Constitutional cases October 2022

Publication date: 20 October 2022

Constitutional cases October 2022 [PDF 1.4MB]
In this issue

High Court holds section 36B of the Australian Citizenship Act 2007 (Cth) to be invalid

By a 6:1 majority, the High Court (Kiefel CJ and Keane, Gleeson, Gageler, Gordon and Edelman JJ; Steward J dissenting) held that s 36B of the Australian Citizenship Act 2007 (Cth) (Citizenship Act) is invalid because it infringes Ch III of the Constitution. Chief Justice Kiefel and Keane and Gleeson JJ, together with Gageler, Gordon and Edelman JJ, who each wrote separate judgments, found that s 36B, which empowered the Minister for Home Affairs to determine that a person ceases to be an Australian citizen if satisfied that they engaged in certain terrorist conduct which repudiated their allegiance to Australia, invalidly reposes the exclusively judicial function of punishing criminal guilt in the Minister.

Alexander v Minister for Home Affairs & Anor
High Court of Australia, 8 June 2021
[2022] HCA 19


Section 36B was introduced into the Citizenship Act by the Australian Citizenship Amendment (Citizenship Cessation) Act 2020 (2020 Amending Act).

Section 36B(1) relevantly empowered the Minister to determine in writing that a person aged 14 or older ceases to be an Australian citizen if the Minister is satisfied that:

  • the person engaged in the conduct specified in s 36B(5) while outside Australia (s 36B(1)(a))
  • the conduct demonstrates that the person has repudiated their allegiance to Australia (s 36B(1)(b))
  • it would be contrary to the public interest for the person to remain an Australian citizen (s 36B(1)(c)), based on the factors in s 36E(2).

The conduct specified in s 36B(5) incorporated various terrorist activities, including, relevantly for present purposes, 'engaging in foreign incursions and recruitment' (s 36B(5)(h)). It may have occurred any time after 29 May 2003.

The words and expressions in s 36B(5) were given the same meaning as in corresponding offence provisions in the Criminal Code (Cth) but did not include the fault elements that applied in relation to those offences (s 36B(6)). The offence corresponding to s 36B(5)(h) was the offence in s 119.2 of the Criminal Code of entering and remaining in a 'declared area'.

The Minister could not make a determination under s 36B(1) if the result would be to render the person stateless (s 36B(2)).

The 2020 Amending Act also inserted s 36D into the Citizenship Act. Section 36D(1) empowers the Minister to determine in writing that a person ceases to be an Australian citizen if, amongst other factors, the person has been convicted of certain terrorist offences and sentenced to a period of imprisonment of at least 3 years. Together, ss 36B and 36D replaced the former 'operation of law' provisions ss 33AA and 35, which operated to cease a person's citizenship automatically (that is, without a determination being made by the Minister) if they engaged in certain terrorist conduct or were convicted of a terrorism offence.

The plaintiff

Mr Alexander was born in Australia on 5 August 1986. He acquired Australian citizenship at birth pursuant to s 10(1) of the Citizenship Act as it then stood. Both his parents were Turkish citizens and, as a result, he also acquired Turkish citizenship at birth under Turkish law.

In April 2013, Mr Alexander departed Australia for Turkey. Australian Security Intelligence Organisation (ASIO) reporting from June 2013 indicated that he subsequently travelled to Syria, his travel facilitated through a network developed by a convicted terrorist Hamdi Alqudsi. Mr Alexander was reported to be part of a group that was taken to Syria by a senior Syria-based Australian member of the Islamic State. At this time, the Islamic State had been designated in the Criminal Code as a terrorist organisation. ASIO later reported that it was 'likely' that Mr Alexander had joined Islamic State by August 2013 and that he had 'likely engaged' in foreign incursions and recruitment by entering or remaining in al-Raqqa Province in Syria on or after 5 December 2014. At that time, al-Raqqa Province was a 'declared area' for the purposes of s 119.2 of the Criminal Code.

In 2017, Mr Alexander was captured in Syria by Kurdish militia. He was transferred into the custody of Syrian authorities and, in 2019, convicted of terrorist offences by a Syrian court. He has remained imprisoned in Syria ever since.

On 2 July 2021, the Minister determined under s 36B(1) that Mr Alexander ceased to be an Australian citizen. The determination was based (amongst other things) on the Minister's satisfaction that Mr Alexander had engaged in foreign incursions within the meaning of s 36B(5)(h) and, in doing so, repudiated his allegiance to Australia.

The High Court's decision

On 23 July 2021, by his litigation guardian, Mr Alexander commenced proceedings in the High Court against the Minister and the Commonwealth, arguing that s 36B is constitutionally invalid because:

a. it is not supported by a head of Commonwealth legislative power in its operation in respect of the plaintiff
b. it reposes in the Minister the exclusively judicial function of punishing criminal guilt
c. it is inconsistent with asserted implied limitations on Commonwealth power preventing the involuntary deprivation of citizenship and/or disenfranchising the plaintiff otherwise than for a substantial reason.

The High Court delivered 5 separate judgments (Kiefel CJ, Keane and Gleeson JJ; Gageler J; Gordon J; Edelman J; Steward J dissenting) primarily directed to the question of whether s 36B was supported by the aliens power and whether it contravened Ch III. Mr Alexander's arguments regarding the asserted implied limitations were dismissed without detailed consideration by any member of the Court except Steward J, who dismissed them based on very brief reasons ([45]-[46], [132], [186], [316]-[324]).

The aliens power supports s 36B in its operation in respect of the plaintiff

Each member of the High Court except Gordon J, who did not finally decide the point ([132]), considered that s 36B was supported by s 51(xix) of the Constitution. The reasons of Kiefel CJ and Keane, Gleeson and Gageler JJ differed in their approach to those of Gordon and Edelman JJ, reflecting pre-existing divisions in the Court regarding the scope and interpretation of the power. Justice Steward's analysis appeared to align at least in several respects with Edelman J's.

Parliament can decide who has the status of 'alien': Chief Justice Kiefel, Justices Keane and Gleeson, and Justice Gageler

Chief Justice Kiefel and Keane and Gleeson JJ, with whom Gageler J in substance agreed ([98]), held that s 51(xix) enables Parliament to create a 'status of citizenship' that allows for 'the exclusion of persons from the membership of the body politic' ([63]).

The joint judgment emphasised that s 51(xix) supports Commonwealth laws determining 'who shall have the status of Australian citizenship' and providing that 'persons who do not share that status are aliens' ([34]). Because s 51(xix) thus supports the provisions of the Citizenship Act conferring citizenship and constituting the source of a person's rights as a citizen, it must also support a law which limits those rights, including by establishing circumstances in which they will be lost ([38]). Furthermore, as citizenship is a status of 'reciprocal rights and obligations', s 51(xix) must be taken to support a law that treats voluntary conduct demonstrating a repudiation of allegiance to Australia as an implied renunciation of citizenship, much as it would support a law enabling a person to voluntarily renounce their citizenship ([50]).

Their Honours considered that, in light of the threats posed by terrorism, it was open to Parliament to treat a person who engaged in the conduct specified in s 36B(5)(h) as having done so ([62], [63]). To withdraw citizenship from a person who voluntarily engaged in such conduct 'cannot be said to pursue an eccentric understanding of the meaning of "aliens" in s 51(xix) of the Constitution' ([63]).

Parliament can withdraw non-alien status in limited circumstances: Justices Gordon and Edelman

Consistent with her Honour's position in other recent cases on s 51(xix), Gordon J emphasised that the constitutional meaning of 'alien' is not fixed by the statutory concept of 'citizen' ([133]-[134]). Noting that s 51(xix) will not support a law which applies to persons who 'could not possibly answer the description of aliens in the ordinary understanding of the word' (the 'Pochi limit' (Pochi v Macphee (1982) 151 CLR 101) ([136]), Gordon J framed the Commonwealth's position on the 'irreducible minimum' for that category of non-aliens as having 'changed without explanation over the last two decades'. It had shifted from persons born in Australia to Australian parents (or perhaps Australian permanent residents) to a much narrower category of 'persons who were born in Australia, to two Australian persons, who are not citizens of any other country, and who have not renounced or repudiated their allegiance to Australia' ([151]).

However, Gordon J held that, in specific circumstances, s 51(xix) will support laws withdrawing the status of non-alienage from non-aliens. Those circumstances include (but may not be limited to) changes in sovereign territory or identity; and, relevantly, where the person has renounced their allegiance to Australia, expressly or impliedly, by engaging in specified conduct ([139]-[141]).

In Gordon J's view, s 36B was clearly within the scope of the aliens power on this basis, at least insofar as it is directed at conduct which 'plainly constitutes renunciation of allegiance'. That conduct included 'spying or fighting' for an enemy state at war with Australia and the closely related conduct involving 'service in the armed forces of a declared terrorist organisation' ([154], [156]). It was unnecessary to decide whether s 36B was supported by s 51(xix) in all of its possible operations ([156]).

Similarly, Edelman J held that s 51(xix) supports laws regulating some circumstances in which a person can cease to be a member of the political community of the Australian body politic (that is, become an alien) ([185], [229]). His Honour agreed with Gordon J that, amongst others, these circumstances included where the person's conduct is so 'wrongful and extreme' that it can be judged to be 'inconsistent with continuing membership of the political community' ([185], [229]).

Justice Edelman held that the conduct caught by ss 36B(1)(a) and (5), involving matters related to terrorism, foreign incursions and services in enemy armed forces, provided a 'baseline' for the extreme nature of conduct necessary to repudiate a person's allegiance. However, his Honour found that conduct falling within these provisions would not in all cases be wrongful and extreme enough to meet that threshold. Section 36B's validity as a law under s 51(xix) in this sense hinged on 'the extreme wrongdoing' that is required by, and inherent in, the notion of 'repudiation of allegiance', required to be determined by the Minister in s 36B(1)(b) ([234]). As noted below, Steward J also placed the same importance on s 36B(1)(b).

Justice Edelman also generally agreed with Steward J that 'repudiation of allegiance' required no subjective intention to sever the bond of membership ([233]). All that was required was actions or steps that were objectively 'indelibly inconsistent' with membership of the Australian political community, including actions and steps of the kind set out in Steward J's reasons at [286] (and quoted below) ([233]).

In addition, Edelman J's judgment includes detailed explanation of why Mr Alexander could not properly have been considered an alien from birth. Like Gordon J, his Honour emphasised that the constitutional meaning of 'alien' must not be conflated with the statutory concept of 'citizen' ([195]). His Honour explained at some length why previous High Court authority on s 51(xix) is, in his Honour's view, erroneous ([183]-[225]), rejecting the idea supported by that authority ([202]) that the aliens power encompasses the power to legislate in respect of persons who are not aliens but 'capable of being treated as aliens' – a category which includes, on the Commonwealth's view, dual citizens and people born outside Australia or in Australia to non-Australian parents ([196]-[202]). Justice Edelman stressed that the result of that approach would be that more than half of the Australian population, including Mr Alexander, could be 'treated as an alien' ([200], [226]). He concluded that Mr Alexander, born in Australia to parents who were, at the time, permanent members of the Australian community, could not possibly be treated as within the reach of s 51(xix) at the time of his birth ([226]).

Reasons of Justice Steward

Justice Steward's starting point was that s 51(xix) authorises Parliament to make laws which prescribe the conditions on which citizenship 'may be acquired and lost' (quoting Re Minister for Immigration and Multicultural and Indigenous Affairs; Ex parte Ame (2005) 222 CLR 439 at [35]) ([286]).

