A Human Rights Charter for Australia – Part 2
This is part 2 of a 2-part series from Dr Paul Taylor. See part 1 here.
Dr Paul Taylor is an Honorary Senior Lecturer in the T.C. Beirne School of Law, a Fellow of the Centre for Public, International and Comparative Law, and an Adjunct Professor at the School of Law, The University of Notre Dame Australia. His principal academic interests are international human rights law and conflict of laws (private international law). He has held Visiting Fellowships at Wolfson College, Cambridge and at the Centre for International and Public Law, College of Law, Australian National University. He is the author of A Commentary on the International Covenant on Civil and Political Rights: The UN Human Rights’ Committee’s Monitoring of ICCPR Rights (Cambridge University Press, 2020); and Freedom of Religion: UN and European Human Rights Law and Practice (Cambridge University Press, 2005).
The comments are written in a personal capacity, and do not reflect any views of any institution to which he is or has been affiliated.
Orginally posted on Law and Religion Australia
Chart a better course?
Since my last guest blog on this subject (A smarter charter?), the Attorney-General has referred the question of whether there should be a federal charter to the Parliamentary Joint Committee on Human Rights (PJCHR). The Terms of Reference can be found at https://www.aph.gov.au/Parliamentary_Business/Committees/Joint/Human_Rights/HumanRightsFramework. Submissions are due by 1 July 2023.
The Position Paper produced by the AHRC, setting out its proposals for a federal charter, is a valuable resource, providing detailed coverage across a broad range of human rights issues. It is informational and constructive. It is also makes for heartening reading because of the many positive results achieved in the experience of different charters here and overseas, for so many people.
However, the issue that has now become central, as this inquiry proceeds, is not whether a charter is merely capable of yielding benefits. A charter, of necessity, must be dependable both in delivering the right outcomes, and in avoiding errant ones. I am far from convinced that existing ACT, Victoria and Queensland charters, and the one now proposed at federal level, which in many respects is similar but replete with home-spun oddities, meets that criterion – even though some good may have come out of charters so far.
Introduction
This supplemental blog post will focus on what this charter is unable to achieve (given that some of the normal “corrective” charter mechanisms are weak or unavailable in Australia), and what it is capable of achieving, through the apparatus of public authorities charged with a new and far-reaching obligation to “act human rights-compatibly”, if the AHRC’s ambitions are to be realised. This will certainly be an extremely powerful agent of change in Australia, principally through profuse bureaucratic output. There is no question about it, the charter will bring about a radical transformation. Whether it is for good or harm is answered by the settings established in the charter definitions of protected human rights, and the scope available for restricting rights, since they above all else set the course of the charter. If they correspond with those in the International Covenant on Civil and Political Rights (ICCPR) and the International Covenant on Economic, Social and Cultural Rights (ICESCR) all should be well, provided there is also a suitable way of monitoring whether public authorities are sticking to the script. If charter rights are defined and treated in ways that depart materially from the requirements of the ICCPR and ICESCR, or if public authorities go off mission, then Australia has a fundamental human rights problem.
Unfortunately, the charter definitions of protected human rights are at variance with the ICCPR and ICESCR, as is the scheme of permissible restriction particularly for ICCPR rights. There is no convincing explanation for any of this. It would have been a simple task to schedule the convention rights, just as the UK charter does with European Convention rights. This would provide a completely accurate definition of each of the rights, including applicable thresholds, qualifications (such as “arbitrary”), and the terms on which they may be restricted (limitation terms). This is the only way to avoid elevating some rights over others, and introducing other distortions. Unaccountably, in this charter the freedom from discrimination receives unique and privileged treatment.
The ICCPR prohibition against subjecting a person to torture, cruel, inhuman or degrading treatment or punishment has been recast in different ways, to generate what appears to be a new low-threshold right not to be “treated in a degrading way.” It is reminiscent of vilification prohibitions and could be recruited in support of claims analogous to discrimination for a broader range of conduct. The right to recognition as a person before the law is now combined with the discrimination provision when it has far broader, autonomous scope. The charter applies a single “limitations clause” generally across rights, when in the ICCPR such a provision (in tighter form) only applies to the freedoms of religion, of expression, of association and of assembly (plus a component of the freedom of movement). Because of this, the charter then has to disapply the “limitations clause” where it has no place.
