies of rights.[186] Moreover, as the High Court has pointed out in the context of the implied freedom,[187] you do not know the answers to the questions you do not ask. If you apply a categorisation test that assumes the limit is necessary or strikes a fair balance, you will not know whether the assumption built into the test is correct. In any given case, the only way to know whether the assumption is correct would be to engage in full structured proportionality. While a proportionate categorisation approach is theoretically possible, the High Court’s reasoning in the context of the implied freedom suggests we should be reluctant to conclude that a particular category will always produce a proportionate outcome.

Nonetheless, the word ‘may’ in s 13(2) provides room for the courts to develop a categorisation approach to justification under the Human Rights Act, should the case law one day move in that direction. Conversely, a categorisation approach would give the word ‘may’ work to do. Importantly, any categorisation approach that may be adopted will itself need to be proportionate, and therefore cohere with the structured proportionality logic underlying s 13.

D Procedural Exceptions to the General Rule

The final way that the word ‘may’ in s 13(2) can be given work to do is by recognising that the procedural obligations under the Human Rights Act do not always demand a rigorous and structured approach to justification. The culture of justification introduced by the Human Rights Act has both a substantive and a procedural dimension. In considering exceptions to the general rule (that all factors in s 13(2) must be considered), so far, we have only considered exceptions that relate to substantive compatibility with human rights. Internal limitations, absolute rights and a categorisation approach are all capable of producing outcomes that are substantively compatible with human rights.

However, legislators are not merely required to produce legislation which is substantively compatible with human rights (albeit under threat of a declaration of incompatibility by the Supreme Court of Queensland under s 53, rather than threat of invalidity). Legislators are also required to think about the impact of their proposed legislation on human rights. Under s 38, a member of the Legislative Assembly who introduces a Bill must prepare a statement of compatibility setting out ‘whether, in the member’s opinion, the Bill is compatible with human rights and, if so, how it is compatible’.[188]

Likewise, public entities are not merely required to arrive at acts and decisions that are substantively compatible with human rights under s 58(1)(a). They also have a distinct obligation under s 58(1)(b) to think about the impact of their decisions on human rights by giving ‘proper consideration’ to relevant human rights.[189] As s 58(5) clarifies, the procedural obligation to give ‘proper consideration’ to human rights involves identifying the human rights that may be affected and then considering whether the decision would be ‘compatible’ with those human rights. Remember that ‘compatible with human rights’ is defined in s 8 in a way that brings in the proportionality analysis in s 13. That gives rise to the question of whether legislators and public entities are required to apply structured proportionality in order to discharge their procedural obligations, or whether they can think about limits on human rights in a less rigorous way.

There are two principled bases for departing from the general rule that all factors in s 13(2) are mandatory when it comes to the procedural obligations under the Human Rights Act. First, the procedural obligation in s 58(1)(b) is not concerned with whether the decision is compatible with human rights, assessed using structured proportionality. That is the role played by the substantive obligation, which ‘requires the act or decision to be compatible with human rights’.[190] ‘[W]hat matters [for the substantive obligation] is the practical outcome, not the quality of the decision-making process that led to it’.[191] By contrast, the procedural obligation is concerned with ensuring that a genuine attempt is made to wrestle with human rights questions. It is an additional stick aimed at speeding up culture change. If a public entity does not turn its mind to relevant human rights, its act or decision will be automatically unlawful under s 58, regardless of whether the act or decision may turn out to be substantively compatible with human rights.[192]

Similarly, when it comes to legislation, the role played by the statement of compatibility is to force the responsible member of the Legislative Assembly to confront the human rights impacts of their proposed legislation. When the Supreme Court is considering whether to issue a declaration of incompatibility under s 53, the Court is not bound by the opinion of the member in his or her statement of compatibility under s 38. Thus, whether legislation is compatible with human rights is not definitively answered by the member’s procedural obligation to table a statement of compatibility. Whether it is compatible with human rights is a separate question, which is open for discussion in a dialogue between the courts and Parliament.[193]

The second principled basis for departing from the general rule (that all factors in s 13(2) must be considered) is that the standard of proper consideration required cannot be so onerous that it proves unworkable in the day-to-day reality of government decision-making. Unrealistically high expectations have the potential to thwart attempts to embed a human rights culture across the Queensland public sector.[194] As Emerton J held in Castles v Secretary, Department of Justice, in the context of the Victorian Charter:

The requirement ... to give proper consideration to human rights must be read in the context of the [Victorian] Charter as a whole, and its purposes. The [Victorian] Charter is intended to apply to the plethora of decisions made by public authorities of all kinds. The consideration of human rights is intended to become part of decision-making processes at all levels of government. It is therefore intended to become a ‘common or garden’ activity for persons working in the public sector, both senior and junior. In these circumstances, proper consideration of human rights should not be a sophisticated legal exercise. Proper consideration need not involve formally identifying the ‘correct’ rights or explaining their content by reference to legal principles or jurisprudence. Rather, proper consideration will involve understanding in general terms which of the rights of the person affected by the decision may be relevant and whether, and if so how, those rights will be interfered with by the decision that is made. As part of the exercise of justification, proper consideration will involve balancing competing private and public interests. There is no formula for such an exercise, and it should not be scrutinised over-zealously by the courts.[195]

Justice Emerton’s observation that ‘[t]here is no formula for such an exercise’ is strictly true in the Victorian context. The Victorian Charter does not contain an equivalent of s 58(5) expressly linking the standard of ‘proper consideration’ to proportionality under s 13.[196] While the link to s 13 means there is a formula in Queensland, the need for a workable standard nevertheless means there should not necessarily be a strict formula for giving proper consideration in Queensland.[197]

The logic of proportionality underlying s 13 suggests it is not enough that there is a principled basis for departing from the general rule. The departure from the general rule must itself be proportionate. While it will be permissible for a police officer on the front line to justify limits on human rights in an abbreviated way, more will be required of a Minister making a major policy decision with deep impacts on human rights, such as a decision to imprison children in an adult maximum security prison. The Minister’s consideration of human rights will not be ‘proper’, for example, if they wilfully ignore obvious alternative housing options (in effect skipping the necessity test under s 13(2)(d)).[198] The reason is that, although the procedural and substantive obligations are distinct, they are related.[199] Both are ultimately concerned with the reasonableness or rationality of the measure.[200] ‘The better ... the consideration given to human rights at first instance, the harder it will be to challenge the act or decision concerned’.[201] If legislators and public entities engage in a robust structured proportionality analysis, they are more likely to arrive at an outcome that is substantively compatible with human rights.[202] The further legislators and public entities depart from structured proportionality, the more chances there will be that they will miss a fatal flaw in their reasoning.

However, the exception to the general rule for the procedural obligation will itself be proportionate if the standard of ‘proper consideration’ is calibrated to the circumstances. For example, the more serious the decision, the bigger the impact on human rights, and the more senior the decision-maker, the closer a decision-maker may be required to adhere to structured proportionality.[203] A calibrated approach along these lines can be seen in the Victorian case of Certain Children v Minister for Families and Children [No 2].[204] Justice John Dixon held that the decision by the Minister to house children in an adult maximum security prison demanded a higher standard of ‘proper consideration’.[205] When it comes to statements of compatibility, ordinarily the same calibrating factors would suggest a closer adherence to structured proportionality for principal legislation. Most Bills are proposed by a Minister, and legislation will necessarily apply to large numbers of people. On the other hand, technical amendments that impose very minor limits on human rights may not demand a full structured proportionality analysis.[206]

It follows that the two criteria posited by this article for an exception to the general rule are satisfied when it comes to the procedural obligations under the Human Rights Act. There are good reasons why legislators and public entities should not be forced to work meticulously through the steps of structured proportionality in all cases. Further, the quality of decision-making called for can be calibrated to the circumstances, such that the departure from the general rule will itself be proportionate. The word ‘may’ in s 13(2) offers the flexibility needed to avoid an overzealous approach to the procedural obligations in s 38 and s 58(1)(b). In many if not most circumstances, it will be enough that legislators and public entities have seriously turned their minds to the impact on human rights, even if they do not order their reasoning using structured proportionality. In exceptional circumstances — such as a decision about whether to house children in an adult maximum security prison — it may be that the only way to discharge the procedural obligation is to apply full structured proportionality. This variable standard for the procedural obligations under the Human Rights Act gives the word ‘may’ in s 13(2) work to do.