However, that power is not 'unfettered'. Membership of the Australian body politic is 'inextricably bound up' with the concept of allegiance to Australia, meaning that the 'power to denationalise' in s 51(xix) must (as Edelman J agreed at [232]) be limited to laws 'that recognise and accept' a loss of citizenship arising from voluntary actions or steps that are 'indelibly inconsistent with that allegiance and with membership of that community' ([286]). This, in Steward J's view, was consistent with how the law of 'denationalisation' had developed in the law of the United Kingdom and United States prior to federation; and with a historical consideration of the aliens power (canvassed in his Honour's judgment at [259]-[275] and [276]-[284] respectively).

The actions and steps that might be 'indelibly inconsistent' with allegiance would include ([289]):

actions which seek to destroy or gravely harm the fundamental and basal features of the nation guarded by its Constitution, such as representative democracy and the rule of law, and actions directed at overthrowing state institutions where such conduct amounts to a clear rejection of allegiance to Australia. Terrorist attacks might also be included.

No subjective intention to repudiate allegiance was required ([287]).

Justices Steward and Edelman agreed that whether s 36B was supported by s 51(xix) accordingly 'hinged' on the requirement in s 36B(1)(b) that the Minister be satisfied that the person had repudiated their allegiance to Australia ([297]). The existence of that requirement (and the requirement in s 36B(1)(c) that it would not be in the public interest for the person to remain an Australian citizen) meant, for Steward J, that a person who had engaged in blameless conduct could not fall within the scope of s 36B ([311]-[312]).

In Steward J's view, it was also the case that, while it may be possible to enter innocently into a declared area, it is possible to do so with 'evil intent' ([314]). The voluntary act of entering and remaining in al-Raqqa province in 2014 could in itself 'well justify' a conclusion that the person had 'repudiated their allegiance to Australia' and that it was in 'the public interest' for that person not to remain a citizen ([315]). Section 36B was, for all of these reasons, a law that was authorised by s 51(xix) of the Constitution.   

Notably, in addition to the above analysis, at [291] Steward J expressly agrees in his reasons with passages from Edelman J's reasons at [193]-[199] and [225], to the effect that who is an alien for the purposes of s 51(xix) is 'ultimately a matter entrusted to the Court', with the aliens power accordingly having 'a constitutional field of application determined by this Court'. His Honour also adopts Edelman J's analysis regarding a proposed analogy between the immigration and aliens powers (at [204]-[211] of the latter's judgment) and agrees that a person who is unconditionally absorbed into the community should not be considered an alien for the purposes of s 51(xix).

Section 36B infringes Ch III of the Constitution

In contrast to the aliens power question, the various Justices forming the majority were largely aligned in their analysis of why s 36B of the Citizenship Act was inconsistent with Ch III of the Constitution.

Section 36B involves punishing criminal guilt

In Chu Kheng Lim v Minister for Immigration, Local Government and Ethnic Affairs (1992) 176 CLR 1 (Lim), the High Court held that the function of 'adjudging and punishing criminal guilt' is exclusively judicial. Each of the majority judges concluded that s 36B was invalid as conferring an exclusively judicial power on the Minister contrary to the principle in Lim, even though, unlike the law at issue in Lim, s 36B did not impose detention in custody.

The joint reasons of Kiefel CJ and Keane and Gleeson JJ accepted that s 36B might be said to pursue a protective purpose of 'protecting the community' from the risk posed by returning foreign fighters ([75]). However, their Honours found that was not the 'principal purpose of the provision' ([75]). Justice Edelman likewise considered that s 36B's sole purpose could not be said to be 'purely protective' ([247]). Neither Gordon J nor Gageler J appeared to accept that the provision could properly be construed as having a protective purpose at all (see in particular Gageler J's judgment at [99], [120] and [126]).

Rather, the 6 majority judges characterised s 36B in the following ways:

  • Chief Justice Kiefel and Keane and Gleeson JJ (Gageler J agreeing) considered that the principal purpose of s 36B was to impose 'retribution for conduct deemed to be so reprehensible as to be incompatible with the shared values of the Australian community' ([75], [82]). The retribution was meted out in the form of depriving the person of 'the entitlement to be at liberty in Australia' ([74], [82]). The Minister's discretion in meting out the retribution arose upon the Minister themselves being satisfied that conduct elements of an offence had occurred ([86]).
  • In addition to agreeing with the joint reasons, Gageler J considered that s 36B's true purpose was one of 'denunciation and exclusion from formal membership of the Australian community' of persons who had engaged in certain conduct 'limited to criminal conduct found to have been engaged in … in the past'. That purpose could 'only be characterised as punitive' ([119]).
  • Justice Gordon also characterised s 36B's purpose as 'retribution' ([163]). Section 36B effectively authorised the Minister to impose a 'sanction' upon a person for engaging in conduct identified as warranting 'the condemnation of the Australian community' ([164]). It operated with respect to 'identified and articulated wrongdoing', and cessation of citizenship was essentially a consequential step after the Minister's adjudication that the person has engaged in past acts which, if accompanied by fault elements, would have involved criminal guilt ([165]).  
  • Justice Edelman considered that s 36B 'has a punitive character with a purpose to sanction particular conduct' ([247]).

The statement of purpose in s 36A of the Citizenship Act supported a construction of s 36B's purpose as punitive ([80], [120], [163], [251]).

The majority also emphasised the harsh consequences of the exercise of the Minister's power in s 36B ([77], [166], [248]). Justice Gordon characterised the consequences as 'significant' ([166]), with Edelman J referring to them as 'extreme' ([248]). Chief Justice Kiefel and Keane, Gleeson, Gageler and Edelman JJ expressly distinguished s 36B from laws dealing with the revocation of licences or other statutory privileges, upheld as compatible with Ch III in arguably analogous cases ([77], [110], [248]). Justice Edelman considered that the consequences imposed by s 36B were 'in a wholly different league' to the revocation of statutory rights with which those cases were concerned ([248]). Justice Gageler considered that to compare s 36B to those cases was to 'stretch the concept of protection to breaking point' ([110]).

Finally, the joint reasons and both Gordon and Edelman JJ pointed to the long-held understanding of exile, banishment and other similar acts as forms of punishment as a factor supporting a characterisation of s 36B as punitive ([75], [167]-[172], [250]).

Consideration of s 36D

Aside from the above factors, the joint reasons of Kiefel CJ, Keane and Gleeson JJ (with which Gageler J agreed in this respect) also placed significant weight on the distinction between s 36B, and s 36D of the Citizenship Act ([70]). Their Honours noted that the power in s 36D only arises where a person has been tried, convicted and sentenced by a court in an 'orthodox exercise of judicial power' ([86], [93]). By contrast, s 36B, despite having the same purpose as s 36D and resulting in the same outcome, turned on the Minister's own discretion and a process of ministerial fact-finding which had none of the protections of a criminal trial ([86]-[87]). This 'incongruity' was not dispelled by the possibility that a determination under s 36B could subsequently be revoked under s 36H or that the revocation provisions ss 36J and 36K may be engaged ([87]).

Ultimately, the joint judgment noted that the validity of s 36D was 'not in issue in this case' ([80]) and framed their ultimate conclusion on the Ch III question in terms that seemed to deliberately avoid encompassing that provision, stating ([96]):

The power to determine the facts which enliven the power to impose [the punishment of depriving a person of the entitlement to enter and live at liberty in Australia] is one which, in accordance with Ch III of the Constitution, is exercisable exclusively by a court that is a part of the federal judicature.

Justice Gordon noted that, unlike s 36B, s 36D 'conferred power to cease citizenship by reference to the "factum" of an earlier adjudication by a court of a person's criminal guilt'. Her Honour said it was unnecessary to consider whether s 36D would be invalid on the basis that it authorises the imposition of a 'new or additional punishment' for a person committing an offence ([174]).

Justice Steward: s 36B is not an invalid conferral of judicial power

In dissent on this point, Steward J observed, at the outset, that 'denationalisation' could be penal in nature in some circumstances ([326]). However, the power exercised by the Minister in cancelling the plaintiff's citizenship in this case was not judicial power ([332]) for 3 reasons:

  1. It has never been an essentially judicial function to make orders which denationalise a person ([332]-[335]).
  2. The task of the Minister under s 36B is not to determine guilt or innocence or whether the conduct in s 36B(5)(a)-(h) constitutes the commission of a crime. The Minister's task is a distinctly different one: of determining whether the conduct repudiates the person's allegiance and whether it is in the public interest for the person to retain their citizenship ([336]).
  3. The purpose of s 36B is not to punish: it serves a legitimate, non-punitive purpose of recognising a person's repudiation of their allegiance and imposing a consequence ([337]).

The ability of courts to ensure that the power in s 36B is exercised lawfully was expressly preserved by the Citizenship Act, including by the note to s 36H and by s 36K. This safeguard supported the conclusion that the power under s 36B was not judicial ([341]-[342]).

Justice Steward emphasised that s 36B applies not to 'victims' but to people who are 'in substance and as a matter of practical effect repudiators of Australia' ([338]).

The Commonwealth's legal team

The AGS Constitutional Litigation Unit (Simon Thornton, Danielle Gatehouse, Liam Boyle and Tasha McNee) acted for the Commonwealth and the Minister for Home Affairs, with the Solicitor-General, Dr Stephen Donaghue KC, Perry Herzfeld SC, Julia Watson, and Luca Moretti as counsel.

The text of the decision is available at:

  Back to top


Thornton, Simon

Senior Executive Lawyer

Boyle, Liam

A/g Senior Executive Lawyer

State tribunals are denied jurisdiction if a party raises a genuine 'federal' claim

The High Court unanimously (7:0) held that the merits of a federal claim or defence are irrelevant in determining whether a case is a matter of a kind described in s 76(i) or s 76(ii) of the Constitution (except in the case of truly nonsensical constitutional arguments). So long as the claim or defence is genuinely raised and not incapable on its face of legal argument, a state tribunal will be denied jurisdiction to exercise judicial power in the matter before it (in line with the High Court's decision of Burns v Corbett (2018) 265 CLR 304 (Burns)) if a party raises an issue that falls within a subject of federal jurisdiction.

Citta Hobart Pty Ltd v Cawthorn
High Court of Australia, 4 May 2022
[2022] HCA 16


Mr Cawthorn uses a wheelchair and lodged a complaint with the Tasmanian Anti-Discrimination Tribunal relating to Parliament Square - a major building development in Hobart. One of the 3 entrances to Parliament Square has stairs and is not wheelchair accessible. Mr Cawthorn argued that the developer and owner (the builders) discriminated against him and other persons reliant on wheelchairs contrary to s 16(k) of the Anti-Discrimination Act 1998 (Tas) (the Tasmanian Act). The builders claimed that the development complies with the Commonwealth Disability (Access to Premises – Buildings) Standards 2010 (Cth Standards) and that if the Tasmanian Act makes their conduct unlawful it is inconsistent with the Disability Discrimination Act 1992 (Cth) (DDA) and invalid under s 109 of the Constitution.

The Tribunal decided it did not have jurisdiction to resolve the complaint because of the High Court's decision in Burns that state tribunals that are not courts cannot exercise judicial power in matters falling within s 75 or s 76 of the Constitution (federal matters). The Tribunal acknowledged that a federal claim or defence raised 'colourably', in the sense of being made solely for the purposes of fabricating jurisdiction, would not deprive it of jurisdiction but concluded that the builders' claims were not colourable.

Mr Cawthorn appealed the Tribunal's decision. The Full Court of the Tasmanian Supreme Court held that, while the builders' s 109 argument was made in good faith and was therefore not colourable, it was 'misconceived'. The Tasmanian Full Court held that the Tasmanian Act was not inconsistent with the DDA and that the matter was within the Tribunal's jurisdiction.