A better solution would be to confine such a clause only to specified rights, as the ICCPR directs (as the result of 20 years of careful drafting deliberation). This would avoid the charter designating some rights “absolute” and elevating them (more than optically) merely because they are not subject to a limitation provision or other obvious qualification (they may e.g. be defined solely in terms of a prohibition against specified conduct, such as the prohibition against the retrospective operation of criminal laws). The charter would explicitly give “absolute” protection to (among other rights) the right to recognition as a person before the law, and the right not to be treated in a degrading way (which is not entitled to absolute protection to the extent that it strays beyond its ICCPR origins), both of which complement the freedom from discrimination. Certain rights which are well established as absolute receive no acknowledgement of that in this charter, namely freedom of opinion, and key aspects of the freedom of thought, conscience and religion. In relation to the latter, the charter either deletes the right or so messes with the ICCPR text as to render it meaningless. That is very peculiar.
Under the AHRC’s proposal, the reconfiguration of ICCPR and ICESCR rights would reinforce existing disparities in rights protection, and lean even more heavily than now in favour of protection from discrimination, at the expense of other rights. The whole point of a charter is that it should faithfully reflect the protection due to each and every right in an even-handed way. This charter certainly does not do that. The simple adoption of ICCPR and ICESCR text would. By adapting rights in departure from convention text the charter is not merely undesirable or inexpedient. It is idiosyncratic and at variance with well understood international law principles. It sets Australian human rights protection on a completely different, untenable, trajectory.
Does this charter achieve treaty implementation? No.
(1) Some of the usual corrective mechanisms of a charter are weak or unavailable in Australia
It is important to appreciate the limits of any dialogue-form charter like this, and the additional constraints imposed by Australian constitutional law.
A charter is meant to set up a warning system to alert Parliament to certain human rights deficiencies in proposed legislation, measured by convention standards such as those in European Convention or ICCPR. A charter typically does so first, through a “statement of compatibility,” which speaks to how well a Bill accords with human rights when tabled by the responsible Minister; and secondly, through the work of a Parliamentary scrutiny committee (like the PJCHR at federal level), which assesses that “statement of compatibility.” These procedures have value but are limited in practical reach, especially since they assiduously avoid interfering with the supremacy of Parliament. They are not that effective. The “statement of compatibility” is often of questionable utility, especially where it offers only a light human rights analysis and is written in a defensive posture. There are no direct adverse consequences for tabling a deficient one. When the scrutiny committee gets involved it is hampered by an overwhelming throughput of legislation, much of it with no human rights implications at all. Even when it exposes shortcomings, this may provoke some questioning but often not much more.
The role of the court in this model of charter can be far more productive, though not in Australia. If a matter before a court exposes particular human rights shortcomings, a court would usually have power to construe legislation in a rights-favourable way, and if that is not possible it may grant a “declaration of incompatibility.” These powers have been particularly valuable under the UK charter, where both have been exercised to considerable effect. However, this charter proposal is necessarily weak, as far as the corrective power of courts is concerned, not for any failing in design, but for Australian constitutional law reasons.
The following quick comparison between the UK model and existing Australian charters reveals the limits of what is possible in Australia. Under the UK Human Rights Act 1998, courts are empowered to “read and give effect” to legislation to render it rights-compatible, and may depart from the statutory text to achieve that result. Australian charters adopt much weaker terminology (“so far as it is possible to do so consistently with its purpose”), and the most a court can do is choose the most rights-favourable among available constructional choices. That leaves the court with very little scope. Where a rights-compatible interpretation of a provision is not possible in the circumstances, a UK court may make a declaration to that effect, and has done on scores of occasions. The power to make such a declaration would not be available under the federal charter proposal, again for constitutional law reasons. Instead, it is proposed that the court’s judgment may include reasoning which discloses that a rights-compatible interpretation is not possible. The value of a charter in Australia to effect positive change through the courts’ interpretive and declaratory functions is therefore limited.
(2) There is no substitute for implementing international convention rights in legislation
None of these charter mechanisms (whether in Australia or elsewhere) is capable of making good a fundamental legislative or other deficiency that exists in the legal protection for a human right. Only Parliament can do that, and then only if there is political will. A charter may advert to such a deficiency, say when support for a particular right is most needed in court but found wanting, but nothing can remedy the underlying failure in the law. The canary is already dead in the mine shaft.