VI With One Exception, the s 13(2) Factors Are Sufficient

Whether all the factors in s 13(2) are necessary was only one of the questions posed by Applegarth J in AIP. The converse question his Honour raised was whether the factors in s 13(2) are sufficient considerations.[207] Is it possible that a measure which helps to achieve a proper purpose, goes no further than necessary, and strikes a fair balance, is nonetheless for some reason not justified? The Canadian origins of the overall test of justification in s 13(1) show that, with one caveat, the factors in s 13(2) merely unpack what is already implicit in s 13(1). That caveat is the bedrock requirement of legality, which s 13(1) spells out in requiring that the limit be ‘under law’. In a society that respects the rule of law, limits on human rights must have ‘lawful authority’ or a ‘legal foundation’.[208] Beyond that basal requirement of legality, s 13(1) says nothing that s 13(2) does not say in greater detail. In all jurisdictions applying proportionality, a measure that passes through all of the hurdles of legality and structured proportionality will necessarily be a reasonable and justified limit on human rights. In Palmer, Gageler J raised the prospect that there may be extraneous factors that do not neatly fit the sub-tests of structured proportionality, but his Honour did not give any examples.[209] Justice Edelman’s response was to ask simply: ‘But what are these factors and how would they lead to a conclusion of invalidity?’[210] Indeed, in the

several decades since proportionality was first introduced to the Western legal systems [from Germany] no significant changes have occurred in its components or its understanding as a constitutional tool.[211]

While there are entirely different ways of thinking about justification — such as a categorisation approach or treating human rights as absolute — under the rubric of structured proportionality, countless judges, theorists and drafters from across the world have not conceived of any super-added requirement for justification in addition to the factors listed in s 13(2) of the Human Rights Act (with the sole exception of the requirement of legality). Despite the hints from Applegarth J in AIP and from the extrinsic material, s 13(2) is exhaustive. Where a limit is authorised by law, satisfaction of each of the factors in s 13(2) will be sufficient to justify the limit under s 13(1).

VII Conclusion

The factors in s 13(2) of the Human Rights Act align closely with structured proportionality, a tool of analysis used almost universally to work through the question of whether a limit on human rights is justified. The factors in s 13(2) are logical requirements, despite the provision itself stating that the factors only ‘may’ be relevant. As a plurality of the High Court has recently explained in the context of constitutional freedoms, each of the four steps of structured proportionality is logically indispensable.[212] It is irrational to say that a measure is justified even though it pursues an improper purpose (per s 13(2)(b)). It is irrational to say that a measure is justified by reference to a purpose that the measure does not even help to achieve (per s 13(2)(c)). It is irrational to say that a measure is justified even though its purpose could have been achieved as effectively without harming the human right at all, or to a lesser extent (per s 13(2)(d)). Finally, it is irrational to conclude that a measure is justified even though the importance of preserving the human right outweighs the importance of achieving the measure’s purpose (per ss 13(2)(e)–(g)).

Accepting that all of the factors of s 13(2) are relevant, work must nonetheless be found for the word ‘may’ to do. This article proposes four possible exceptions to the general rule which will give meaning and effect to the word ‘may’. First, if an internal limitation in a human right modifies the test of justification, depending on the modification, it is possible that not all of the factors in s 13(2) will be relevant. Secondly, if it is accepted that some of the rights enshrined in the Human Rights Act are absolute, then for those rights, only s 13(2)(a) will be a relevant factor. Thirdly, s 13(2) does not foreclose the possibility that a categorisation approach may emerge, with Queensland courts fashioning more specific tests of justification that apply to specific human rights or specific kinds of limits. Only time (and a significant body of case law) will tell whether justification analysis in Queensland develops in that direction. Finally, the factors in s 13(2) may be optional for the purposes of the procedural obligations of legislators and public entities in s 38 and s 58(1)(b). On that reading, in some circumstances, legislators and public entities can give proper consideration to human rights by seriously turning their minds to the impact that their proposed act, decision or statutory provision will have on human rights, even though they do not engage in a detailed proportionality analysis.

Importantly, each of these exceptions is itself proportionate. Internal limitations, absolute rights and a categorisation approach are permissible under s 13, because they ensure proportionate outcomes. That is, properly applying any of those approaches would produce the same conclusion that would be reached were one instead to take the time to work through all of the factors of structured proportionality set out in s 13(2). Allowing unstructured reasoning for the procedural obligations of legislators and public entities also facilitates proportionate outcomes, provided the standard of proper consideration required is calibrated to the circumstances. A flexible, workable standard for frontline workers is more likely to cultivate a culture of justification across the Queensland public sector. However, the greater the potential impact on human rights, the higher the standard of proper consideration will become. The higher the standard, the closer the process of reasoning will need to accord with full structured proportionality. In this way, each of the exceptions identified in this article proves the general rule that all of the factors in s 13(2) are logically indispensable.

The queries raised by Applegarth J in AIP have clear answers once it is appreciated that s 13 imports structured proportionality as a matter of text, historical background and the underlying logic of justification analysis. Unless one of the exceptions identified in this article applies, the factors in s 13(2) of the Human Rights Act are both necessary and sufficient to conclude that a limit on a human right is justified.


* Senior Principal Lawyer, Crown Law, Queensland. The views expressed in this article are my own and not those of Crown Law. My thanks are due to the Crown Law librarians for their assistance, and to Kevin Bell, Tamara Walsh, Kylie Evans, Bruce Chen, James Potter, Brenna Booth-Mowat and the anonymous reviewers for their comments on earlier versions of this article.

[1] See generally Etienne Mureinik, ‘A Bridge to Where? Introducing the Interim Bill of Rights’ (1994) 10(1) South African Journal on Human Rights 31, 32; Moshe Cohen-Eliya and Iddo Porat, ‘Proportionality and the Culture of Justification’ (2011) 59(2) American Journal of Comparative Law 463; Murray Hunt, ‘Introduction’ in Murray Hunt, Hayley J Hooper and Paul Yowell (eds), Parliaments and Human Rights: Redressing the Democratic Deficit (Hart Publishing, 2015) 1, 15–16.

[2] Human Rights Act 2019 (Qld) s 13(1) (‘Human Rights Act’).

[3] Momcilovic v The Queen (2011) 245 CLR 1, 170 [429] (Heydon J) (‘Momcilovic’).

[4] Canada Act 1982 (UK) c 11, sch B, pt I s 1 (‘Canadian Charter of Rights and Freedoms’).

[5] R v Oakes [1986] 1 SCR 103, 138–40 (Dickson CJ for Dickson CJ, Chouinard, Lamer, Wilson and Le Dain JJ) (‘Oakes’).

[6] See, eg, Alec Stone Sweet and Jud Mathews, ‘Proportionality Balancing and Global Constitutionalism’ (2008) 47(1) Columbia Journal of Transnational Law 72, 73–6; Adrienne Stone, ‘The Limits of Constitutional Text and Structure: Standards of Review and the Freedom of Political Communication’ [1999] MelbULawRw 26; (1999) 23(3) Melbourne University Law Review 668, 677–8 (‘The Limits of Constitutional Text and Structure’).

[7] Human Rights Act (n 2) s 13(2).

[8] Ibid s 13(1).

[9] See, eg, Innes v Electoral Commission (Qld) [No 2] [2020] QSC 293; (2020) 5 QR 623, 682–3 [287], 683–4 [295]–[300] (Ryan J); SF v Department of Education [2021] QCAT 10, [47]–[50] (Member Hughes).

[10] See, eg A-G (Qld) v Sri [2020] QSC 246, [28], [43] (Applegarth J); Queensland (Department of Housing and Public Works) v Tenant [2020] QCAT 144, [158], [161] (Adjudicator Alan Walsh).

[11] [2020] QSC 54; (2020) 4 QR 31 (‘AIP’). Shortly before this article was published, Martin J also considered s 13 in Owen-D’Arcy v Chief Executive, Queensland Corrective Services [2021] QSC 273 (‘Owen-D’Arcy’). His Honour held that s 13 should be regarded as embodying a proportionality test, and that when considering whether a limit on human rights is justified ‘the factors set out in s 13(2) should be addressed’: at [104], [244]. However, his Honour did not specifically address whether the factors are necessary or sufficient.

[12] AIP (n 11) 40–1 [2]–[4], 42–3 [11] (Applegarth J).

[13] Ibid 73 [119] (Applegarth J).

[14] Ibid 72 [114].

[15] This article does not draw a distinction between ‘reasonableness’ and ‘demonstrable justifiability’. As will be seen, justification by structured proportionality subsumes questions of reasonableness: see especially below nn 123–33 and accompanying text.

[16] Explanatory Notes, Human Rights Bill 2018 (Qld) 16 (‘Human Rights Bill Explanatory Notes’).