The High Court granted the builders special leave to appeal. The 2 grounds of appeal were, in summary:

  1. whether the Tribunal would have been exercising judicial power contrary to Ch III if it determined the complaint and had therefore rightly dismissed the complaint
  2. whether the Tasmanian Act was inconsistent with the DDA and the Cth Standards within the meaning of s 109 of the Constitution.

The Attorney‑General of the Commonwealth intervened to put submissions in support of the appellant on both grounds.

The Australian Human Rights Commission (AHRC) was also granted leave to intervene as amicus curiae.

The High Court's decision

The High Court delivered 2 judgments – a plurality judgment of Kiefel CJ and Gageler, Keane, Gordon, Steward and Gleeson JJ; and the separate reasons of Edelman J.

The Court unanimously allowed the appeal but held that it was not necessary or appropriate to resolve the s 109 inconsistency argument in ground 2, despite the parties making extensive written and oral submission on that ground (plurality at [9]; Edelman J at [86]).

The plurality judgment

The plurality characterised the specific issue presented by ground 1 as 'the exclusion from State jurisdiction conferred on a State tribunal of matters, referred to in s 76(i) or s 76(ii) of the Constitution, arising under the Constitution or arising under laws made by the Commonwealth Parliament' ([3]), the question being: 'how is a justiciable controversy to be identified as a matter answering one or other of those descriptions?' ([3]).

The plurality held that on any view the Tasmanian Full Court 'would have been wrong to discern appealable error on the part of the Tribunal' ([9]). This was because ([10]):

The constitutional defence was genuinely raised in answer to the complaint in the Tribunal and was not incapable on its face of legal argument. That being so, the complaint and the defence together formed parts of a single justiciable controversy comprising a matter within the description in each of s 76(i) and s 76(ii) of the Constitution. The Tribunal was on that basis correct to order that the complaint be dismissed for want of jurisdiction.

In reaching that conclusion the plurality reasoned as follows.

The Tribunal would exercise judicial power

The plurality first considered a 'threshold issue' raised by the AHRC (and adopted by Mr Cawthorn and the Attorney‑General of Queensland). This issue was 'whether the jurisdiction conferred on the Tribunal by the State Act to hear and determine a complaint of discrimination referred to it in truth involves the exercise of judicial power' ([11]).

Given that the decision in Burns limits the exercise of judicial power by tribunals, the AHRC contended that that the Tasmanian Act 'on its proper construction makes the binding effect of the Tribunal's order contingent on registration of that order in the Supreme Court', but the prior step of the Tribunal hearing and determining matters was an exercise of non-judicial power ([13]). On this argument, the Tasmanian Act would comply with the limit in Burns by 'construing the provision permitting registration of the Tribunal's order to exclude an order that, if registered, would result in an exercise of judicial power with respect to a [federal] matter' ([13]) - that is, the Tribunal could have heard and determined the matter, even though a federal defence had been raised, so long as s 90 of the Tasmanian Act was construed so that it did not include registration of orders relating to federal matters.

The plurality rejected this argument. Their Honours noted that the same argument had been put and rejected in Commonwealth v Anti-Discrimination Tribunal (Tas) (2008) 169 FCR 85. The plurality pointed out that the 'essential flaw in the AHRC's argument is that it confuses the order of the Tribunal with the mechanism for enforcement of that order' ([15]). The Tasmanian Act makes clear that an order made by the Tribunal on finding a complaint established 'takes immediate effect as an order with which the person to whom it is directed is bound to comply' ([16], emphasis added). Thus, registration of the order in the Supreme Court is not a precondition which makes the order binding ([16]). The Tribunal would therefore be exercising judicial power if it heard and determined the matter ([12]).

The Tribunal has state judicial power to determine the limits of its state jurisdiction

All of the parties and interveners accepted that the Tribunal has a duty and concomitant authority to ensure that a complaint referred to it is and remains within its jurisdiction.

The plurality emphasised that a tribunal's power to ensure it remains within jurisdiction is inherently neither judicial nor non-judicial but rather 'takes its nature from the nature of the power to which it is incidental' ([22]). Thus, a court in which judicial power is invested has jurisdiction to decide its own jurisdiction and can do so exercising judicial power, while a tribunal that is not a court has authority to determine 'in the sense of forming an opinion' the limits of its own jurisdiction ([24]). The tribunal does 'reach a conclusion having legal effect' but forms an opinion which then informs the tribunal's conduct ([24]).

Thus, a 'State tribunal must be taken to have incidental jurisdiction to determine whether the hearing and determination of a particular claim or complaint would be within the legislated limits of its State jurisdiction' ([25]). This power, the plurality held, is judicial in nature because the Tribunal's power is gained through a conferral of state judicial power ([26]). To the extent that the decision in Sunol v Collier (2012) 81 NSWLR 619 may be understood to have held otherwise, the plurality held that the decision should not be followed ([26]).

In this case, the plurality held that 'the opinion formed by the Tribunal that the complaint referred to it was beyond its jurisdiction to hear and determine was accordingly a judicial opinion and the order made by the Tribunal dismissing the complaint for want of jurisdiction was an order made in the exercise of State judicial power' ([28]).

Determining the limit on the Tribunal's state jurisdiction

The plurality held that determining whether the Burns limit applies to exclude jurisdiction conferred on a non-court state tribunal by a state law requires the same assessment as determining the existence and scope of a federal matter conferred on the High Court or a federal court, or invested in a court of the state under s 77(iii) of the Constitution ([30]). In each case it is the 'assertion' of a federal matter which will be relevant, and an 'objective' assessment is required ([31], [33]).

Mr Cawthorn argued that the Tasmanian Full Court's judgment could be supported on the basis that the Tribunal should have adjourned the proceeding, once the s 109 issue was raised, and had that argument determined by the Supreme Court exercising federal jurisdiction ([32]).

The plurality rejected this argument because the allegation that the appellants had discriminated against the respondent on the ground of disability, and the appellants' assertion, by way of defence, that the Tasmanian Act was inoperative by operation of s 109, were inextricably part of one justiciable controversy ([33]). The resolution of the defence was essential to the determination of the claim. They would remain one controversy even if the defence were later considered and rejected by the Supreme Court ([33]).

The irrelevance of the merits of the constitutional defence

The plurality next considered whether, in order to give rise to a matter described in s 76(i) or s 76(ii) of the Constitution, 'the constitutional defence asserted by the appellants needed to meet some threshold degree of arguability and if so, what that threshold was' ([34]). The Tasmanian Full Court had held that because the claim was 'misconceived' the Tribunal could have heard and determined the matter. In doing so the Tasmanian Full Court had essentially considered the merits of the constitutional defence.

The plurality held that 'it is enough that the claim or defence be genuinely in controversy and that it give rise to an issue capable of judicial determination' or, put another way, that it is 'genuinely raised and not incapable on its face of legal argument' ([35]). Their Honours stated that this test is what is meant when a claim or defence is said to be 'unarguable' or 'colourable' in that it is made for the purpose of 'fabricating jurisdiction' ([36]).

However, the prospects of success of a claim (if determined on its merits) does not form part of this assessment ([37]). The plurality rejected a submission of the Commonwealth (supported by other intervenors), which the Court characterised as an argument that a 'matter' under s 76(i) or s 76(ii) 'must meet a threshold of arguability consistent with the raising of the claim or defence in a court not amounting to an abuse of process of that court' ([41]). The Attorney‑General of the Commonwealth had proposed a test that the claim or defence be neither colourable nor so clearly untenable that it could not possibly succeed. Their Honours said that to introduce the latter element would 'blur the distinction between the existence of jurisdiction and the exercise of jurisdiction' ([42]). For a state tribunal, the plurality held, such an exercise would 'inevitably involve the tribunal being drawn down the forbidden path of judicially determining the merits of a matter within a description in s 76(i) or s 76(ii) of the Constitution' ([42]). However, the plurality made it clear that the rejection of a threshold of arguability was not a suggestion that 'an incomprehensible or nonsensical claim or defence' is capable of giving rise to a matter within the description of s 76(i) or s 76(ii) ([43]).

In the circumstances of this case, the plurality noted that there was no suggestion that the builders' constitutional defence in the Tribunal 'was not genuinely raised or is so incoherent as to be insusceptible of judicial determination on those merits' ([45]). Accordingly, the Tribunal was correct that it did not have jurisdiction to hear and determine the entire claim ([46]).

Justice Edelman's judgment

Justice Edelman agreed with the orders made by the plurality; however, his Honour's reasoning and conclusions differed somewhat.

Justice Edelman agreed that the Tribunal would have exercised judicial power if it had resolved the dispute (contrary to the AHRC's argument) because the orders of the Tribunal impose a binding norm requiring things to be done ([56]). The question of whether a matter arises under federal jurisdiction does not vary depending on whether the question is asked by a court or a tribunal – the scope of the subject matter remains the same ([61]).

Justice Edelman also agreed that tribunals have the capacity to consider the legal limits of their jurisdiction. But, in contrast with the plurality, Edelman J held that for a tribunal this determination is anterior to, but is not, an exercise of judicial power ([62]).

There must be a 'real question' as to whether there is a federal matter

Justice Edelman's reasoning differed from the plurality most prominently in propositions 5, 6 and 7. His Honour held that a mere assertion is not enough for a matter to arise; rather, there must be a 'real question' ([67]). A matter will not arise where 'one party raises the issue merely for jurisdictional reasons without any genuine dispute or where the issue is preposterous or manifestly hopeless' ([67]).

Justice Edelman then expanded on the concept of when a 'real question' has been raised by a party. His Honour held that a court may consider whether a 'real question' has been raised through the exercise of its inherent jurisdiction to consider whether it would be an abuse of process to address the question ([69]). Unlike the plurality, Edelman J accepted the Attorney‑General of the Commonwealth's submission that there is an additional threshold so that the claim or defence must not be 'so clearly untenable that it cannot possibly succeed', although his Honour stated that 'manifestly hopeless' was a better descriptor ([73]).

Section 109 issue in this case involved a 'real question' about a matter arising under the Constitution

Justice Edelman characterised Mr Cawthorn's submissions about whether s 109 involved a 'real question' as one solely related to whether the argument was so clearly untenable it could not possibly succeed (or as his Honour preferred, 'manifestly hopeless') ([77]). His Honour accepted that a determination of whether a claim or defence was manifestly hopeless 'could not be entirely independent of an assessment of the "merits" of that issue' ([78]). However, it was his view that this threshold is an 'extreme conclusion' that 'requires the lack of merit to be so obvious, and so apparent' that this was acceptable ([78]). The s 109 issue raised by the builders was not manifestly hopeless on even a brief consideration of the Commonwealth and state legislation ([79]-[81]).

The appeal was allowed and the substantive orders of the Tasmanian Full Court were set aside. In their place, the appeal from the Tribunal to the Tasmanian Full Court was dismissed.

The Commonwealth's legal team

The AGS Constitutional Litigation Unit (Niamh Lenagh-Maguire, Liam Boyle and Aurora Crain) acted for the Commonwealth, with the Solicitor‑General, Dr Stephen Donaghue KC, Frances Gordon and Rachel Amamoo as counsel.

The text of the decision is available at:

  Back to top


Lenagh-Maguire, Niamh

Senior Executive Lawyer

Constitutional Litigation Unit
Boyle, Liam

A/g Senior Executive Lawyer

High Court holds prohibition on use of confusing or misleading party names and logos is valid

By a 4:3 majority, the High Court (Gordon, Edelman, Gleeson and Steward JJ; Kiefel CJ and Keane and Gageler JJ) upheld the constitutional validity of amendments to the Commonwealth Electoral Act 1918 (Electoral Act) which prevent registered political parties from using names or logos on the ballot paper that have words in common with the name of a political party registered earlier in time.