A charter cannot therefore substitute for implementation of convention rights, as is demanded under the ICCPR (with particular requirements) and certain other conventions, to give effect to rights in law, where protection is otherwise lacking. The goal of ICCPR implementation is to enshrine human rights protection in law so that the intended beneficiary may invoke the law to withstand and/or remedy incursion, whether from public or private sources. There are probably few rights in need of such substantial remediation in Australian law. I would suggest those that do should be identified and have legislative priority over a charter. (Obvious candidates would be freedom of religion and freedom of expression). Simply requiring public authorities to act in a rights-compatible way does not “implement” rights, even if the rights in question are correctly established in the charter (which in this case they are not). A right is not implemented until it exists substantively in the legal system, such that it can be asserted independently of a charter, with provision for a remedy.
What would a federal charter do if it were in place now?
(1) The duty on public authorities to act compatibly with human rights, and to consider those rights when making decisions
The AHRC’s Free and Equal proposals have strong ambitions for a rights culture change by introducing a ‘positive duty’ on public authorities to act compatibly with human rights, and to give proper consideration to human rights when making decisions and implementing legislation. The much narrower “participation duty” (applicable to First Nations peoples, children and those with disability) recognises a broad, fundamental problem in the development of federal policies and decisions: inadequate engagement with the very people to whom those decisions directly apply.
Of these the most potent by far is the duty to act compatibly with human rights. Given the limited corrective power of a charter in its other functions, the AHRC understandably places this one “at the centre”. What makes the AHRC’s charter vision distinctive is the sheer force and enthusiasm which it puts behind this duty. As the AHRC explained,
“A positive duty must be accompanied by intensive measures to ensure cultural change and the adoption of a preventative approach to human rights protection within public authorities”.
The AHRC clearly intends a powerful shot in the arm for this sleeping giant. The purpose would be “to develop proficiency within the public service,” including through a “whole-of-government education program,” and “permanent routine educational requirements at all levels of government to maintain fluency” with the charter.
Compared with other charter mechanisms, this one, aimed at vivifying the entire public sector, is the least circumscribed. This obligation on public authorities to act compatibly with human rights is much more powerful and open-ended. The AHRC expresses it as a “positive duty” (while existing charters such as in the ACT and in the UK express it negatively, in terms of an obligation not to act incompatibly with the listed human rights). The outer boundary, where the obligation no longer applies, is typically the point at which the obligation is specifically contradicted by legislation in force. For example, in the ACT the obligation on public authorities does not apply where a law “expressly requires the act to be done or decision made in a particular way and that way is inconsistent with a human right”; and in the UK it does not apply if “as the result of one or more provisions of primary legislation, the authority could not have acted differently.”
The scope of this particular public authority obligation is therefore expansive. The AHRC’s ambitions in this respect are clear. Every aspect of public authority service delivery, policy and decision-making, and public service functionality, extending to the private sector where it undertakes public functions, may be required to be thoroughly compatible with the standards specified in the charter. Charter standards (not ICCPR standards) would become pervasive and bureaucracy-driven.
It really does not comfort the Australian public to say that the AHRC would be well placed to develop general and specific guidance for public authorities about how all limbs of the positive duty are to be applied (as the Position Paper does), including the application of the “limitations clause,” when it devised a charter which so inexplicably and extensively adapted ICCPR and ICESCR text, as if that were a matter of no great importance. This is a fundamental but easily avoidable design fault.
Whatever human rights standards are ultimately enshrined in the charter, a crucial question that remains is how ever will it be possible to ensure that they are promulgated, in appropriate balance, having regard to all rights, even when they are in tension, throughout all public authorities?
This brings us back to the primacy of clear compass-settings for this charter. Charter rights must be capable of being universally understood. For too many of the rights proposed by the AHRC I have no idea of their meaning, or intended meaning, especially in light of the degree of drift from the original convention text. The starting point must be the clearest specification of charter rights. It is crucially important that the charter list accurately, without any distortion, gap or bias, the ICCPR and ICESCR rights which are intended as the reference point. A charter is capable of being powerfully counter-productive, and rights-destructive, if it is able to press home an unbalanced or distorted conception of rights. An ill-conceived charter could well do just that.