[17] Ibid.

[18] Ibid 16–17.

[19] See Human Rights Act 2004 (ACT) ss 28, 30 (‘ACT Human Rights Act’); Charter of Human Rights and Responsibilities Act 2006 (Vic) ss 7, 32 (‘Victorian Charter’). In relation to the Victorian Charter (n 19), see also Momcilovic (n 3) 42–3 [30]–[31] (French CJ), 209 [540] (Crennan and Kiefel JJ).

[20] See Human Rights Act (n 2) s 8.

[21] See Slaveski v Smith [2012] VSCA 25; (2012) 34 VR 206, 214–15 [21]–[22] (Warren CJ, Nettle and Redlich JJA); Director of Consumer Affairs Victoria v Operation Smile (Aust) Inc [2012] VSCA 91; (2012) 38 VR 569, 575–7 [26]–[31] (Warren CJ and Cavanough AJA); WBM v Chief Commissioner of Police [2012] VSCA 159; (2012) 43 VR 446, 473 [122] (Warren CJ, Hansen JA agreeing at 475 [133]); DPP (Vic) v Leys [2012] VSCA 304; (2012) 44 VR 1, 46–7 [138] (Redlich, Tate JJA and T Forrest AJA); Victoria Police Toll Enforcement v Taha [2013] VSCA 37; (2013) 49 VR 1, 62 [191] (Tate JA); DPP (Vic) v Kaba [2014] VSC 52; (2014) 44 VR 526, 587–9 [211]–[215] (Bell J) (‘Kaba’); Andrews v Thomson [2018] ACTCA 53; (2018) 340 FLR 439, 449–50 [52]–[54] (Elkaim, Loukas-Karlsson JJ and Robinson AJ) (‘Andrews’).

[22] A question which did not arise in AIP (n 11) was whether the link drawn between ss 13 and 48 by s 8 of the Human Rights Act (n 2) breaches the principle in Kable v DPP (NSW) [1996] HCA 24; (1996) 189 CLR 51, 101–3 (Gaudron J), 114–16 (McHugh J), 138, 143 (Gummow J), such that the provisions are constitutionally invalid. In Momcilovic (n 3), of the six judges who upheld the validity of the equivalent of s 48 (s 32 of the Victorian Charter (n 19)), three did so on the basis that there was no link between the equivalents of ss 13 and 48 (ss 7(2) and 32 of the Victorian Charter (n 19)): Momcilovic (n 3) 44 [35] (French CJ), 219–20 [574]–[575] (Crennan and Kiefel JJ). The other three did so on the basis that there was such a link: at 92 [168] (Gummow J, Hayne J agreeing at 123 [280]), 249–50 [683]–[684] (Bell J). Justice Heydon also agreed that there was a link, albeit in dissent as to the validity of s 32(1): at 170 [427].

[23] AIP (n 11) 73 [121] (Applegarth J). His Honour repeated this approach to s 13 more recently in TRKJ v DPP (Qld) [2021] QSC 297, [170]: ‘One or more of the factors in s 13(2) may be relevant in deciding whether a limit on a human right is reasonable and “demonstrably justified”’.

[24] Queensland, Parliamentary Debates, Legislative Assembly, 31 October 2018, 3185 (Yvette D’Ath, Attorney-General). See also Human Rights Bill Explanatory Notes (n 16) 5.

[25] AIP (n 11) 76 [134]–[135] (Applegarth J).

[26] Human Rights Bill Explanatory Notes (n 16) 16.

[27] Human Rights Act (n 2) s 13(1).

[28] Canadian Charter of Rights and Freedoms (n 4) s 1. In turn, s 1 is evidently based on art 29(2) of the Universal Declaration of Human Rights, GA Res 217A (III), UN GAOR, UN Doc A/810 (10 December 1948). Note that New Zealand’s general limitations clause also follows the Canadian model: New Zealand Bill of Rights Act 1990 (NZ) s 5.

[29] Canadian Charter of Rights and Freedoms (n 4) s 1.

[30] Oakes (n 5) 138–40 (Dickson CJ for Dickson CJ, Chouinard, Lamer, Wilson and Le Dain JJ).

[31] See Justice Susan Kiefel, ‘Proportionality: A Rule of Reason’ (2012) 23(2) Public Law Review 85, 87–8; Clubb v Edwards [2019] HCA 11; (2019) 267 CLR 171, 331–2 [466] (Edelman J) (‘Clubb’).

[32] See, eg, Robert Alexy, ‘Constitutional Rights and Proportionality’ (2014) 22 Revus 51, 51; Stone Sweet and Mathews (n 6) 74; Moshe Cohen-Eliya and Iddo Porat, ‘American Balancing and German Proportionality: The Historical Origins’ (2010) 8(2) International Journal of Constitutional Law 263, 263–4; Grégoire CN Webber, The Negotiable Constitution: On the Limitation of Rights (Cambridge University Press, 2009) 63–4. See also Re Kracke and Mental Health Review Board (2009) 29 VAR 1 (‘Re Kracke’), in which Bell J stated that proportionality is ‘applied universally across the various jurisdictions in determining whether limitations on human rights are justified’: at 35 [111]; Palmer v Western Australia [2021] HCA 5; (2021) 388 ALR 180, 193–4 [53] (Kiefel CJ and Keane J), 213 [141] (Gageler J), 247 [264] (Edelman J) (‘Palmer’).

[33] See, eg, Kai Möller, ‘Constructing the Proportionality Test: An Emerging Global Conversation’ in Liora Lazarus, Christopher McCrudden and Nigel Bowles (eds), Reasoning Rights: Comparative Judicial Engagement (Hart Publishing, 2014) 31, 33; Francisco J Urbina, A Critique of Proportionality and Balancing (Cambridge University Press, 2017) 5. However, in some articulations of structured proportionality, the first step of proper purpose or legitimate aim is assumed: see, eg, Kiefel (n 31) 88. In others, it is treated as a separate test prior to structured proportionality: see, eg, Oakes (n 5) 138–9 (Dickson CJ for Dickson CJ, Chouinard, Lamer, Wilson and Le Dain JJ). More recently, Edelman J grouped proper purpose and suitability together as a single step in reasoning: Palmer (n 32) 249 [269]–[270].

[34] R (Tigere) v Secretary of State for Business, Innovation and Skills [2015] UKSC 57; [2015] 1 WLR 3820, 3834 [33] (Baroness Hale DPSC, Lord Kerr JSC agreeing) (emphasis added), citing Huang v Secretary of State for the Home Department [2007] UKHL 11; [2007] 2 AC 167 (‘Huang’), R (Aguilar Quila) v Secretary of State for the Home Department [2011] UKSC 45; [2012] 1 AC 621, Bank Mellat v Her Majesty’s Treasury [No 2] [2014] AC 700 (‘Bank Mellat’).

[35] See Constitution of the Republic of South Africa Act 1996 (South Africa) s 36 (‘Constitution of South Africa’).

[36] The drafters adopted these factors from the judgment in S v Makwanyane [1995] ZACC 3; [1995] 3 SA 391, 436 [104] (Chaskalson P). See Niels Petersen, ‘Proportionality and the Incommensurability Challenge in the Jurisprudence of the South African Constitutional Court’ (2014) 30(3) South African Journal on Human Rights 405, 406.

[37] S v Manamela [2000] ZACC 5; [2000] 3 SA 1, 19–20 [32] (Madala, Sachs and Yacoob JJ) (‘Manamela’). See also Re Kracke (n 32) 40 [133] (Bell J).

[38] Carolyn Evans and Simon Evans, Australian Bills of Rights: The Law of the Victorian Charter and ACT Human Rights Act (LexisNexis Butterworths, 2008) 177 [5.48].

[39] Oakes (n 5) 138–40 (Dickson CJ for Dickson CJ, Chouinard, Lamer, Wilson and Le Dain JJ); Carter v Canada (A-G) [2015] 1 SCR 331, 378–9 [94] (The Court); R v KRJ [2016] 1 SCR 906, 938 [58] (Karakatsanis J for McLachlin CJ, Cromwell, Moldaver, Karakatsanis, Wagner, Gascon and Côté JJ).

[40] De Freitas v Permanent Secretary, Ministry of Agriculture, Fisheries, Lands and Housing [1998] UKPC 30; [1999] 1 AC 69, 80 (Lord Clyde for the Judicial Committee); Huang (n 34) 187 [19] (Lord Bingham for the Judicial Committee); Bank Mellat (n 34) 790–1 [72]–[74] (Lord Reed JSC); R (Nicklinson) v Ministry of Justice [2014] UKSC 38; [2015] AC 657, 807–9 [167]–[168] (Lord Mance JSC).