Ruddick v Commonwealth
High Court of Australia, 15 February 2022
[2022] HCA 9


The Electoral Act regulates the registration of political parties. Part XI of the Electoral Act governs the registration of political parties. Where a candidate is endorsed by a political party registered under Pt XI, that political party's registered name and logo may be printed next to the candidate's name on the ballot paper. There is no requirement in the Electoral Act for a registered party to use its name and logo registered under Pt XI in contexts outside of the ballot paper, such as in advertising or campaigning.

Since 2004, ss 129 and 129A have required the Australian Electoral Commission (AEC) to refuse an application for registration if, in the AEC's opinion, the party's proposed name 'so nearly resembles' the name, or an abbreviation or acronym of the name, of another, unrelated political party that it is 'likely to be confused with or mistaken for' that name, abbreviation or acronym (ss 129(1)(d), 129A(1)(c)).

Once a political party is registered (s 133), s 134A sets out the circumstances in which the AEC must deregister it if a political party registered earlier in time (the parent party) objects to the party's name or logo.

Amendments to prevent the use of confusing or misleading party names

Items 7, 9, 11 and 14 in Sch 1 to the Electoral Legislation Amendment (Party Registration Integrity) Act 2021 (Cth) (Amending Act) commenced on 3 September 2021. The items introduced several amendments to Pt XI, with the purpose of 'minimis[ing] the risk that a voter might be confused or potentially misled in the exercise of their choice at an election due to a political party having a registered name or abbreviation similar to that of an unrelated registered political party' (see Explanatory Memorandum, Electoral Legislation Amendment (Party Registration Integrity) Bill 2021, [19]).

Items 7 and 9 amended ss 129 and 129A to impose an additional requirement for registration of a new party: that the name, abbreviation or logo of a party must not, without the consent of the prior registered political party, contain a word that is in the name, or the abbreviation of the name, of a prior registered political party.

Items 11 and 14 amended s 134A to require the AEC to deregister an existing party (the second party) if:

  • the party's name or logo contains a word that is in the name, or the abbreviation of the name, of the parent party
  • the parent party objects to the continued use of the name or logo
  • the second party does not make an application under s 134 for a change of name or logo within one month of being notified of the objection or makes an application and that application is refused.

For the purposes of these new restrictions, grammatical forms and commonly accepted variants of a word, including abbreviations, contractions and alternative forms, must be treated in the same way as a 'word' (new s 129(6), inserted by item 7). However, the restrictions do not apply to words in a party's name that are a function word, a collective noun for people, the name of a country, the word 'country', or a recognised geographical place in Australia, or the word 'democratic' (new s 129(5), also inserted by item 7).

Mr Ruddick challenged the validity of the amendments

The Liberal Democratic Party (LDP) was first registered under the name 'Liberal Democratic Party' in December 2008. It fielded candidates in the 2010, 2013, 2016 and 2019 federal elections under that name. Mr Ruddick was a member and endorsed Senate candidate of the LDP.

On his own behalf and on behalf of the LDP, the plaintiff, Mr Ruddick, challenged each of items 7, 9, 11 and 14 on the basis that the amendments were incompatible with the constitutionally implied freedom of political communication and inconsistent with the requirement in ss 7 and 24 of the Constitution that senators and members of the House of Representatives be 'directly chosen by the people'.

The effect of the Amending Act was that, if the Liberal Party of Australia (Liberal Party) objected to the LDP's ongoing registration under s 134A of the Electoral Act as amended by items 11 and 14 in Sch 1 of the Amending Act, the use of the word 'Liberal' in the LDP's name would lead to the LDP being deregistered if it did not quickly change its name. That would, in turn, prevent Mr Ruddick from having his name associated with the LDP on the ballot paper or at least require him to be associated with the LDP's new name on the ballot paper.

After the matter commenced, but prior to the hearing, the Liberal Party did in fact object to the LDP's ongoing use of the name 'LDP' under s 134A of the Electoral Act as amended ([137]). The LDP accordingly applied under s 134 to change its name to 'Liberty & Democracy Party' ([144]). At the time of the hearing, the AEC had not yet dealt with the name change application.

Mr Ruddick admitted that, at least at the 2013 election for the Senate in New South Wales, some voters who intended to vote for the Liberal Party instead unintentionally voted for the LDP because they were confused as to the party affiliation of LDP candidates. (In that election, the LDP had appeared first on the New South Wales Senate ballot paper and substantially to the left of the Liberal Party, which appeared in the 25th column. The LDP's candidate, Mr David Leyonhjelm, received 9.5% of votes for the New South Wales Senate, far exceeding the party's vote in any other New South Wales Senate election ([117])).

The High Court's decision

The High Court delivered 4 sets of reasons: joint reasons of Gordon, Edelman and Gleeson JJ; separate concurring reasons of Steward J; a joint dissenting judgment of Kiefel CJ and Keane J; and separate dissenting judgment of Gageler J.

The problem of voter confusion and the purpose of the amendments

A majority of the High Court accepted that there was evidence of the phenomenon of voter confusion and that Parliament had endeavoured to respond to that problem. Justices Gordon, Edelman and Gleeson JJ (Steward J relevantly agreeing at [174]) outlined the background to the impugned amendments, including:

  • evidence that voter confusion between the LDP and the Liberal Party and other parties, including the Democratic Labor Party and the Labor Party, was ongoing ([115], [124])
  • Mr Ruddick's admission about voter confusion at the 2013 Senate election ([122], [140]) and a separate radio interview from 2013 in which Senator Leyonhjelm himself had admitted that some electors would have voted for the LDP in that election because they mistook the LDP for the Liberal Party ([108], [118])
  • the previous amendments to the Electoral Act designed to avoid confusion of parties by electors, passed in 1983, 2004, and 2016 ([113], [114], [123]).

Against this background of voter confusion, and 'evolving legislative responses to minimise confusion' ([130]), their Honours concluded:

  • First, it was an 'available inference' that confusion caused by the similarity between the Liberal Party's name and the LDP's name was responsible for the LDP's result in the 2013 New South Wales Senate election ([120], [122]). Their Honours expressly rejected the idea, advanced by Mr Ruddick, that the confusion arose solely from the order of the parties on the ballot paper and not from any similarity in their names ([134]).
  • Second, the impugned amendments had a legitimate purpose of 'reducing the risk of voter confusion' ([133]-[134]). Their Honours rejected Mr Ruddick's submission that, despite references in the extrinsic materials to a purpose of avoiding confusion, the 'real mischief' and the 'true purpose' of the impugned provisions is 'anti-competitive' (that is, to reduce competition between major parties and minor parties) ([131], [132]). They held ([132]):

    It can be accepted that statements of purpose in an explanatory memorandum or a second reading speech are not conclusive, but it is a significant step to conclude that express statements in such extrinsic materials are a pretence designed to conceal an anti-competitive purpose.

Justice Gageler held that, as a matter of 'logic or linguistics', it did not follow from 'the mere fact that one name contains a word contained in another name' that confusion of those names might occur ([90]); nor did that follow from Australia's 'national experience' ([91]), Mr Ruddick's admission in his pleadings or the facts in the special case.

In Gageler J's view, the picture arising from the admission and the special case was more 'complex' than that ([97]). Mr Ruddick's admission did not go so far as to admit that electors confused the LDP with the Liberal Party merely on the basis of the common word used in their names ([96]); nor did the special case. The highest any of the material in the special case rose was to demonstrate that a real issue can from time to time arise as to whether the particular name of some other party which has the word 'Liberal' in it might be confused with the Liberal Party ([103]). It did not demonstrate that the mere fact that one political party contains in its registered name a word contained in the registered name of another, earlier registered political party in itself generates a meaningful risk of confusion on the part of electors as to the party affiliation of candidates for election ([104]).

To similar effect, Kiefel CJ and Keane J held (at [15]):

Nothing in the facts or opinions stated in the special case demonstrates that a word contained in the name of another, registered, political party of itself gives rise to a real risk of confusion on the part of voters as to the party affiliation of a candidate for election. What they do confirm is what political scientists refer to as "the donkey vote", namely the effect that a party's position on a ballot paper may have on its share of the vote. That in turn may be thought to be a consequence of compulsory voting. But the special case does not demonstrate that the potential for confusion is attributable to the use of a word in a party's name alone.

No basis to consider the validity of items 7 and 9

The joint judgment emphasised that Mr Ruddick was not permitted to 'roam at large' over the impugned provisions or to advance grounds of challenge other than those which bore on the validity of the impugned provisions in their application to him ([144]). In that context, the majority found that Mr Ruddick had failed to put any facts before the Court against which to test the validity of items 7 and 9 in Sch 1 to the Amending Act ([145]). The validity of those items was an 'abstract and hypothetical question', 'outside the scope of the suit', and there was no basis to determine the validity of items 7 and 9 ([145]).

In contrast, Gageler J, with Kiefel CJ and Keane J agreeing, would have held that all of the impugned provisions of the Amendment Act are invalid.

Threshold for establishing that items 11 and 14 are inconsistent with 'direct choice'

Addressing the requirement, derived from ss 7 and 24 of the Constitution, that members of the Parliament be 'directly chosen by the people', Gordon, Edelman and Gleeson JJ characterised the 'threshold issue' as whether the plaintiff could establish that items 11 and 14 imposed 'some burden' on the franchise ([150]). Their Honours noted that, in contrast to cases like Murphy v Electoral Commissioner (2016) 261 CLR 28 that were concerned with a burden on universal adult suffrage, this case was concerned with a burden on 'the people's' ability to make an 'informed choice', and 'convey and receive opinions, arguments and information concerning matter intended or likely to affect voting' ([151]).

The plurality rejected as 'inapposite' the Commonwealth's contention that there is a distinction between laws which impose an impediment on universal adult suffrage and laws which regulate the electoral system without excluding any electors from the franchise ([151]). However, like the restriction implied by the notion of 'the people', the restriction imposed by the requirement of 'choice' must not be applied in an 'overbroad' manner which would 'fail to respect the constitutional design of leaving to Parliament the choice of how to legislate for every aspect, except the bare foundations, of the electoral system' ([152]).

Threshold for establishing that items 11 and 14 are incompatible with the implied freedom

As for the implied freedom of political communication, their Honours noted that, provided a law has a legitimate purpose, a threshold issue for determining the validity of a law said to infringe the implied freedom is whether the law effectively burdens communication about government or political matters either in its terms, operation or effect ([155]).

Mr Ruddick failed to establish either threshold

Against that background, Gordon, Edelman and Gleeson JJ found that Mr Ruddick had failed to establish any burden on electoral choice or on the implied freedom. Even if the impugned items resulted in the deregistration of the LDP or a requirement for it to change its name, that would not preclude any communication with the public except via the ballot paper. The only effect would be that, as a candidate endorsed by the LDP, Mr Ruddick would not be entitled to have that party name printed next to his name on the ballot paper ([165]).

It was not 'self-evident' to Gordon, Edelman and Gleeson JJ that the quality of electoral choice, or the freedom of communication on government or political matters, would be impaired by this effect ([162]). In fact, their Honours found, the expected conclusion would be the opposite ([162]):

The likely effect of the narrow restrictions imposed by [items 11 and 14] is, overall, to improve the clarity, and hence the quality, of electoral choice and communication on government or political matters.