A charter is meant to stand apart from the rest of a jurisdiction’s domestic law, to signify the standards set by international conventions. A charter draws attention to parochial deficiencies, and points to something better. It respects parliamentary sovereignty by not upsetting legislation even where it is irretrievably incompatible with those international standards. That is why the charter mechanisms concerned with parliamentary process and statutory interpretation are limited in what they can achieve. However, the impact of the charter through the influence of this particular public authority obligation is potentially massive. Even if the rights listed in this charter can be fixed, the practical emanation of the public authority duty must be rendered susceptible to oversight through a charter rights lens, in combination with accountability and correction.
One of the greatest hazards liable to be created by this charter is an aberrated human rights culture permeated through bureaucracy-developed policies, guidelines and norms, unchecked by effective scrutiny measures.
(2) What impact would this charter have if it were in effect now?
It is instructive to consider what work this federal charter would do, if it were already in force. One way to test that is to ask what difference would it make if the outcome of the current ALRC inquiry on religious schools’ exemptions were the entire removal of those exemptions (taking an extreme scenario)? Assume also, for the sake of simplicity, that this outcome is contrary to international convention standards.
The first step in a charter analysis would be to identify the human rights engaged. This should identify the charter rights which support the operation of religious schools in accordance with the exemptions, and the freedom from discrimination which is the basis for removing them. If the right to establish religious schools is an aspect of the “freedom to manifest one’s religion or beliefs…individually or in community with others” (which the UN Human Rights Committee says it is), it corresponds closest with the proposed charter freedom of religion. However, compared with the terms of limitation applicable to that right in the ICCPR, and the rigour required by the UN Human Rights Committee to justify restrictions on it, the proposed charter “limitations clause” exposes the right to a wider range of restriction than is permissible under the ICCPR article 18(3).
The right to religious manifestation should be read in light of “the liberty of parents…to ensure the religious and moral education of their children in conformity with their own convictions” in ICCPR article 18(4), which benefits from absolute protection. Except that it is missing from the charter’s freedom of religion. Why? The Position Paper offers no explanation. It is protected under the UK charter, by virtue of European Convention, First Protocol, article 2, where it must be respected in the exercise of any functions assumed in relation to education and teaching.
The federal charter includes in its adaptation of the ICESCR 13(3) right to education the right of parents to “choose schooling for the child to ensure the religious and moral education of the child in accordance with their convictions,” but it has no value if such schooling is unavailable with the demise of the anti-discrimination exemptions required to support it. In any case, ICESCR article 13(3) has two separate elements, of which that is only one. The other is the parental right “to ensure the religious and moral education of their children in conformity with their own convictions”! This too was deliberately left out by the AHRC. Because it is so clearly stated in ICCPR article 18(4) and ICESCR article 13(3) it cannot simply be an oversight. Article 13(4) also states in the clearest terms that “[n]o part of this article shall be construed so as to interfere with the liberty of individuals and bodies to establish and direct educational institutions,” etc. This is obviously highly relevant to the treatment of religious school exemptions. It also is missing. (So also are key components of ICESCR article 13(1) which place a high value on human rights and tolerance in education by emphasising that education “shall strengthen the respect for human rights and fundamental freedoms…and…shall enable all persons to participate effectively in a free society, promote understanding, tolerance and friendship among all nations and all racial, ethnic or religious groups.”)
A properly functioning charter should be able to detect domestic law inadequacies in its protection of all such fundamental rights. This charter’s capacity to do that is frustrated by removing or diminishing key rights from the charter (e.g. as we see here, those in support of religious schools). The rights exist at international law but are not visible on the charter radar.
This charter also does something extraordinary to prevent the detection of excessive protection of one right at the expense of another. Normally a charter would define all its rights by reference to those in the relevant international convention. The UK charter does exactly what it should, when it schedules all the European Convention rights in full, as they appear in the Convention text, including any limitation text where applicable to a right. It thereby subjects all rights to the same European Convention standard. Even though the Equality Act 2010 (UK) contains extensive domestic law prohibitions on discrimination, just as Australian anti-discrimination law does, the UK charter does not single out freedom from discrimination in its list of applicable human rights and substitute a UK domestic law standard (e.g. under the Equality Act 2010). Yet the charter makes exactly this kind of substitution.