[41] R v Hansen [2007] 3 NZLR 1, 28 [64] (Blanchard J), 40–1 [103]–[104] (Tipping J), 69 [203]–[204] (McGrath J). Cf the recent case of Four Aviation Security Service Employees v Minister of COVID-19 Response [2021] NZHC 3012, in which Cooke J described the structured proportionality test as ‘only a framework’, and instead opted to address justification in that case by answering a series of evidentiary questions: at [91]. Nevertheless, having answered those questions, his Honour returned to the structured proportionality test as a capstone: at [124].

[42] Aharon Barak, Proportionality: Constitutional Rights and Their Limitations, tr Doron Kalir (Cambridge University Press, 2012) 461 (citations omitted) (‘Proportionality’).

[43] Ariel L Bendor and Tal Sela, ‘How Proportional Is Proportionality?’ (2015) 13(2) International Journal of Constitutional Law 530, 538.

[44] See Re Application for Bail by Islam [2010] ACTSC 147; (2010) 4 ACTLR 235, 292 [248] (Penfold J) (‘Re Islam’). Cf at 292 [250].

[45] Clubb (n 31) 331 [463]. See also Palmer (n 32) 248 [266] (Edelman J).

[46] The factors in s 7(2) of the Victorian Charter (n 19) were included in the Act as passed. In the ACT, the factors in s 28(2) were added by an amendment in 2008: Human Rights Amendment Act 2008 (ACT) s 4.

[47] Frederick Schauer, ‘Freedom of Expression Adjudication in Europe and America: A Case Study in Comparative Constitutional Architecture’ (Research Paper No RWPO5-019, John F Kennedy School of Government, Harvard University, February 2005) 6, quoted in Webber (n 32) 62.

[48] ACT Human Rights Act (n 19) s 28(2); Victorian Charter (n 19) s 7(2); Constitution of South Africa (n 35) s 36(1).

[49] Bruce Chen, ‘The Human Rights Act 2019 (Qld): Some Perspectives from Victoria’ (2020) 45(1) Alternative Law Journal 4, 5.

[50] Constitution of South Africa (n 35) s 36(1)(a); Victorian Charter (n 19) s 7(2)(a); ACT Human Rights Act (n 19) s 28(2)(a); Human Rights Act (n 2) s 13(2).

[51] Human Rights Bill Explanatory Notes (n 16) 17; Re Kracke (n 32) 41 [137]–[142] (Bell J); Kaba (n 21) 557 [106]–[108] (Bell J); Lawyers for Human Rights v Minister of Home Affairs [2017] ZACC 22; [2017] 5 SA 480, 488 [28] (Jafta J, Mogoeng CJ, Nkabinde ADCJ, Cameron, Froneman, Khampepe, Madlanga, Mhlantla JJ, Mojapelo, Pretorius AJJ and Zondo J agreeing).

[52] Human Rights Bill Explanatory Notes (n 16) 17.

[53] AIP (n 11) 73–4 [122], quoting Spence v Queensland [2019] HCA 15; (2019) 268 CLR 355, 416 [93] (Kiefel CJ, Bell, Gageler and Keane JJ) (‘Spence’).

[54] See AIP (n 11) 73–4 [122]–[124] (Applegarth J).

[55] See Constitution of South Africa (n 35) ss 36(1)(b)–(c); ACT Human Rights Act (n 19) ss 28(2)(b)–(c); Victorian Charter (n 19) ss 7(2)(b)–(c). See also Momcilovic (n 3) 213 [553] (Crennan and Kiefel JJ).

[56] Barak, Proportionality (n 42) 246–7.

[57] Ibid 245–51. See also Re Kracke (n 32) 42–3 [152] (Bell J); McCloy v New South Wales [2015] HCA 34; (2015) 257 CLR 178, 212–13 [67] (French CJ, Kiefel, Bell and Keane JJ) (‘McCloy’).

[58] Constitution of South Africa (n 35) s 36(1)(d); Victorian Charter (n 19) s 7(2)(d); ACT Human Rights Act (n 19) s 28(2)(d).

[59] See, eg, R (T) v Chief Constable of Greater Manchester Police [2014] UKSC 35; [2015] AC 49, 101 [142] (Lord Reed JSC). Note that Lord Wilson JSC appears to have decided the point at the necessity stage rather than the suitability stage: at 74 [40], 77 [50]. Lord Neuberger PSC, Baroness Hale DPSC and Lord Clarke JSC agreed with both Lord Reed JSC and Lord Wilson JSC: at 105 [158]. The headnote, at least, interprets the case as having been decided at the suitability stage: at 51.

[60] See, eg, Human Rights Bill Explanatory Notes (n 16) 17.

[61] Unions NSW v New South Wales (2013) 252 CLR 530, 557 [50], 558 [54], 559 [56], 560 [60], 561 [64]–[65] (French CJ, Hayne, Crennan, Kiefel and Bell JJ) (‘Unions NSW’).

[62] Barak, Proportionality (n 42) 303.

[63] AIP (n 11) 74 [125].

[64] Unions NSW (n 61) 559–60 [56]–[60] (French CJ, Hayne, Crennan, Kiefel and Bell JJ).

[65] McCloy (n 57) 208–10 [51]–[56] (French CJ, Kiefel, Bell and Keane JJ); Spence (n 53) 499 [322] (Edelman J). Queensland was also entitled to respond to the risk of corruption identified in New South Wales: at 416–17 [93]–[97] (Kiefel CJ, Bell, Gageler and Keane JJ).

[66] Constitution of South Africa (n 35) s 36(1)(e); Victorian Charter (n 19) s 7(2)(e); ACT Human Rights Act (n 19) s 28(2)(e).

[67] See, eg, Human Rights Bill Explanatory Notes (n 16) 17.

[68] AIP (n 11) 42 [11], 55 [68]–[70] (Applegarth J).

[69] Ibid 71 [108]–[109].

[70] Ibid 75 [126] (Applegarth J).

[71] Barak, Proportionality (n 42) 340; Human Rights Bill Explanatory Notes (n 16) 17–18.

[72] Alberta v Hutterian Brethren of Wilson Colony [2009] 2 SCR 567, 605 [76] (McLachlin CJ for McLachlin CJ, Binnie, Deschamps and Rothstein JJ).

[73] Ibid.

[74] Constitution of South Africa (n 35) s 36(1)(b); Victorian Charter (n 19) s 7(2)(b); ACT Human Rights Act (n 19) s 28(2)(b).

[75] But see Momcilovic (n 3) 213 [553], where Crennan and Kiefel JJ suggested that ‘at least the framework for’ the final balancing test may be gleaned from the other paragraphs.

[76] See, eg, Re Application under the Major Crime (Investigative Powers) Act 2004 [2009] VSC 381; (2009) 24 VR 415, 449 [148]–[149] (Warren CJ) (‘Re Major Crime Act’); Re Kracke (n 32) 40 [134] (Bell J); Re Islam (n 44) 290–1 [242]–[243], 292 [247] (Penfold J); Certain Children v Minister for Families and Children [No 2] [2017] VSC 251; (2017) 52 VR 441, 505 [205] (John Dixon J) (‘Certain Children [No 2]’).

[77] See, eg, AB v CD [2017] VSCA 338, [174]–[176] (‘AB’), where Ferguson CJ, Osborn and McLeish JJA merely considered the express factors in s 7(2) of the Victorian Charter (n 19), and therefore never expressly considered how important the human right was, nor whether the limit on the human right was outweighed by the proper purpose. Arguably, the Court may have considered the importance of the human right when considering its nature under s 7(2)(a), however that was neither express nor developed: see AB (n 77) [174] (Ferguson CJ, Osborn and McLeish JJA).

[78] Kai Möller, ‘Proportionality: Challenging the Critics’ (2012) 10(3) International Journal of Constitutional Law 709, 711; Dieter Grimm, ‘Proportionality in Canadian and German Constitutional Jurisprudence’ (2007) 57(2) University of Toronto Law Journal 383, 393, noting that most cases in Germany are decided at the fair balance stage. Cf Paul Yowell, Constitutional Rights and Constitutional Design: Moral and Empirical Reasoning in Judicial Review (Hart Publishing, 2018) 31, noting that most cases in Canada have been decided at the necessity stage.

[79] AIP (n 11) 75 [127]–[132].

[80] Ibid 76 [133].

[81] Human Rights Act (n 2) ss 13(2)(b)–(g).