In other words, 'as the 2013 election demonstrated', the effect of items 11 and 14 would be to reduce confusion and thus 'enhance the quality of electoral choice by the public' ([165], emphasis in original).

Their Honours rejected Mr Ruddick's claim that the LDP was precluded from using its name to communicate a 'message of political philosophy' ([166]), accepting the Commonwealth's submission that items 11 and 14 do not preclude registration of names which use derivatives of the word 'Liberal' ([129], [166]).

The plurality judgment also considered that Mr Ruddick's submissions on the implied freedom were indistinguishable from the basis upon which 5 judges in Mulholland v Australian Electoral Commission (2004) 220 CLR 181 (Mulholland) had upheld the validity of earlier amendments to the registration scheme in the Electoral Act. Mr Ruddick did not seek leave to challenge the correctness of Mulholland, meaning that his implied freedom challenge also failed on that basis ([171]-[172]).

While Steward J observed that the impugned amendments might 'rightly be characterised as heavy handed' ([174]), given the Court's 'traditional deference' to the legislature in the case of laws regulating federal elections, his Honour concurred with the joint reasons of Gordon, Edelman and Gleeson JJ ([174]). Justice Steward also referred briefly to his reasons in LibertyWorks Inc v Commonwealth of Australia (2021) 95 ALJR 490, where his Honour had suggested, at [298]-[304], that there may be no constitutional basis for the implied freedom of political communication.

Dissenting judgments: the impugned amendments infringe the direct choice requirement and the implied freedom

Chief Justice Kiefel and Keane and Gageler JJ diverged substantially from the plurality in their approach to the evidence of voter confusion and the way that the impugned amendments sought to achieve their stated purpose. As a result, the dissenting judges approached the question of whether the impugned amendments breached the implied freedom, and the direct choice requirement arising from ss 7 and 24, quite differently from the plurality. 

Justice Gageler: there is no 'compelling justification' for the impugned amendments

In Gageler J's view, both of the limitations on legislative power relied on by Mr Ruddick were directly engaged. In precluding information about the party affiliation of candidates from appearing on ballot papers, the impugned amendments create a 'legal impediment' to elector's receipt of information which bears on the making of an informed choice between candidates in the casting of their ballots. The impugned amendments also impose a further, consequential practical impediment to political communication by 'rendering less effective' the communication to electors, both during and outside of an election period, of information about the party affiliation of those candidates whose party affiliation cannot appear on the ballot paper ([31]-[32]). As Gageler J put it later in his reasons ([78]):

If the name of the political party with which a candidate is affiliated cannot appear with the name of that candidate on a ballot paper, standard methods of communication with electors by political advertising such as billboards, corflutes and how-to-vote cards linking the candidate with the political party will inevitably be less effective … as fairly put on behalf of the plaintiff, … 'no political party would spend money on campaign advertising using one party name, when the ballot contains a different party name'.

Justice Gageler considered that, to the extent that this practical impediment could be demonstrated, the decision in Mulholland does not mean that excluding party affiliation from a ballot paper can never impose a burden on political communication ([78]).

The same test is to be applied in respect of both limitations

Justice Gageler considered that, in order to identify whether the impugned amendments infringed either limitation, the same test was to be applied ([29], [70]). His Honour outlined the test established in Lange v Australian Broadcasting Corporation (1997) 189 CLR 520 (Lange) that, for a law to be justified as compatible with the implied freedom, 2 conditions must be demonstrated: both the purpose of the law and the manner in which the law pursues that purpose must be shown to be compatible with the constitutionally prescribed system of government ([28], [69]). '[C]oherence of constitutional principle and methodology' required that ([29]):

a law made by the Commonwealth Parliament which imposes a legal or practical impediment to receipt by electors of information bearing on them making an informed choice between candidates for election, both during and outside of an election period, must satisfy the same two conditions in order to be justified as compatible with the core operation of ss 7 and 24 of the Constitution – that senators and members of the House of Representatives be directly chosen by the people.

Justice Gageler considered that the 'appropriateness' of approaching the core operation of ss 7 and 24 consistently with the approach taken to the implied freedom was also supported by observations in Mulholland and Australian Capital Television Pty Ltd v The Commonwealth (1992) 177 CLR 106 ([79]-[81]).

Consistent with his Honour's position in other implied freedom cases, Gageler J emphasised that the nature and extent of the impediment that a law imposed on receipt by electors of information capable of bearing on informed electoral choice can bear on the intensity of the judicial scrutiny warranted in applying the Lange test ([82]). The 'intensity of judicial scrutiny' required in this case was a test of strict necessity, consistent with the need for a 'compelling justification', because the impugned amendments directed to political communication 'discriminated between political viewpoints' and 'in favour of incumbency' ([83]).

There is no 'compelling justification' for the impugned amendments

Justice Gageler found that there was no 'compelling justification' for the impugned amendments ([32], [104]). On one hand, if the purpose of the impugned amendments was identified simply by reference to their immediate effect, that purpose would 'plainly enough' not be compatible with the constitutionally prescribed system of government. The Constitution 'does not admit of incumbent political parties giving themselves monopolies over words that express political ideologies' ([86]). However, the Commonwealth identified the purpose of the impugned provisions as minimising voter confusion about party affiliation of candidates, which 'is not only compatible with the constitutionally prescribed system of government; it is positively supportive of that system'. Justice Gageler accepted that was the purpose of the provisions and, critically, also the purpose of the other pre-existing and continuing provisions dealing with party names. However, in light of his conclusions about the extent to which voter confusion was a problem established on the evidence before the Court, Gageler J disagreed that the impugned provisions were necessary, much less that the necessity for them rose to the level of a compelling justification.

Accordingly, Gageler J would have declared that items 7, 9, 11 and 14 of the Amendment Act are wholly invalid on the grounds that they infringed the implied freedom of political communication or preclude direct choice by the people.

Chief Justice Kiefel and Justice Keane

Chief Justice Kiefel and Keane J's judgment was short and, for the most part, expressed agreement with Gageler J's reasons and conclusions ([3]).

Like Gageler J, their Honours considered that both the implied freedom, and the 'direct choice' requirement in ss 7 and 24, were engaged. In their Honours' view ([21]):

The impugned provisions prevent an elector from identifying a candidate with all that is associated with the name of a political party with which that candidate is affiliated. In their effect, they are apt to restrict or distort the choice presented to an elector… It is not in dispute that the name of a political party appearing next to the name of a candidate on a ballot paper is a form of political communication. There can be no dispute that the impugned provisions burden the freedom and do so to a considerable extent.

Their Honours expressly agreed with Gageler J that Mulholland did not affect this conclusion ([22]).

While Kiefel CJ and Keane J did not adopt Gageler J's precise reasoning as to the test for determining whether the impugned amendments breached the implied freedom and the direct choice requirement in ss 7 and 24, their Honours similarly concluded that the test in respect of both limitations was practically the same, finding that ([19]):

An impairment of choice has been said to require that there be a substantial reason for it to be constitutionally valid. An impairment of or burden on the implied freedom of political communication requires that it be justified. The level of justification must be commensurate with the extent of the burden. In reality the requirements of a substantial reason and justification amount to the same thing. They both require that the legislative measure be proportionate to the purpose it seeks to achieve.

Their Honours did not consider that the impugned measures were proportionate to their stated purpose of reducing voter confusion. Neither the special case nor Mr Ruddick's admission could be relied upon for a conclusion that 'a word contained in the name of another, registered, political party of itself gives rise to a real risk of confusion on the part of voters as to the party affiliation of a candidate for election' ([15]). In Kiefel CJ and Keane J's view, all the special case confirmed is 'what political scientists refer to as "the donkey vote"' - namely, the effect that a party's position on a ballot paper may have on its share of the vote. The Joint Standing Committee on Electoral Matters reports cited in the special case merely supported a view that 'voters are misled because of the random allocation of the location of the party name on the ballot paper' ([13]-[15]).

As such, the impugned amendments could not be 'justified' by reference to the stated purpose of preventing voter confusion. They were 'directed principally to a problem which does not arise from a word in a party name, but rather from the location of the name on a ballot paper'. Insofar as they did seek to reduce voter confusion, they did not need to go as far as they do. There was an 'obvious and compelling alternative' which did not unnecessarily burden the implied freedom - namely, the provisions ss 129(1), 129A and 134A(1) as they had stood before the introduction of the amendments ([23]).

The Commonwealth's legal team

The AGS Constitutional Litigation Unit (Niamh Lenagh‑Maguire, David Rowe and Tasha McNee) acted for the Commonwealth, instructed by the Department of Finance, with the Solicitor‑General, Dr Stephen Donaghue KC, Brendan Lim and Christine Ernst as counsel.

The text of the decision is available at:

  Back to top


Lenagh-Maguire, Niamh

Senior Executive Lawyer

Constitutional Litigation Unit

Two challenges to the Iron Ore Processing (Mineralogy Pty Ltd) Agreement Amendment Act 2020 (WA) fail

The High Court unanimously rejected challenges to the validity of the Iron Ore Processing (Mineralogy Pty Ltd) Agreement Amendment Act 2020 (WA) (WA Act). The challenges were brought in 2 separate cases that were heard and determined together. In the first case the plaintiffs were Mineralogy Pty Ltd (Mineralogy) and International Minerals Pty Ltd (International Minerals). The plaintiff in the second case was Mr Clive Palmer.

There were 2 separate sets of reasons in each case: one set of reasons comprising the judgment of Kiefel CJ and Gageler, Keane, Gordon, Steward and Gleeson JJ (the plurality); and a separate set of reasons by Edelman J. Edelman J agreed with the result reached by the plurality in each case but differed slightly in his reasoning.

Mineralogy Pty Ltd v Western Australia
High Court of Australia, 13 October 2021
[2021] HCA 30

Palmer v Western Australia
High Court of Australia, 13 October 2021
[2021] HCA 3


The 2 cases had their origin in an agreement called the 'Iron Ore Processing (Mineralogy Pty Ltd) Agreement' (the State Agreement). The State Agreement deals with the exploitation of certain iron ore deposits in the Pilbara. The parties to the State Agreement are Western Australia (WA) and a number of corporations beneficially owned by Mr Clive Palmer, including Mineralogy and International Minerals.

In 2012, the parties to the State Agreement fell into dispute. The dispute was referred to arbitration. The arbitrator made 2 awards. Both awards were in favour of Mineralogy and International Minerals and against WA. A third arbitration was set in motion to determine the plaintiffs' claims for damages against WA, including the quantum of those damages. Mineralogy and International Minerals claimed a large sum.

In late 2020, WA enacted the WA Act to terminate the third arbitration and immunise WA with respect to all claims against it. The WA Act seeks, in multiple ways, to nullify or prevent any present or future legal claims or liability on the part of WA in relation to the State Agreement. The response of Mineralogy, International Minerals and Mr Palmer was to commence proceedings in the High Court challenging the constitutional validity of the WA Act.

The High Court's decision

The first challenge: by Mineralogy and International Minerals

Mineralogy and International Minerals challenged the validity of the WA Act on 'multifarious legal bases' ([7]). The Court decided that it was unnecessary or inappropriate to decide many of them. The Court held that only 4 of the plaintiff's arguments properly fell for decision (see [67] and [97]):

  • whether the WA Act as a whole is inconsistent with s 6 of the Australia Act 1986 (Cth) (Australia Act)
  • whether the WA Act as a whole exceeds limitations on WA's legislative power said to arise from 'the rule of law and deeply rooted common law rights' ([70])
  • whether ss 9(1) and (2) and ss 10(4) to (7) of the WA Act are inconsistent with Ch III of the Constitution because they amount to the exercise of judicial power by a state Parliament contrary to Ch III (or alternatively are contrary to the Kable principle (Kable v Director of Public Prosecutions (NSW) (1996) 189 CLR 51))
  • whether ss 9(1)-(2) and 10(4)-(7) of the WA Act are incompatible with s 118 of the Constitution.