All rights in the proposed charter are referrable in some way to ICCPR and ICESCR rights. Except that “discrimination” in the charter’s freedom from discrimination is defined by reference to federal anti-discrimination legislation (“[d]iscrimination in the context of the [charter] has the same meaning as discrimination in federal discrimination laws (including any future discrimination legislation)”). The discrimination standards of international human rights conventions do not apply to the charter term “discrimination.” It is not easy to grasp why the charter does this, but the implications are profound. The charter’s freedom from discrimination has different elements, each of which has a counterpart in ICCPR provisions (though significantly adapted from the ICCPR). None of them has any direct equivalent in Australian domestic law. In the charter, “Every person has the right to enjoy the person’s human rights without discrimination” corresponds with ICCPR article 2; “Every person is equal before the law and is entitled to the equal protection of the law without discrimination” and “Every person has the right to equal and effective protection against discrimination” correspond with different clauses within ICCPR article 26. The term “discrimination” has a particular meaning under those ICCPR provisions, which broadly excludes from the prohibition against discrimination any differentiation based on reasonable and objective criteria. The principle applies equally to direct and indirect discrimination.
The difference between the international law and domestic law approaches may be illustrated in reference to the Sex Discrimination Act 1984 (SDA). In an ICCPR analysis the first question one might ask is whether there been differential treatment (i.e. a distinction, exclusion, restriction or preference) on a prohibited ground. This has some parallel with the “less favourable treatment” or “disadvantage imposition” tests in the ss.5(1) and (2) descriptions of sex discrimination in the SDA. The second question in an ICCPR analysis would then be whether the differential treatment (or failure to make appropriate differentiation) is based on reasonable and objective criteria. If yes, there is no discrimination. The SDA does not undertake this enquiry, except partially through the following provisions: s.7B which saves certain s.5(2) conduct from constituting indirect discrimination where it is reasonable in the circumstances (though it does not save any s.5(1) conduct); and s.7D which allows special measures intended to achieve equality. Otherwise the SDA merely offers certain situation-specific exemptions, among which is the section 38 religious schools’ exemption. It is obviously necessary that legislation endeavours to strike such balances but the point of a charter is to provide a mechanism for evaluating its success, measured (usually) by convention standards.
The result of adopting the charter’s definition of “discrimination” (i.e. discrimination as defined e.g. in the SDA) is to conceal any deficiency that may exist in its application in federal legislation, as compared with ICCPR standards. In particular, it prevents an analysis of whether that term as used in federal anti-discrimination legislation makes appropriate allowance, as is required under the ICCPR, for differentiation to allow proper scope for freedom of religion (e.g. in operating schools according to a religious ethos) applying criteria that are reasonable and objective and where the aim is to achieve a purpose which is legitimate.
It means that if federal anti-discrimination law is excessively restrictive of freedom of religion by international standards, e.g. by imposing anti-discrimination prohibitions without legitimate relief (applying reasonable and objective criteria etc), the charter is prevented from detecting this. In any balancing exercise, the charter meaning of “discrimination” is privileged, for no accountable reason.
The suggestion is unavoidable that the list of rights chosen for this charter is the product of cherry picking of provisions from other Australian charters, downgrading some rights while elevating others, and selectively deleting crucial texts within certain rights (notably in the freedom of religion). If a charter in this form were to be enacted, the task of deciphering the protection attributable to rights would be near impossible, particularly since Parliament must be taken to have intended to depart deliberately from the clear objective standards expressed in the ICCPR and ICESCR. In the charter, fundamental rights are reworded and rearranged, in a codification that departs from decades of guidance from the relevant treaty bodies (especially the Human Rights Committee responsible for the ICCPR). In relation to the charter obligation on public authorities to act rights-compatibly, the risk is particularly wide that functionaries may generate subjective, expectation-based or wishful thinking interpretations of rights with widespread impact. The situation is exacerbated by inadequate transparency, accountability and scrutiny of formal and informal guidance which generates a culture suffusing the public sector environment in the name of charter rights.
The short answer to the hypothetical question (What impact would this charter have if it were in effect now?) is that it would do far less than a charter should do, if it would do anything at all, to correct the removal of religious school exemptions – even if (as we assumed) their removal were contrary to international law. Key elements of freedom of religion and the right to education in the ICCPR and ICESCR are missing or devalued in the charter. At the same time, the obligation on public authorities in the education sector would have an immediate, powerful and enabling influence. It has the capacity to enlarge state interference with the operations of private sector schools, to ill-defined and unstable charter standards. These will direct government departments, and in turn influence the regulation and supervision of schools through new policies, practices, procedures, guidelines and checklists.