[82] Barak, Proportionality (n 42) 161–4; Julie Debeljak, ‘Balancing Rights in a Democracy: The Problems with Limitations and Overrides of Rights under the Victorian Charter of Human Rights and Responsibilities Act 2006[2008] MelbULawRw 14; (2008) 32(2) Melbourne University Law Review 422, 424.

[83] McCloy (n 57) 194–5 [2] (French CJ, Kiefel, Bell and Keane JJ). See also Brown v Tasmania (2017) 261 CLR 328, 368–9 [123]–[127] (Kiefel CJ, Bell and Keane JJ), 416–17 [278] (Nettle J) (‘Brown’); Unions NSW v New South Wales [No 2] [2019] HCA 1; (2019) 264 CLR 595, 615 [42] (Kiefel CJ, Bell and Keane JJ), 638 [110] (Nettle J) (‘Unions NSW [No 2]’); Clubb (n 31) 186 [6] (Kiefel CJ, Bell and Keane JJ), 264–5 [266] (Nettle J), 311 [408], 330–1 [462]–[463] (Edelman J); Comcare v Banerji [2019] HCA 23; (2019) 267 CLR 373, 400 [32]–[33], 401–2 [35], 402–3 [38] (Kiefel CJ, Bell, Keane and Nettle JJ), 451 [188] (Edelman J).

[84] Palmer (n 32) 195–6 [61]–[62] (Kiefel CJ and Keane J), 247 [264] (Edelman J). Cf at 202 [94] (Gageler J), 228 [198] (Gordon J).

[85] See, eg, Brown (n 83) 368–9 [124]–[126] (Kiefel CJ, Bell and Keane JJ); Clubb (n 31) 199 [61] (Kiefel CJ, Bell and Keane JJ).

[86] See, eg, McCloy (n 57) 234–7 [140]–[146] (Gageler J), 281–2 [308]–[311], 288–9 [339] (Gordon J); Brown (n 83) 376–9 [157]–[166] (Gageler J), 464–8 [429]–[438] (Gordon J); Clubb (n 31) 223–5 [157]–[160] (Gageler J), 305–10 [390]–[404] (Gordon J); Palmer (n 32) 202 [94], 210–16 [129]–[151] (Gageler J), 228–9 [198]–[199] (Gordon J).

[87] Palmer (n 32) 213 [140] (Gageler J).

[88] McCloy (n 57) 217 [78]–[79] (French CJ, Kiefel, Bell and Keane JJ). See also at 234 [140] (Gageler J).

[89] See Alistair Pound and Kylie Evans, Annotated Victorian Charter of Rights (Lawbook, 2nd ed, 2019) 63 [7.140] (albeit in relation to s 7(2) of the Victorian Charter (n 19)).

[90] Note, in Murphy v Electoral Commissioner [2016] HCA 36; (2016) 261 CLR 28 (‘Murphy’), a majority of the High Court declined to apply structured proportionality to the implied voting rights arising from the mandate in ss 7 and 24 of the Constitution that the Parliament be directly chosen by the people: Murphy (n 90) 52–3 [37]–[39] (French CJ and Bell J), 71–2 [98]–[102] (Gageler J), 94 [202] (Keane J), 122–4 [294]–[305] (Gordon J). See also Palmer (n 32) 216 [150] (Gageler J). One explanation may be that in Murphy (n 90) the Court was dealing with a positive obligation to ensure direct choice by the people (rather than a negative right) and that the impugned electoral laws did not detract from that positive obligation sufficiently to engage the implied voting rights, such that there was no limit requiring justification: Amelia Simpson, ‘Section 92 as a Transplant Recipient?: Commentary on Chapter 8’ in John Griffiths and James Stellios (eds), Current Issues in Australian Constitutional Law: Tributes to Professor Leslie Zines (Federation Press, 2020) 283, 289. However, structured proportionality is capable of being applied to positive rights: see Murphy (n 90) 60–4 [60]–[74] (Kiefel J); Robert Alexy, ‘On Constitutional Rights to Protection’ (2009) 3(1) Legisprudence 1, 5–17.

[91] Brown (n 83) 368–9 [124]–[125] (Kiefel CJ, Bell and Keane JJ); Tasmania, ‘Submissions’, Submission in Brown v Tasmania, H3/2016, 21 March 2017, 11–13 [54]–[60]; Attorney-General (Qld), ‘Annotated Submissions of the Attorney-General for the State of Queensland (Intervening)’, Submission in Brown v Tasmania, H3/2016, 28 March 2017, 2–3 [8], 9–10 [30]–[32], 20 [69] (‘Qld Submissions in Brown’).

[92] From Lange v Australian Broadcasting Corporation [1997] HCA 25; (1997) 189 CLR 520, 561–2 (Brennan CJ, Dawson, Toohey, Gaudron, McHugh, Gummow and Kirby JJ) (‘Lange’).

[93] Qld Submissions in Brown (n 91) 9 [31].

[94] Manamela (n 37) 20 [32] (Madala, Sachs and Yacoob JJ).

[95] Brown (n 83) 369 [125]. See also Kiefel (n 31) 87, referring to the risk of ‘an unconstrained value judgment’ if an unstructured approach is taken.

[96] McCloy (n 57) 216 [75] (French CJ, Kiefel, Bell and Keane JJ), relied upon in rejecting Queensland’s proposed test in Brown (n 83) 369 [125], 369 [125] n 93 (Kiefel CJ, Bell and Keane JJ). See also Palmer (n 32) 193 [51], 194 [55] (Kiefel CJ and Keane J), 232 [217], 247–8 [262]–[266] (Edelman J).

[97] Gertrude Lübbe-Wolff, ‘The Principle of Proportionality in the Case-Law of the German Federal Constitutional Court’ (2014) 34(1– 6) Human Rights Law Journal 12, 16, quoted with approval in McCloy (n 57) 216 [77] (French CJ, Kiefel, Bell and Keane JJ).

[98] McCloy (n 57) 217 [78] (French CJ, Kiefel, Bell and Keane JJ), citing with approval Bank Mellat (n 34) 790 [72] (Lord Reed JSC).

[99] Palmer (n 32) 214 [146] (Gageler J).

[100] Ibid 214 [143], [146]. See also at 228 [198] (Gordon J).

[101] See also Stavros Tsakyrakis, ‘Proportionality: An Assault on Human Rights?’ (2009) 7(3) International Journal of Constitutional Law 468, 474–5.

[102] McCloy (n 57) 238 [150] (Gageler J).

[103] Brown (n 83) 369 [126] (Kiefel CJ, Bell and Keane JJ) (citations omitted).

[104] Clubb (n 31) 199 [61] (Kiefel CJ, Bell and Keane JJ).

[105] See ibid 262–3 [264] (Nettle J).

[106] Ibid 225 [161] (Gageler J). See also at 262 [264] (Nettle J).

[107] See, eg, Urbina (n 33) 13, 213–52; Tsakyrakis (n 101) 471.

[108] McCloy (n 57) 215 [72] (French CJ, Kiefel, Bell and Keane JJ).

[109] Barak, Proportionality (n 42) 502.

[110] See ibid 542–5.

[111] See, eg, United States v Playboy Entertainment Group Inc, [2000] USSC 46; 529 US 803, 813 (Kennedy J for Stevens, Kennedy, Souter, Thomas and Ginsburg JJ, Stephens J agreeing at 828, Thomas J agreeing at 829) (2000), cited in Clubb (n 31) 231 [180] n 178 (Gageler J).

[112] See, eg, Heller v Doe, [1993] USSC 92; 509 US 312, 319–21 (Kennedy J for Rehnquist CJ, White, Scalia, Kennedy and Thomas JJ) (1993).

[113] South Africa v Grootboom [2000] ZACC 19; [2001] 1 SA 46, 67 [38] (Yacoob J, Chaskalson P, Langa DP, Goldstone, Kriegler, Madala, Mokgoro, Ngcobo, O’Regan, Sachs JJ and Cameron AJ agreeing) (‘Grootboom’).

[114] Katharine G Young, ‘Proportionality, Reasonableness, and Economic and Social Rights’ in Vicki C Jackson and Mark Tushnet (eds), Proportionality: New Frontiers, New Challenges (Cambridge University Press, 2017) 248, 252–5. See also Associated Provincial Picture Houses Ltd v Wednesbury Corporation [1947] EWCA Civ 1; [1948] 1 KB 223.