Each of those 4 bases of challenge was unanimously rejected.

WA Act is not contrary to s 6 of the Australia Act (manner and form challenge)

The plaintiffs argued that the WA Act is inconsistent with s 6 of the Australia Act, which provides that:

a law made after the commencement of this Act by the Parliament of a State respecting the constitution, powers or procedure of the Parliament of the State shall be of no force or effect unless it is made in such manner and form as may from time to time be required by a law made by that Parliament.

The Australia Act is an important Commonwealth Act. It has quasi-constitutional status. It was enacted following a request from all states under s 51(xxxviii) of the Constitution. Section 6 enhances the legislative powers of current state Parliaments by allowing them, if they see fit, to entrench certain kinds of laws (that is, laws respecting the constitution, powers and procedure of Parliament) at the expense of the powers of future parliaments (which must thereafter follow a special 'manner and form' to amend an entrenched law).

The State Agreement is, of itself, merely a contract that was entered into in 2001. It contains a provision that specifies that the State Agreement may be amended by agreement of the parties. However, by special act of the WA Parliament enacted in 2002, the State Agreement was 'ratified' and authorised 'to operate and take effect … according to its terms notwithstanding any other Act or law'.

The plaintiffs said that the effect of 2002 Act was to transform the State Agreement into a law of the WA Parliament. It followed (the plaintiffs claimed) that any attempt to amend the State Agreement by a means other than that provided for in the State Agreement itself   specifically, by agreement of the parties   amounted to an attempt to amend a law of the WA Parliament. The amendment of the State Agreement in 2020 by the WA Act was not by agreement of the parties: the WA Act was the result of the WA Parliament legislating unilaterally. This was said to be contrary to a 'manner and form' restriction laid down by the State Agreement. Section 6 of the Australia Act requires manner and form requirements to be complied with. The WA Act did not comply and so was said to be invalid because it was inconsistent with s 6 of the Australia Act ([75]).

The plurality rejected this argument. It held that the State Agreement merely lays down a procedure by which the parties might amend the State Agreement by agreement. The State Agreement says nothing one way or another to prevent the WA Parliament from amending the State Agreement unilaterally by legislating ([78] and [79]). There was therefore no applicable manner and form requirement at all.

Justice Edelman essentially agreed that the plaintiffs had failed in that step of their argument ([154]). However, his Honour was prepared to accept that some or all of the other steps in the plaintiffs' manner and form argument had been made out; this latter point was something that the plurality did not rule on ([147]-[153]).

Asserted limitations on the scope of WA's legislative power

The plaintiffs argued that the WA Act as a whole exceeds asserted limitations on the legislative power of the WA Parliament said to arise from 'the rule of law and deeply rooted common law rights'. The plurality dismissed this argument on the basis that no operative limitation such as alleged arises from the rule of law. The plaintiffs had also failed to identify any relevant 'deeply rooted common law right' ([70]).

Ch III argument

The Ch III argument concerned s 9(1) and (2) of the WA Act (which deem that certain actions previously taken by the plaintiffs under the State Agreement have no legal effect) and s 10(4)-(7) (which deems that earlier agreements between WA and the plaintiffs to arbitrate their disputes, and the resulting arbitral awards, have no force or effect). The argument was that these provisions amounted to the WA Parliament purporting to decide the dispute between the plaintiffs and WA rather than leaving it for the courts to do. Thus, it was said, the WA Act amounted to the WA Parliament purporting to exercise judicial power itself. This was said to be contrary to Ch III of the Constitution. Alternatively, the WA Act was said to be contrary to Ch III on the basis of the Kable principle- that is,the WA Act undermined the institutional integrity of the WA courts.

The plurality held that ss 9(1)-(2) and 10(4)-(7) are not contrary to Ch III. They do not amount to the exercise of judicial power. The provisions merely alter the law in a way that changes existing rights. It is well settled that parliaments can do this. To do so does not amount to the exercise of judicial power; nor does it undermine the institutional integrity of courts. In short, the provisions '[go] no further than to ascribe new legal consequences to past events and thereby to alter substantive legal rights' ([84]). That this might have led to harsh outcomes for the plaintiffs made no difference ([86]). Justice Edelman agreed but noted that the result might (in his opinion) have been different if the WA Act had been enacted at a time when proceedings between WA and the plaintiffs were already on foot (which was not the case) ([159]).

Section 118 argument

The s 118 argument was that the WA Act is in conflict with, and fails to give full force and effect to (as it was said s 118 of the Constitution requires), a Queensland statute   namely, the Commercial Arbitration Act 2013 (Qld) (Qld Act). The Qld Act (which is part of a national scheme: every state has enacted a statute in similar terms) provides for Queensland courts to recognise and enforce arbitral awards made anywhere in Australia. The plurality rejected the challenge. The plurality held that no conflict between the WA Act and the Qld Act has yet been demonstrated to have arisen ([92]). Justice Edelman basically agreed ([165]). Thus the question of the proper purpose and construction of s 118 of the Constitution (a vexed question) was left for another day.

All other challenges rejected as not properly arising

In addition to the above challenges, the plaintiffs brought a wide range of other challenges to particular provisions of the WA Act. The plurality and Edelman J held that it was unnecessary and inappropriate to consider these other challenges. The plurality said ([10]; see also [67], [112]):

The procedure adopted by the parties is inappropriate to obtain judicial resolution of legal issues which might or might not arise in relation to the numerous other legal bases on which the plaintiffs seek declarations of invalidity…The reason is that the facts agreed between the parties for the purpose of the special case provide an inadequate foundation upon which to crystallise those legal issues or to demonstrate the necessity of their resolution to the determination of any immediate right, duty or liability in controversy between the parties.

The second challenge: by Mr Clive Palmer

Mr Palmer raised essentially the same arguments put by the plaintiffs in Mineralogy, as well as a number of additional arguments. The plurality and Edelman J adopted their reasoning from Mineralogy and dismissed the arguments already put in that case on the same bases. Their Honours then dealt with the additional arguments raised by Mr Palmer.

Section 117 argument

Section 117 of the Constitution provides:

A subject of the Queen, resident in any State, shall not be subject in any other State to any disability or discrimination which would not be equally applicable to him if her were a subject of the Queen resident in such other State.

This was an argument that the WA Act was invalid because it singled out Mr Palmer for a 'disability' or 'discrimination' of a kind forbidden by s 117 of the Constitution. Both the plurality and Edelman J rejected this argument ([6], [18]). The WA Act does not discriminate against Mr Palmer by reason of his residency. It would affect him just the same way even if he were a resident of WA:

The prohibition in s 117 is against subjecting an Australian citizen resident in one State to a disability or discrimination in another State which would not be equally applicable to that Australian citizen if the Australian citizen were resident in that other State. Even if the [WA Act] could be characterised as singling the plaintiff out for a disability or discrimination, that disability or discrimination could not be characterised as being of a kind forbidden by s 117 of the Constitution. Neither the legal operation nor the practical operation of the whole or any part of the [WA Act] would be any different if the plaintiff were resident in Western Australia instead of being resident in Queensland.

Ch III argument

Mr Palmer argued that the WA Act constitutes an invalid attempt to exercise authority to determine a justiciable controversy within the scope of s 75(iv) of the Constitution. This kind of authority is capable of being exercised only by a court. The plurality rejected this argument on the basis that it presupposed that the WA Act involved an exercise of judicial power (which, for the reasons given in Mineralogy at [86], it did not) ([7]). Justice Edelman found similarly ([13]). His Honour gave more attention to the plaintiff's argument that the WA Act is a 'bill of pains and penalties' than the plurality, who dismissed it outright, but he ultimately dismissed it ([14]-[16]).

Rule of law argument

This was an argument that the WA Act violates an asserted 'cardinal principle' of the rule of law that the government should be 'under law', with the law applying to and being observed by the government and its agencies just as it applies to the ordinary citizen. The plurality dismissed this argument on the basis that it sought to impermissibly 'constitutionalise' the term 'rule of law' as if it were some freestanding limit on legislative power over and above what is stated in the text of the Constitution itself. The plurality noted that it had been 'acknowledged repeatedly' that the rule of law supports neither expansion of judicial power nor contraction of legislative or executive power, beyond those limits that inhere in the text and structure of the Constitution ([7]). Justice Edelman found similarly ([26]).

Section 109 inconsistency argument

Finally, the plaintiff argued that the WA Act is inconsistent with various Commonwealth laws. As with the arguments in Mineralogy mentioned above, the plurality and Edelman J found that this argument did not properly arise. The special case agreed by the parties provided no basis for the Court be satisfied of the necessity of determining whether any such inconsistency exists ([9], [12]).

The Commonwealth's legal team

The AGS Constitutional Litigation Unit (Gavin Loughton, Emily Kerr and Anthony Hall) acted for the Attorney‑General, with the Solicitor‑General Stephen Donaghue KC, Frances Gordon, Tom Wood and Jackson Wherrett as counsel.

The text of the decisions is available at:

  Back to top


Migration detention validly authorised where removal should have occurred

By a 4:3 majority, the High Court (Kiefel CJ, Gageler, Keane and Steward JJ; Gordon and Gleeson JJ and Edelman J dissenting) allowed the Commonwealth's appeal from the judgment of the Federal Court in AJL20 v The Commonwealth [2020] FCA 1305 and held that the Migration Act 1958 (Cth) (Migration Act) authorised the respondent's detention, even where the Executive had failed to comply with the statutory duty to remove the respondent 'as soon as reasonably practicable'.

Commonwealth of Australia v AJL20
High Court of Australia, 23 June 2021
[2021] HCA 21; 95 ALJR 567; 391 ALR 562



AJL20 is a citizen of Syria who came to Australia in 2005 holding a child visa. The visa was cancelled in October 2014 under s 501(2) of the Migration Act and from that time AJL20 was detained pursuant to s 189 of the Act, which imposes a duty to detain a person known, or reasonably suspected to be, an unlawful non‑citizen. In the Federal Court, AJL20 alleged that his detention since 26 July 2019 was unlawful. He sought damages for false imprisonment and an order in the nature of habeas corpus.

Justice Bromberg found that the Commonwealth had failed in its duty under s 198 of the Act to remove AJL20 'as soon as reasonably practicable' because:

a. between 26 July 2019 and 27 November 2019 it took no steps to remove him anywhere
b. from 28 November 2019 it took steps to remove him to Lebanon but took no steps to remove him to Syria.

Justice Bromberg held that this failure meant that the detention of the respondent was no longer for an authorised purpose and thus was unlawful. His Honour ordered AJL20's release and a hearing on unlawful imprisonment was to follow.

On appeal (which was removed into the High Court on the application of the Commonwealth Attorney‑General), the Commonwealth conceded that it had not complied with the duty under s 198. The reason for not taking steps to remove AJL20 to Syria was that he was found to be owed protection in relation to that country. However, s 197C of the Migration Act (as in force at the relevant time) made non‑refoulement considerations irrelevant to the performance of the duty to remove under s 198 of the Act.

The issue on appeal was whether ss 189 and 196 of the Act, consistently with Ch III of the Constitution, validly required and authorised the detention of an unlawful non‑citizen until, relevantly, he or she is in fact removed under s 198(1) irrespective of whether there has been compliance with duty under those provisions to remove the unlawful non‑citizen from Australia 'as soon as reasonably practicable'.