One way of assessing whether the proliferation of such materials is likely to produce charter-compatible outcomes, or the opposite, is to require a searchable public register to be maintained of all such materials. This way they may be inspected and any concerns brought to the AHRC’s attention.
(Is it also possible that one day independent schools may themselves be treated as “public authorities,” especially given the intention that human rights should apply broadly beyond government to bodies exercising “functions of public nature.” Considerations relevant to that question are whether the function is conferred under a statutory provision, and whether it is publicly funded to perform the function. An entity might also become a public authority by being declared one by regulation. Independent schools would not only then be bound to enculturate and observe charter standards, but would also be exposed to claims from victims of a charter breach.)
Remedies
The right to an effective remedy is required by article 2(3) of the ICCPR. In principle I support the proposal for remedies in the Position Paper. The right to a remedy would have been of some concern if combined with a more expansive role for courts than is proposed, e.g. if courts were treated as public authorities in their own right, including when making judgments (as in the UK). The concern with that approach is that it would promote the horizontal application of the charter, as courts would be obliged to consider human rights when making decisions in private litigation.
The Position Paper interestingly notes the view that a charter purporting to implement the ICCPR, which does not include a domestic right to a remedy through a direct cause of action “might be inconsistent with the ICCPR” and beyond constitutional power under s.51(xxix). I would suggest that a far greater risk for the charter under the present proposal, with its idiosyncratic restatement of rights and the fundamentally inappropriate use of a blanket “limitations clause” (among other shortcomings, related to implementation) is that it risks not being “reasonably capable of being considered as appropriate and adapted to implementing a treaty to which Australia is a party.”
Is judicial discretion under the charter a problem?
I recall that more than a decade ago there was a great deal of scepticism about human rights charters in some quarters, because of the perceived risks of excessive judicial discretion. I did not share those concerns, at least for an Australian charter in which rights were defined with certainty, and with objectively discernible meaning. The obvious reference point for Australia would be the ICCPR (given Australia’s ratification of the ICCPR). The simplest and most faithful way to incorporate convention rights, with interpretive certainty, would be to do what the UK Human Rights Act 1998 did when “bringing home” European Convention rights. It scheduled all the convention rights which were relevant, in their original form, including their respective terms of limitation or qualification.
I agree that a charter should be powerful and effective. I would enthusiastically support one that was. But I could not remotely support one (like this) in which the fundamental rights which give it the steer, and the principles by which rights may be limited, have undergone such extensive adaptation. This justifiably enliven fears of judicial discretion.
In the 2017 review of Australia’s compliance with the ICCPR UN Human Rights Committee recommended that Australia “adopt comprehensive federal legislation giving full legal effect to all Covenant [ICCPR] provisions across all state and territory jurisdictions. It should also step up efforts to raise awareness about the Covenant and ensure the availability of specific training on the Covenant at the state and territory levels for judges, lawyers, prosecutors, law enforcement officers and public servants and for federal immigration staff.” Why then adopt divergent standards under this charter? Why not at least allow Australian courts to apply the clearly discernible, objective standards of the ICCPR? Why does the AHRC’s proposal go so far in the opposite direction, away from the ICCPR text, and clear Human Rights Committee General Comment guidance?