[115] Although the Grootboom (n 113) test may go to limitation rather than justification, the Constitutional Court of South Africa has noted that all of the work of justification is done by the Grootboom test, leaving little work for the general limitations clause: Khosa v Minister of Social Development [2004] 6 SA 505, 540 [83]–[84] (Mokgoro J, Chaskalson CJ, Langa DCJ, Goldstone, Moseneke, O’Regan and Yacoob JJ agreeing), 549 [105]–[107] (Ngcobo J, Madala J agreeing); Phaahla v Minister of Justice and Correctional Services [2019] 7 BCLR 795, 812 n 56 (Dlodlo AJ, Mogoeng CJ, Khampepe, Mhlantla, Theron JJ, Basson, Goliath and Petse AJJ agreeing). On the interaction between Grootboom (n 113) and the general limitations clause, see NW Orago, ‘Limitation of Socio-Economic Rights in the 2010 Kenyan Constitution: A Proposal for the Adoption of a Proportionality Approach in the Judicial Adjudication of Socio-Economic Rights Disputes’ (2013) 16(5) Potchefstroom Electronic Law Journal 170, 199–206.

[116] [2016] 1 SCR 631 (‘Jordan’).

[117] Ibid 657 [46]–[47], 666 [76], 668 [81] (Moldaver, Karakatsanis and Brown JJ for Abella, Moldaver, Karakatsanis, Côté and Brown JJ). For the ongoing fallout from Jordan (n 116), see generally R v KGK (2020) 443 DLR (4th) 361; R v Thanabalasingham (2020) 447 DLR (4th) 310.

[118] See Frederick Schauer, ‘The Exceptional First Amendment’ in Michael Ignatieff (ed), American Exceptionalism and Human Rights (Princeton University Press, 2005) 29, 47–50; Stone, ‘The Limits of Constitutional Text and Structure’ (n 6) 704–5; Adrienne Stone, ‘Proportionality and Its Alternatives’ in John Griffiths and James Stellios (eds), Current Issues in Australian Constitutional Law: Tributes to Professor Leslie Zines (Federation Press, 2020) 170, 200–1.

[119] Tajjour v New South Wales [2014] HCA 35; (2014) 254 CLR 508, 580–1 [151] (Gageler J); McCloy (n 57) 238–9 [150]–[152] (Gageler J); Brown (n 83) 378 [164] (Gageler J).

[120] McCloy (n 57) 213 [68], 215–16 [74] (French CJ, Kiefel, Bell and Keane JJ). See also Brown (n 83) 370 [131] (Kiefel CJ, Bell and Keane JJ), 416–17 [278]–[280] (Nettle J). Cf Edelman J’s judgment in Clubb (n 31) 332–3 [468]: ‘It is, indeed, a tool. But its nature as a tool does not make it dispensable.’

[121] Palmer (n 32) 194 [56].

[122] Barak, Proportionality (n 42) 458. See also at 240–1.

[123] Cohen-Eliya and Porat, ‘Proportionality and the Culture of Justification’ (n 1) 466–7, 475.

[124] Robert Alexy, A Theory of Constitutional Rights, tr Julian Rivers (Oxford University Press, 2002) 74 [trans of: Theorie der Grundrechte (1986)]. See also Stone, ‘Proportionality and Its Alternatives’ (n 118) 179–80.

[125] David M Beatty, The Ultimate Rule of Law (Oxford University Press, 2004) 163. The current Chief Justice appears to have been influenced by this school of thought, having equated proportionality with ‘reasonableness of action’ and with ‘a rule of reason in public law’ in extrajudicial writings: Kiefel (n 31) 86. See also Murphy (n 90) 61 [65] (Kiefel J): ‘[t]he aim of any testing for proportionality is to ascertain the rationality and reasonableness of a legislative restriction’; Palmer (n 32) 194 [56] (Kiefel CJ and Keane J): ‘No one could doubt that proportionality is necessary to justification.’

[126] Lübbe-Wolff (n 97) 16 (emphasis added), quoted with approval in Pham v Secretary of State for the Home Department [2015] UKSC 19; [2015] 1 WLR 1591, 1622 [96] (Lord Mance JSC, Lord Neuberger PSC, Baroness Hale DPSC and Lord Wilson JSC agreeing), McCloy (n 57) 216 [77] (French CJ, Kiefel, Bell and Keane JJ).

[127] See Mattias Kumm, ‘The Idea of Socratic Contestation and the Right to Justification: The Point of Rights-Based Proportionality Review’ (2010) 4(2) Law & Ethics of Human Rights 141, 150.

[128] See Unions NSW [No 2] (n 83), where, on one reading, Kiefel CJ, Bell and Keane JJ assumed legitimate aim and proceeded to the later stages of the justification analysis: at 613 [35]–[36]. For an example in a human rights context, see Re Islam (n 44), where Penfold J proceeded to necessity even though there was a ‘real question’ as to whether the measure passed suitability: at 311 [357].

[129] Clubb (n 31) 199 [64].

[130] Brown (n 83) 370 [130] (Kiefel CJ, Bell and Keane JJ).

[131] Clubb (n 31) 201 [70] (Kiefel CJ, Bell and Keane JJ).

[132] Palmer (n 32) 195–6 [61]. See also at 248 [267], 250–1 [275]–[276] (Edelman J).

[133] Clubb (n 31) 263 [265]. See also Palmer (n 32) 194 [56] (Kiefel CJ and Keane J).

[134] Unions NSW (n 61) 554 [36] (French CJ, Hayne, Crennan, Kiefel and Bell JJ), regarding the implied freedom of political communication. There are exceptions to that general rule. For example, s 117 of the Constitution confers an individual right on interstate residents to non-discrimination: Street v Queensland Bar Association [1989] HCA 53; (1989) 168 CLR 461, 486 (Mason CJ), 503–4 (Brennan J), 541 (Dawson J).

[135] Lange (n 92) 561 (Brennan CJ, Dawson, Toohey, Gaudron, McHugh, Gummow and Kirby JJ).

[136] Robert Alexy, ‘Proportionality and Rationality’ in Vicki C Jackson and Mark Tushnet (eds), Proportionality: New Frontiers, New Challenges (Cambridge University Press, 2017) 13, 14, quoting Alexy, A Theory of Constitutional Rights (n 124) 47.

[137] Clubb (n 31) 307–8 [397], citing Alexy, ‘Proportionality and Rationality’ (n 136) 14. Note that proper purpose is assumed in this analysis.

[138] Murphy (n 90) 74 [110] (Gageler J), citing Alexy, A Theory of Constitutional Rights (n 124) 105, Barak, Proportionality (n 42) 364–5.

[139] Murphy (n 90) 74 [110].

[140] See Re Major Crime Act (n 76) 438 [97] (Warren CJ):

[T]he only difference of importance between such instruments [entrenching constitutional rights] and the [Victorian] Charter is in the remedial powers of the courts under such instruments. This fact alone should not impact on the question of whether jurisprudence from these jurisdictions may be of assistance in determining comparable principle.

[141] McCloy (n 57) 213 [68] (French CJ, Kiefel, Bell and Keane JJ).

[142] Commonwealth v Baume [1905] HCA 11; (1905) 2 CLR 405, 414 (Griffith CJ); Project Blue Sky Inc v Australian Broadcasting Authority (1998) 194 CLR 355, 382 [71] (McHugh, Gummow, Kirby and Hayne JJ); Saeed v Minister for Immigration and Citizenship [2010] HCA 23; (2010) 241 CLR 252, 266 [39] (French CJ, Gummow, Hayne, Crennan and Kiefel JJ).

[143] On the operation of s 15(3) generally, albeit not in relation to s 13(2), see Re Lifestyle Communities Ltd [No 3] (2009) 31 VAR 286, 344 [285] (Bell J); Matsoukatidou v Yarra Ranges Council [2017] VSC 61; (2017) 51 VR 624, 640 [52], 642–3 [58] (Bell J).

[144] Palmer (n 32) 194 [55] (Kiefel CJ and Keane J).

[145] Barak, Proportionality (n 42) 379, quoted with approval in McCloy (n 57) 216 [74] (French CJ, Kiefel, Bell and Keane JJ). See also Palmer (n 32) 194 [56] (Kiefel CJ and Keane J).

[146] See Palmer (n 32) 247 [263] (Edelman J).

[147] Barak, Proportionality (n 42) 542–5. See also McCloy (n 57) 236–7 [146]–[147] (Gageler J).

[148] See, eg, Barak, Proportionality (n 42) 33, 153–4; Johan de Waal, ‘A Comparative Analysis of the Provisions of German Origin in the Interim Bill of Rights’ (1995) 11(1) South African Journal on Human Rights 1, 25–6; AJ van der Walt, ‘The Limits of Constitutional Property’ (1997) 12(2) South African Public Law 275, 279–84. Cf Evans and Evans (n 38) 166–7 [5.20]–[5.22], where the authors do not appear to distinguish between internal limitations and the slightly different concept of internal modifiers (which are concerned with the scope of the right rather than justification).