The High Court's decision

The High Court delivered 3 judgments: a majority joint judgment of Kiefel CJ, Gageler, Keane and Steward JJ; a dissent by Gordon and Gleeson JJ; and a separate dissent by Edelman J. The majority and minority differed in their view of the effect that the actions of the Executive could have on the purpose, and therefore lawfulness, of detention that was required under the Act.

The legislative scheme authorising detention is constitutionally valid

The majority began by setting out the established principles regarding the scope of Parliament's power to make laws with respect to aliens and the qualifications on that power that arise by implication Ch III of the Constitution ([20]-[32]). They then directly responded to the reasons of the primary judge, noting that Bromberg J's approach had conflated questions of constitutional validity with questions of statutory interpretation ([42]). Their Honours explained that, when the Executive exercises statutory power, the 'constitutional question' asks whether the statutory authority conferred on the Executive is constitutionally valid and the 'statutory question' asks whether, as a matter of administrative law, the executive action was authorised by the statute ([43]).

Turning to the 'constitutional question', the majority observed that it is well settled that detention that is reasonably capable of being seen as necessary for the legitimate non‑punitive purposes of segregation pending investigation and determination of any visa application or removal from Australia is consistent with the requirements of Ch III of the Constitution ([44]).

Furthermore, the scheme of Executive detention set up by ss 189, 196 and 198 of the Migration Act is hedged about by enforceable duties, such as the duty to remove the detainee as soon as reasonably practicable, that connect the duty to detain unlawful non‑citizens with these legitimate, non‑punitive purposes and ensure that detention does not occur at the unconstrained discretion of the executive ([44]-[45]; see also [25] and [28]). It followed that the detention authorised by the scheme was valid in all its potential applications ([45]) and there was no need to read down the provisions to save their validity as the primary judge had done ([46]).

Once it is accepted that the scheme of the Act requiring and authorising detention was 'within Parliament's competence because of the imposition of duties on officers of the Executive', a failure on the part of the officers to 'diligently perform the duties' could not take the law outside Parliament's competence or erase those duties or the statutory purposes which those duties support ([48]). The alternative would mean that 'the supremacy of the Parliament over the Executive would be reversed and the rule of law subverted' ([48]).

The detention is validly authorised until an event in s 196(1) actually occurs

Having found that the detention regime in the Migration Act was valid in all of its operations, the majority concluded that the detention authorised by s189(1) must continue, and remains lawful, up until one of the 'terminating events' (broadly, removal or grant of visa) in 196(1) actually occurs, not the time by which the action could or should have occurred ([49]). In short (51):

The duty imposed upon officers of the Executive by ss 189(1) and 196(1) of the Act is to detain the unlawful non‑citizen until the occurrence of one of the events referred to in s 196. The duty so imposed by the Act is neither conditional upon, nor co‑extensive with, the intents or purposes of officers of the Executive towards the detainee.

Accordingly, 'the notion that pursuit by the Executive of a purpose that is unauthorised or even prohibited by the Act' might render the detention required by the Act invalid is 'problematic' ([71]). Furthermore, the authorities, including Plaintiff S4/2014 v Minister for Immigration and Border Protection (2014) 253 CLR 219, do not support 'the leap from a finding of a failure by the Executive to remove the respondent as soon as reasonably practicable … to the conclusion that the Executive was detaining the respondent for a purpose other than his removal from Australia' ([67]). Finally, where there has been a failure to perform the duty to remove, the proper remedy in these circumstances is mandamus to compel performance of the duty; habeas corpus is not available ([52]). As the majority noted, the duties are 'not things written in water', and failure to comply with an order for mandamus may result in officers being committed to prison for contempt of court ([52]).
On the basis of this reasoning, the majority concluded ([72], emphasis added):

It is enough for the duty to detain imposed by s 189(1) to be sustained in accordance with s 196(1)(a) until completion of the performance of the duty to remove imposed by s 198 that the officer keeping or causing the person to be kept in immigration detention knows or reasonably suspects that the person is an unlawful non‑citizen. Provided the requisite knowledge or suspicion continues to exist throughout the period of detention, an unauthorised or prohibited purpose on the part of the officer in prolonging the period of detention can affect neither the duty to detain nor the duty to remove nor the appropriate remedy for non‑performance of the duty to remove.

The dissenting judgments

Justices Gordon and Gleeson

In contrast to the majority's approach, which hinged on the view that the authority to detain conferred by the Act was separated from the duty to remove, Gordon and Gleeson JJ divide the power to detain unlawful non‑citizens into 'two interlocking dimensions': taking into detention and the duration of detention ([87]).

When understood by reference to these dimensions and the requirements of the Constitution, the authority that the Migration Act confers on the Executive to keep an unlawful non‑citizen in immigration detention 'stops when time for removal as soon as reasonably practicable has expired' – detention beyond this time is unlawful and could not be authorised consistently with the requirements of the Constitution ([84], [89]). This is because the justification for the duration of detention was set at the commencement of detention ([92]) and 'the relevant limit is removal as soon as reasonably practicable, not the bare fact of removal' ([90]). The Commonwealth's approach - that detention remained lawful up until the time that a 'terminating event' such as removal actually occurred - would have elided the distinction between taking into detention and duration ([86]) and 'would enable detention of unlawful non‑citizens at the unconstrained discretion of the Executive' and render the Ch III limits on Executive detention meaningless ([83]).

Accordingly, the requirement that an officer must maintain their knowledge or reasonable suspicion that a person is an unlawful non‑citizen throughout the duration of the person's detention is not – and could not validly be – the only limit on the duration of detention ([92]). Lawful detention depends on compliance with the time limit imposed by the duty to remove as soon as reasonably practicable ([89]). If there is noncompliance with the duty to remove, the detention becomes unlawful at the time when the detainee should have been removed ([90]). Prolongation of detention beyond this time is not authorised by the Act and is therefore unlawful.

Further, unlawful executive inaction may also be characterised as a departure from the purpose of detention ([103]). The purpose of detention is assessed objectively, with reference to all of the circumstances, including the conduct of officers. When officers considered non‑refoulement obligations contrary to s 197C and failed to remove AJL20 as soon as reasonably practicable, this altered the purpose of his detention from one that was permissible ([103]).

As to the appropriate remedy when there had been a failure to comply with the duty to remove, their Honours said wrongful detention by the Executive compelled the availability of habeas corpus, which protects individual liberty in Australian law ([92]). The availability of mandamus did not preclude the availability of habeas corpus, as these remedies have different purposes ([93]). Equally, the fact that that the Migration Act requires an unlawful non‑citizen to be immediately re‑detained upon release did not preclude habeas corpus([89], [92]-[94]).

Justice Edelman

Justice Edelman identified the key question as whether the Executive could perform the duty to detain AJL20 for an objective purpose that was inconsistent with the scope and purposes of the Migration Act ([108]). In answering this question, his Honour identified the 2 distinct categories of duty: a duty to keep unlawful non‑citizens in immigration detention for proper purposes; and a duty to remove unlawful non‑citizens as soon as reasonably practicable. He noted that a breach of one duty did not mean a breach of the other and that the legal response to a breach of each duty is different ([110]).

His Honour stated that the performance of these duties must be consistent with the scope and purpose of the statutory enactment ([125]) and that an implication of this kind was necessary to preserve the constitutional validity of the Act ([137]). Furthermore, in the case of the duty to detain, the purpose of the detention is to be assessed objectively, with reference to the conduct of executive officers ([138]). Analysing AJL20's detention in light of these principles, Edelman J held that the Executive's objective purpose was to remove him in a manner consistent with its non‑refoulement obligations. This purpose was expressly inconsistent with s 197C, meaning the detention of AJL20 was not authorised by the Act ([148]-[149]). The primary remedy for unlawful detention was habeas corpus, which was available to AJL20 as long as the Executive pursued its illegitimate purpose ([130], [143], [157]).

Separately, his Honour noted the Executive had also breached its duty to remove AJL20 as soon as reasonably practicable ([113]). This breach was distinct to AJL20's unlawful detention and required a different remedy ([142]). Noncompliance with the duty to remove could be remedied by mandamus([143]).

The Commonwealth's legal team

The AGS Constitutional Litigation Unit (Simon Thornton and Danielle Gatehouse) and AGS Dispute Resolution (Dejan Lukic and Lizzy McCallum) acted for the Commonwealth, with the Solicitor‑General, Dr Stephen Donaghue KC, Geoffrey Kennett SC, Chris Tran and Naomi Wootton as counsel.

The text of the decision is available at:

  Back to top


Thornton, Simon

Senior Executive Lawyer

Lukic, Dejan

Senior Executive Lawyer

Western Australian legislation conferring power to close borders during a pandemic held valid

The High Court unanimously (5:0) held that Western Australian legislation relied upon to 'close' Western Australia's border in response to the COVID-19 pandemic (ss 56 and 67 of the Emergency Management Act 2005 (WA) (EM Act)) is not contrary to the requirement in s 92 of the Constitution that 'trade, commerce, and intercourse among the States … shall be absolutely free'.

The Court further held that the exercise of the power to close the border (through the making of the Quarantine (Closing the Border) Directions (WA) (the Directions)) did not itself raise a constitutional question, and no issue was otherwise taken as to whether the Directions were validly authorised by the EM Act.

Palmer v State of Western Australia
High Court of Australia, 24 February 2021
[2021] HCA 5


On 15 March 2020, some 4 days after the World Health Organization declared COVID‑19 a pandemic, the Minister for Emergency Services for Western Australia declared a state of emergency for the whole of Western Australia with effect from 16 March 2020.
The power to declare a state of emergency is conferred on the Minister by s 56 of the EM Act (as then in existence). Before making the declaration, the Minister was required to be satisfied that:

  • an 'emergency' (being a hazard of such nature or magnitude as requires a significant and coordinated response) had occurred, was occurring or was imminent
  • extraordinary measures were required to prevent or minimise loss of life, prejudice to the safety, or harm to the health, of persons or animals.

A state of emergency remains in force for an initial period of only 3 days but can be extended from time to time by a maximum of 14 days at a time. In the present case the state of emergency declared for the whole of Western Australia was repeatedly extended and was still in force at the time of the hearing in the High Court.

Where a state of emergency has been declared, various powers are then available to be exercised under the EM Act. Most relevantly for present purposes, s 67 of the EM Act (as then in existence) authorises the 'State Emergency Coordinator' (being the Commissioner of Police) to 'direct or … prohibit, the movement of persons, animals and vehicles within, into, out of or around an emergency area or any part of an emergency area'.

This power is to be exercised 'for the purpose of emergency management', which relevantly includes to prevent or mitigate the adverse effects of an emergency and to respond to the emergency by combatting its effects.

Soon after the state of emergency declaration, the Commissioner of Police made the Directions pursuant to s 67 of the EM Act ([5]). The Directions took effect from 5 April 2020 and effectively closed the border of Western Australia to all persons from any place unless they had an exemption under the Directions.

The plaintiffs' challenge

The first plaintiff, Mr Palmer, applied for an exemption to enter Western Australia for business and other purposes related to his position as Chairman and Managing Director of the second plaintiff (Mineralogy Pty Ltd). The exemption was refused. On 25 May 2020 both plaintiffs commenced this proceeding in the High Court against the State of Western Australia and the Commissioner of Police for Western Australia (the State), seeking a declaration that either the EM Act or the Directions was invalid by reason of s 92 of the Constitution, which relevantly provides that 'trade, commerce, and intercourse among the States … shall be absolutely free'. (Two proceedings were also commenced against the State of Queensland by Mineralogy and other plaintiffs, but those proceedings were subsequently discontinued.)