Straightening out some AHRC rights (mis)descriptions
As we have seen, one advantage of the UK charter over the AHRC’s proposal is definitional precision of the rights which are protected, in clear European Convention text. Some of the rights in the AHRC and the UK models do not even equate, but the Position Paper gives the strong impression that they do. The so-called right “not to be treated in a degrading way” is a good example. The Position Paper does not clarify, as it should, how different this is from anything found in the ICCPR article 7, or European Convention article 3. In fact, it does the opposite. The Position Paper repeated an account from a British Institute of Human Rights publication of the role played by the UK charter concerning care home obligations towards an elderly couple (Muriel and Robert). There the BIHR loosely described the right engaged as their “right not to be treated in an inhuman and degrading way (Article 3, HRA [Human Rights Act])” (p.15). Elsewhere in the same publication the BIHR used more accurate terminology (p.20, “freedom from inhuman and degrading treatment under Article 3”).” No criticism is made of the BIHR here. It could only have in mind European Convention Article 3, headed “Prohibition of torture,” which is in precisely the same terms as ICCPR article 7: “No one shall be subjected to torture or to inhuman or degrading treatment or punishment.” It is one thing to refer to a well-known and incontrovertible right in colloquial form (as the BIHR did); it is quite another to do what the Position Paper does, and propose to enshrine in the charter a right “not to be treated…in a…degrading way,” detached from its context (which includes torture) and elements of force/imposition, and attracting absolute protection by virtue of an extremely high threshold intended for the right as it existed in unadulterated form. The Position Paper offers no explanation for what it has done to this provision (or the others it has transformed). Among possible explanations is that the AHRC is blissfully unaware, or would rather we were not aware, that such alterations matter crucially.
Before moving on I need to correct something I said in my original post, concerning what the Position Paper described as “the right to recognition before the law.” I criticised this as an inappropriate adaptation of the ICCPR article 16 “right to recognition everywhere as a person before the law”. Indeed it is, but it was probably just a typographical error on the AHRC’s part to mention “the right to recognition before the law” (in section 9.5 when discussing the “limitations clause”) since elsewhere the right appears with the inclusion “as a person”, most importantly in the “list of rights for inclusion.”
However, this still does not explain why this right to recognition is combined with equality before the law and freedom from discrimination. It is an independent right, and there to avoid e.g. the situation where lack of legal existence prevents a person asserting their rights within the domestic legal system. It is a necessary prerequisite to the enjoyment of all rights of the individual. It arises in the context of enforced disappearance, human trafficking, being declared legally dead, and (as the Position Paper rightly acknowledges in the context of the participation duty) the legal capacity of those with disabilities. It begs the question what interpretive significance is to be attached to adding this particular right in a clutch of discrimination-related rights, when this particular one has autonomous value.
Conclusion
It is quite understandable that a charter based on a particular convention may adapt the original text, in order to de-gender it or give it a less archaic rendering, but this proposal (like state and territory charters) goes radically further. The core rights chosen for the charter are all based on the rights in the ICCPR, plus a small number taken from the ICESCR.
A simple comparison between the rights in the charter and those taken from the ICCPR/ICESCR reveals that the charter has so brutalised the original texts as to make the task of interpreting the charter virtually impossible. The charter has selectively deprived key components of some rights of all protection; it has elevated some rights over others; numerous provisions have lost all meaning in the process of adaptation; and the single “limitations clause” adds a layer of complexity across the entire scheme which has more ramifications than the Position Paper has explored.
This federal charter would have immense power to bring about change in the human rights culture in Australia, through the obligation on public authorities to act human rights-compatibly. However, it could not be depended on both in delivering the right outcomes, and in avoiding errant ones. It is not even crafted by reference to the conventional standards. In all likelihood it will be rolled out principally through bureaucratic production of policies, guidelines and advice, quite possibly asserted as if these materials had the force of law but without clear mechanisms that would ensure visibility, accountability or compliance with international human rights convention standards.
I would have no difficulty supporting a charter based on the UK model (even with Australian constitutional law constraints), particularly in its adoption of convention rights law, lock stock and barrel, including convention-applicable terms of limitation and qualification.
The proposed federal charter prompts a range of questions. In this process, how much did the AHRC have regard to its own responsibilities under the Australian Human Rights Commission Act 1986 in reference to the ICCPR, as one the few government bodies responsible for upholding its standards? How acclimated is the AHRC to ICCPR standards (in its recent submission to the Religious Discrimination Bill inquiry it pushed the heresy that religious bodies do not warrant human rights protection because this is only available to protect “the rights of individuals, that is, humans”)? How much is this proposal simply a wish list posting-board, as opposed to the AHRC’s careful curation of feedback with conscientious regard for the ICCPR? Will the charter substitute for ICCPR standards so that they are consigned to obscurity? Finally, the most important question, especially in view of the AHRC’s charter empowering of public authorities, was posed 2000 years ago in the Satires of Juvenal: “Who will watch the watchmen?;” and by John Stuart Mill’s dad when lobbying in support of electoral reform in the mid-1830s “Who is to guard us from the guardians?”