[149] See Human Rights Bill Explanatory Notes (n 16) 18, 25.

[150] International Covenant on Civil and Political Rights, opened for signature 16 December 1966, 999 UNTS 171 (entered into force 23 March 1976) (‘ICCPR’).

[151] See Re Kracke (n 32) 35 [106] (Bell J); McDonald v Legal Services Commissioner [No 2] [2017] VSC 89, [30]–[31] (Bell J) (‘McDonald’).

[152] Evans and Evans (n 38) 166 [5.21], discussing s 15(3) of the Victorian Charter (n 19).

[153] The Young Review described the Victorian Charter (n 19) as a ‘“hybrid” model’ in this sense: Michael Brett Young, From Commitment to Culture: The 2015 Review of the Charter of Human Rights and Responsibilities Act 2006 (Review, September 2015) 157.

[154] Re Kracke (n 32) 35 [109]–[110]. The approach to internal limitations has been a vexed question in Victoria and the ACT. On the various approaches taken, see PJB v Melbourne Health [2011] VSC 327; (2011) 39 VR 373, 392 [74], 419 [213] (Bell J) (‘PJB’); Magee v Delaney [2012] VSC 407; (2012) 39 VR 50, 81 [157] (Kyrou J); McDonald (n 151) [31]–[32], [36] (Bell J); PBU v Mental Health Tribunal (2018) 56 VR 141, 178–9 [124] (Bell J); Andrews (n 21) 450–1 [56]–[57] (Elkaim, Loukas-Karlsson JJ and Robinson AJ). Shortly before publication of this article, the Victorian Court of Appeal handed down judgment in Thompson v Minogue [2021] VSCA 358 (‘Thompson’). The Court landed upon a position that treats internal limitations as internal modifiers, while acknowledging a significant overlap between internal limitations and the general limitations clause: at [44]–[58], [78] (Kyrou, McLeish and Niall JJA). It remains to be seen whether this approach is functionally different from the approach in Re Kracke (n 32).

[155] Note that the right in s 30(2) of the Human Rights Act (n 2) may also be subject to an external limitation: see Corrective Services Act 2006 (Qld) s 5A.

[156] Queensland, Parliamentary Debates, Legislative Assembly, 31 October 2018, 3185 (Yvette D’Ath, Attorney-General); Human Rights Bill Explanatory Notes (n 16) 5. On whether absolute human rights can exist, see Barak, Proportionality (n 42) 27–32.

[157] For a list of these absolute rights, see Debeljak (n 82) 434.

[158] ‘Absolute Rights: Public Sector Guidance Sheet’, Australian Government: Attorney-General’s Department (Web Page) <https://www.ag.gov.au/rights-and-protections/human-rights-and-anti-discrimination/human-rights-scrutiny/public-sector-guidance-sheets/absolute-rights>.

[159] For a list of the key non-derogable rights, see art 4 of the ICCPR (n 150).

[160] Human Rights Committee, General Comment No 29: States of Emergency (Article 4), 72nd sess, UN Doc CCPR/C/21/Rev.1/Add.11 (31 August 2001) 3 [7] (‘General Comment No 29’), noting: ‘Conceptually, the qualification of [an ICCPR] provision as a non-derogable one does not mean that no limitations or restrictions would ever be justified.’ See also Debeljak (n 82) 437 n 65.

[161] Human Rights Committee, General Comment No 24: General Comment on Issues Relating to Reservations Made upon Ratification or Accession to the Covenant or the Optional Protocols Thereto, or in Relation to Declarations under Article 41 of the Covenant, 52nd sess, UN Doc CCPR/C/21/Rev.1/Add.6 (11 November 1994) 4 [10] (‘General Comment No 24’).

[162] Certain Children [No 2] (n 76) 520 [254]–[255] (John Dixon J); Human Rights Committee, General Comment No 20: Article 7 (Prohibition of Torture, or Other Cruel, Inhuman or Degrading Treatment or Punishment), 44th sess (10 March 1992) 1 [3]; William A Schabas, UN International Covenant on Civil and Political Rights: Nowak’s CCPR Commentary (NP Engel, 3rd rev ed, 2019) 172 [4].

[163] International Covenant on Economic, Social and Cultural Rights, opened for signature 16 December 1996, 993 UNTS 3 (entered into force 3 January 1976) (‘ICESCR’).

[164] Committee on Economic, Social and Cultural Rights, General Comment No 3: The Nature of States Parties’ Obligations (Art 2, Para 1 of the Covenant), 5th sess, UN Doc E/1991/23 (14 December 1990) 3 [10].

[165] Amrei Müller, ‘Limitations to and Derogations from Economic, Social and Cultural Rights’ (2009) 9(4) Human Rights Law Review 557, 580.

[166] For the other core minimum obligations with respect to the right to education, see Committee on Economic, Social and Cultural Rights, General Comment No 13: The Right to Education (Article 13 of the Covenant), 21st sess, UN Doc E/C.12/1999/10 (8 December 1999) 12 [57]. For the core minimum obligations that apply to the right to health, see Committee on Economic, Social and Cultural Rights, General Comment No 14: The Right to the Highest Attainable Standard of Health (Article 12 of the International Covenant on Economic, Social and Cultural Rights), 22nd sess, UN Doc E/C.12/2000/4 (11 August 2000) 12–13 [43].

[167] However, the Constitutional Court of South Africa has declined to introduce absolute core minimum obligations for the socio-economic rights enshrined in the South African Bill of Rights: Grootboom (n 113) 65–6 [31]–[33] (Yacoob J, Chaskalson P, Langa DP, Goldstone, Kriegler, Madala, Mokgoro, Ngcobo, O’Regan, Sachs JJ and Cameron AJ agreeing); Minister of Health v Treatment Action Campaign [No 2] [2002] ZACC 15; [2002] 5 SA 721, 739–40 [34]–[39] (The Court) (‘Treatment Action’).

[168] See Pound and Evans (n 89) 97–8 [10.120]. For the possibility of applying minimum core obligations to socio-economic rights in this way, see Grootboom (n 113) 66 [33] (Yacoob J, Chaskalson P, Langa DP, Goldstone, Kriegler, Madala, Mokgoro, Ngcobo, O’Regan, Sachs JJ and Cameron AJ agreeing); Treatment Action (n 167) 739 [34] (The Court).

[169] Human Rights Committee, General Comment No 32: Article 14: Right to Equality before Courts and Tribunals and to a Fair Trial, 90th sess, UN Doc CCPR/C/GC/32 (23 August 2007) 5 [19]. On certain aspects of the right to a fair hearing being non-derogable, see General Comment No 29 (n 160) 5 [11], 6 [16]; General Comment No 24 (n 161) 3 [8].

[170] Mackin v New Brunswick (Minister of Finance) [2002] 1 SCR 405, 439 [72]–[73] (Gonthier J for L’Heureux-Dubé, Gonthier, Iacobucci, Major and Arbour JJ) (‘Mackin’).

[171] Conférence des Juges de Paix Magistrats du Québec v Québec (A-G) [2016] 2 SCR 116, 152 [97] (Karakatsanis, Wagner and Côté JJ for the Court).

[172] Ibid, quoting Mackin (n 170) 439 [72] (Gonthier J for L’Heureux-Dubé, Gonthier, Iacobucci, Major and Arbour JJ).

[173] See also Debeljak (n 82) 434.

[174] For example, s 48(3) cannot be used to import the ‘procedural’ aspects of human rights in circumstances where ss 58 and 59 already provide for remedies: Bare v Independent Broad-Based Anti-Corruption Commission [2015] VSCA 197; (2015) 48 VR 129, 198 [214] (Warren CJ), 279 [457] (Tate JA), 349 [665] (Santamaria JA) (‘Bare’). See generally French CJ’s note of caution with respect to the Victorian equivalent of s 48 in Momcilovic (n 3) 36–8 [18]–[19]. Note also Emerton J’s caution about using the equivalent of s 48(3) to trump a contrary statement in the explanatory memorandum: Castles v Secretary, Department of Justice [2010] VSC 310; (2010) 28 VR 141, 161 [71] (‘Castles’).

[175] Owen-D’Arcy (n 11) [103] (Martin J). See, similarly, in respect of the Victorian Charter (n 19), DPP (Vic) v Mokbel [No 1] [2010] VSC 331, [163] (Whelan J): ‘Under the [Victorian] Charter there are no absolute rights; all rights are subject to s 7’; Momcilovic (n 3) 162–3 [407], 165 [415] (Heydon J), 247–8 [678] (Bell J); R v Independent Broad-Based Anti-Corruption Commissioner [2016] HCA 8; (2016) 256 CLR 459, 478 [71] (Gageler J); Certain Children [No 2] (n 76) 495 [168], 497 [172] (John Dixon J). Cf Wagners Cement Pty Ltd v Boral Resources (Qld) Pty Ltd [2020] QSC 124, where Bond J queried whether s 48(3) might alter the analysis under s 13: at [11].

[176] R v Condon [2006] NZSC 62; [2007] 1 NZLR 300, 332 [77] (Blanchard J for the Court).

[177] See Human Rights Act (n 2) ss 13(2)(e)–(f).

[178] Barak, Proportionality (n 42) 508–9.

[179] See ibid 508–12, 544; Aharon Barak, ‘Proportionality and Principled Balancing’ (2010) 4(1) Law & Ethics of Human Rights 1, 12.

[180] But note the earliest tests of justification employed in Australian constitutional law derived from US authority, which may in turn have derived from Germany: Clubb (n 31) 331 [465] (Edelman J), citing McCulloch v Maryland, 17 US [1819] USSC 5; (4 Wheat) 316, 421 (Marshall CJ for the Court) (1819).

[181] Clubb (n 31) 263 [265] (Nettle J).

[182] The source of the duty of progressive realisation is art 2 of the ICESCR (n 163). On the relationship between justification and the duty of progressive realisation, see Müller (n 165) 570; Fons Coomans, ‘Reviewing Implementation of Social and Economic Rights: An Assessment of the “Reasonableness” Test as Developed by the South African Constitutional Court’ (2005) 65 Heidelberg Journal of International Law 167, 191–4.

[183] Jordan (n 116) 657 [46]–[47] (Moldaver, Karakatsanis and Brown JJ for Abella, Moldaver, Karakatsanis, Côté and Brown JJ).

[184] Intermediate appellate authority on human rights in Victoria and the ACT may be binding in Queensland unless plainly wrong: Baker v DPP (Vic) [2017] VSCA 58; (2017) 270 A Crim R 318, 338–9 [83] (Tate JA, Maxwell P agreeing at 321 [1]); Re Islam (n 44) 268 [137] (Penfold J). Both cases apply the High Court’s decision in Farah Constructions Pty Ltd v Say-Dee Pty Ltd (2007) 230 CLR 89, 151–2 [135] (Gleeson CJ, Gummow, Callinan, Heydon and Crennan JJ). Cf Harding v Sutton [2021] VSC 741, [210] n 71 (Richards J). See also Waratah Coal Pty Ltd v Youth Verdict Ltd [2020] QLC 33, [21] (Kingham P) (‘Waratah Coal’): ‘[I]t is legitimate and instructive to have regard to the text of the Victorian Charter, and relevant jurisprudence about it, paying due regard to any differences in the text of the two Acts.’

[185] Stone Sweet and Mathews (n 6) 74. See also Clubb (n 31) 331–2 [466] (Edelman J). Cf at 306–7 [395] (Gordon J).

[186] See Alexy, A Theory of Constitutional Rights (n 124) 406; Barak, ‘Proportionality and Principled Balancing’ (n 179) 14.

[187] See above nn 126–33 and accompanying text.

[188] Human Rights Act (n 2) s 38(2)(a). Portfolio committees in the Legislative Assembly also have a procedural obligation to ‘consider ... whether the Bill is not compatible with human rights’: at s 39(a). A similar procedural obligation applies to consideration by Ministers of the human rights compatibility of subordinate legislation in the form of ‘human rights certificates’: at s 41.

[189] That these two obligations are distinct was confirmed in Certain Children [No 2] (n 76) 510–11 [225]–[226] (John Dixon J) and Waratah Coal (n 184) [24]–[25] (Kingham P). Cf Thompson (n 154) [101], where Kyrou, McLeish and Niall JJA confirmed that it is possible to simultaneously comply with the procedural limb and breach the substantive limb. However their Honours left open whether the reverse is possible — that is, to simultaneously comply with the substantive limb and breach the procedural limb.

[190] PJB (n 154) 442 [312] (Bell J) (emphasis in original).

[191] R (SB) v Governors of Denbigh High School [2006] UKHL 15; [2007] 1 AC 100, 116 [31] (Lord Bingham), quoted with approval in Certain Children [No 2] (n 76) 508 [215] (John Dixon J).

[192] See, eg, LG v Melbourne Health [2019] VSC 183, [80]–[83] (Richards J); Certain Children v Minister for Families and Children [2016] VSC 796; (2016) 51 VR 473, 510–11 [197]–[199], [202]–[203] (Garde J) (‘Certain Children’). Cf Thompson (n 154) [101] (Kyrou, McLeish and Niall JJA).

[193] On the concept of a ‘dialogue’, see George Williams, ‘The Distinctive Features of Australia’s Human Rights Charters’ in Matthew Groves and Colin Campbell (eds), Australian Charters of Rights a Decade On (Federation Press, 2017) 22, 23; R v Momcilovic [2010] VSCA 50; (2010) 25 VR 436, 462–3 [93]–[96] (Maxwell P, Ashley and Neave JJA). For criticisms of the term ‘dialogue model’, see, eg, Momcilovic (n 3) 67–8 [95]–[96] (French CJ).

[194] Contrary to one of the main objects of the Human Rights Act (n 2): at s 3(b).

[195] Castles (n 174) 184 [185] (Emerton J), cited with approval in Bare (n 174) 199 [221] (Warren CJ), 219–20 [279], 223 [288]–[289] (Tate JA), 297–9 [535]–[538] (Santamaria JA), Hoskin v Greater Bendigo City Council [2015] VSCA 350; (2015) 48 VR 715, 725 [35]–[36] (Warren CJ, Osborn and Santamaria JJA), Owen-D’Arcy (n 11) [137] (Martin J).

[196] Cf Minogue v Thompson [2021] VSC 56, [51]–[53] (‘Minogue’), where Richards J aligned the requirement of proper consideration with the general limitations clause in s 7(2) of the Victorian Charter (n 19). Cf Thompson (n 154) [84]–[89] (Kyrou, McLeish and Niall JJA).

[197] Cf Owen-D’Arcy (n 11) [134]–[137] (Martin J).

[198] See Certain Children [No 2] (n 76) 581 [474]–[475] (John Dixon J), albeit under the rubric of the substantive limb.

[199] Certain Children (n 192) 512 [210] (Garde J).

[200] An analogy can be drawn to Li unreasonableness in the context of judicial review. Unreasonableness has both a ‘procedural’ and ‘substantive character’: Minister for Immigration and Citizenship v Li [2012] HCA 61; (2013) 249 CLR 332, 350 [26] (French CJ). Where the ‘outcome’ is ‘obviously disproportionate’, it may be inferred that the process of reasoning leading to it was flawed: at 366 [74] (Hayne, Kiefel and Bell JJ). For similar statements, see also at 350 [27] (French CJ), 364 [68], 367 [76] (Hayne, Kiefel and Bell JJ), 371 [91], 375 [105] (Gageler J).

[201] PJB (n 154) 442 [310] (Bell J).

[202] See also Barak, Proportionality (n 42) 379–81.

[203] See Minogue (n 196) [54], [66], [69], [75] (Richards J); Thompson (n 154) [90] (Kyrou, McLeish and Niall JJ).

[204] Certain Children [No 2] (n 76).

[205] Ibid 584 [491]. Although that was the context in which John Dixon J demanded a higher standard of proper consideration, the specific reasons his Honour gave were that the human rights assessment had been carried out by lawyers and was guided by the earlier decision of Certain Children (n 192): Certain Children [No 2] (n 76) 584 [491] (John Dixon J).

[206] See similarly Momcilovic (n 3) 248 [679] (Bell J).

[207] See above nn 23–6 and accompanying text.

[208] Kaba (n 21) 647 [468] (Bell J). See also Barak, Proportionality (n 42) 107–27; Borrowdale v Director-General of Health [2020] NZHC 2090; [2020] 2 NZLR 864, 912 [225], 915 [240] (Thomas, Venning and Ellis JJ).

[209] Palmer (n 32) 214 [146].

[210] Ibid 248 [268].

[211] Barak, Proportionality (n 42) 528.

[212] See above nn 126–32 and accompanying text.


AustLII: Copyright Policy | Disclaimers | Privacy Policy | Feedback
URL: http://www.austlii.edu.au/au/journals/MelbULawRw/2022/2.html