Following the determination of a factual issue, remitted by the High Court to the Federal Court, as to the State's claim for the reasonable need for and efficacy of the Directions, the parties agreed on a special case which stated the following question of law for the Full Court of the High Court:

Are the Quarantine (Closing the Border) Directions (WA) and/or the authorising Emergency Management Act 2005 (WA) invalid (in whole or in part, and if in part, to what extent) because they impermissibly infringe s 92 of the Constitution?

The Attorneys‑General of the Australian Capital Territory, Tasmania, Queensland, South Australia, Victoria and the Northern Territory intervened. The Attorney‑General of the Commonwealth initially also intervened but subsequently discontinued that intervention (see Palmer v Western Australia (No 3) [2020] FCA 1220).

The High Court's decision


The High Court, sitting only 5 Justices (Bell and Nettle JJ being about to retire), delivered 4 judgments: a joint judgment of Kiefel CJ and Keane J; and separate judgments of Gageler, Gordon and Edelman JJ.

In summary, the Court decided that:

a. The only constitutional question is the validity of the EM Act; the validity of the Directions turns on whether they were authorised by the EM Act (which the plaintiffs did not put in issue) and not on any constitutional question.
b. Like the 'trade and commerce limb', the 'intercourse limb' of s 92 is concerned with discriminatory burdens on interstate, as compared to intrastate, intercourse.
c. A discriminatory burden on interstate intercourse will not infringe s 92 if it is 'reasonably necessary' to achieve a legitimate object. Whether a burden is reasonably necessary can be determined by use of structured proportionality analysis.
d. Sections 56 and 67 of the EM Act do not infringe s 92 having regard to the limited circumstances in which a state of emergency declaration can be made, the limited duration of such a declaration, the preconditions on the making of directions which restrict cross‑border movement, the limited purpose for which such directions can be made, and the importance of the purpose or object of the provisions.
a) The appropriate level of analysis for s 92 is to consider the Act only

The primary focus of argument in this case had been on the constitutional validity of the Directions, made under the EM Act. This gave rise to a threshold question: does s 92 apply to the Directions, to the EM Act or to both?

While recognising that s 92 constrains both legislative and executive action, the High Court unanimously held that the constitutional question is whether s 92 is infringed by the EM Act and not whether it is infringed by the particular exercise of power under the Act (that is, making the Directions) (Kiefel CJ and Keane J [63]; Gageler J [118]; Gordon J [201]-[202]; see also Edelman J [224]). The Court thus followed and applied the approach put by the Commonwealth and endorsed by a majority of the High Court in Wotton v Queensland (2012) 246 CLR 1 in explaining how constitutionally guaranteed freedoms operate as a limit on both legislative and executive power (Kiefel CJ and Keane J [65]; Gageler J [117]; see also Gordon [201]; Edelman J [225]). Thus, as Gageler J stated ([119], emphasis added):

Where executive action purporting to be taken pursuant to statute imposes a burden argued to infringe the implied constitutional guarantee of freedom of political communication or the express constitutional guarantee of absolute freedom of trade, commerce and intercourse among the States, two distinct questions accordingly arise: one constitutional, the other statutory. The statutory question is whether the executive action is authorised by the statute. The constitutional question is whether the statute complies with the constitutional guarantee if, and insofar as, the statute authorises the executive action.

The Court recognised, however, that difficult questions may arise where 'the power or discretion given by the statute is broad and general' (Kiefel CJ and Keane J [68]; see also Edelman J [219]). In such cases the statutory and constitutional questions 'can converge' (Gageler J [122]).

That was not the case here. The Court held that the relevant provisions of the EM Act were sufficiently constrained so as validly to authorise the making of directions that restricted entry into a declared emergency area, which could be the entire state, and thereby impose a differential burden on interstate intercourse (Kiefel CJ and Keane J [68]; Gageler J [127]; Gordon J [201]; cf Edelman J [228]-[230], deciding only that the provisions were valid in their application to facts of the same general kind as those in issue in the present case).

b) Section 92 is concerned with discrimination against interstate intercourse

For much of the time since federation the High Court struggled to settle on an approach to the meaning and operation of s 92 of the Constitution. However, in Cole v Whitfield (1988) 165 CLR 360 at 394 the Court, after 140 earlier failed attempts, authoritatively determined that the scope of the freedom of interstate trade and commerce guaranteed by the section is a freedom from 'discriminatory burdens of a protectionist kind'. In doing so, the Court effectively divided s 92 into 2 distinct limbs: a freedom of interstate trade and commerce; and a distinct freedom of interstate intercourse.

In the present case the Court largely swept that distinction away, unanimously holding that both limbs of s 92 should now be understood as centrally concerned with the notion of unjustified discriminatory burdens on the relevant freedom (Kiefel CJ and Keane J [48]; Gageler J [97] and [114]; Gordon J [181]; Edelman J [215] and [242]-[249]). For Kiefel CJ and Keane J, that is because there is no textual reason to distinguish between the limbs; the intercourse limb may otherwise largely subsume the trade and commerce; and general laws that burden interstate movement may be held invalid ([48]).

The Court (other than Edelman J, who took a broader view) confirmed the holding in Cole v Whitfield that the trade and commerce limb is concerned with discrimination of a protectionist kind(Gageler J [114]) - that is, with 'a law that treats interstate trade or commerce differently as compared to intrastate trade or commerce, and effects a disadvantage to interstate trade or commerce' (Kiefel CJ and Keane J [31]-[32]) and thus operates 'for the protection of domestic [intrastate] industries against foreign [interstate] competition' (Gordon J [184]; cf Edelman J [237]-[239], [255]-[260]). The intercourse limb is similarly concerned with a law that discriminates againstinterstate intercourse by imposing a differential burden on interstate, when compared to intrastate, intercourse (whether or not it is 'protectionist') (Kiefel CJ and Keane J [47]-[48]; Gageler J [92]; Gordon J [184], [190]; cf Edelman J [241] (again taking a broader view).

c) Differential treatment is justified if 'reasonably necessary' (applying structured proportionality)

Where a law discriminates in the relevant sense, the next (and decisive) question is whether the burden the law imposes on interstate intercourse can be justified. All members of the Court agreed that this requires determining whether the law (or burden) is 'reasonably necessary' to achieve a legitimate object or end (Kiefel CJ and Keane J [50]; Gageler J [139]; Gordon J [192]; Edelman J [271]). A majority further held that the test of structured proportionality should be applied to determine whether a burden is reasonably necessary (Kiefel CJ and Keane J [62]; Edelman J [264]). Justices Gageler and Gordon, however, declined to apply structured proportionality as a tool of legal analysis ([94], [192]-[193]) because of its 'rigidity' ([144], [198]) and as being inappropriate to the task of characterisation ([198]).

d) Sections 56 and 67 of the EM Act do not infringe s 92

Applying their Honours' respective approaches, each judgment held that the relevant provisions of the EM Act did not infringe s 92 of the Constitution.

In reaching that conclusion the Court accepted that s 92 was potentially engaged because, although the EM Act was not directed to movement across the Western Australian border and accordingly did not, in terms, discriminate against interstate movement, an exercise of the power conferred by the Act could impose a differential burden on interstate intercourse (Kiefel CJ and Keane J [72]; Gageler J [165]; Gordon J [204]; Edelman J [282]).

It was therefore necessary to consider whether the power to impose a differential burden on interstate intercourse was justified. And that required, as a first step, identifying the purpose of ss 56(1) and 67 of the EM Act. Each judgment identified that purpose, which was accepted to be legitimate, in slightly different terms:

  • Chief Justice Kiefel and Keane J identified the purpose as being 'the protection of the health of residents of Western Australia' and 'to prevent infectious diseases such as COVID‑19 spreading into the Western Australian community' ([74], [76]).
  • For Gageler J the legislatively identified end to be achieved was 'that of managing the adverse effects of a plague or epidemic of a nature that requires a significant and coordinated response' ([153]).
  • Justice Gordon identified the object of the impugned provisions as being 'concerned with managing a state of emergency' ([205]).
  • According to Edelman J, ss 56 and 67 of the EM Act served the legitimate purpose of 'creat[ing], and … mak[ing] conditional, broad powers … to manage a broad range of emergencies' ([280]).

The critical question was whether the identified purpose of the impugned provisions of the EM Act justified a power to restrict the entry of persons into Western Australia during a state of emergency. In summary:

  • For a majority of the Court, the combined effect of (a) the limited circumstances in which a state of emergency declaration can be made and the limited duration of such a declaration, and (b) the preconditions on the making of directions which restrict cross-border movement, the limited purpose for which such directions can be made and the importance of the purpose or object of the provisions, were sufficient to justify the power to burden interstate intercourse.
  • Unlike the majority, Edelman J did not decide whether the provisions were sufficiently constrained in order to be wholly valid but did find them valid in the general circumstances of the present case.

Chief Justice Kiefel and Justice Keane: structured proportionality

As noted above, Kiefel CJ and Keane J used a structured proportionality analysis to answer the question whether the power to restrict entry into Western Australia was justified. In doing so their Honours had regard to the factual findings by the Federal Court concerning the risk posed by COVID‑19, notwithstanding that the analysis was conducted at the level of the EM Act rather than Directions:

  • Suitability: Their Honours held that 'a law restricting the movement of persons into a State is suitable for the purpose of preventing persons infected with COVID‑19 from bringing the disease into the community' ([77]). In particular, (a) the matters required to be considered before such restrictions can be put in place, and (b) the shortness of the period of an emergency declaration 'suggest that these measures are a considered, proportionate response to an emergency such as an epidemic' ([77]).
  • Necessity: Further, the findings at trial as to the risk posed by COVID‑19 left 'little room for debate about effective alternatives' ([80]).
  • Adequacy of balance: Finally, their Honours held that although 'the restrictions are severe … it cannot be denied that the importance of the protection of health and life amply justifies the severity of the measures' ([81]).

Justice Gageler: reasonable necessity

Justice Gageler held that ss 56 and 67 of the EM Act conformed to the standard of 'reasonable necessity' in that any differential burden that might result from an exercise of the power of direction was justified according to the 'standard of reasonable necessity across the range of potential exercises of the power' ([166]). His Honour noted that the power to make a state of emergency declaration (s 56) is subject to 2 significant limitations: a temporal limit; and the need to be 'satisfied' of the occurrence or imminence of an emergency and the need for extraordinary measures ([156]-[159]). The making of a state of emergency declaration then enlivened the power to give directions prohibiting movement into or out of an emergency (s 67). It was of particular significance for Gageler J that 'the purpose of emergency management is the sole purpose for which the power of direction can be exercised' (emphasis added) and the discretion to exercise the power for that purpose was subject to an implied condition that its exercise must be reasonable ([164]-[165]).

Justice Gordon: reasonable necessity

Like Gageler J, Gordon J considered that any differential burden on interstate trade, commerce or intercourse resulting from a direction prohibiting movement across a state border is 'reasonably necessary for that legitimate object' ([206], [210]). Her Honour held that ss 56 and 67 of the EM Act are not 'provisions where the discretion is insufficiently controlled or … so wide as to be susceptible to being exercised inconsistently with s 92' ([208]). A differential burden can be placed on interstate trade, commerce or intercourse 'only in extraordinary and highly particular circumstances, namely to meet an emergency constituted by, in this case, an epidemic' ([208]).

Justice Edelman: structured proportionality; provisions valid in circumstances

Unlike the other Justices, Edelman J did not determine whether ss 56 and 67 of the EM Act were valid in all their applications. That was because the provisions are 'open‑textured' and could be disapplied from any application which would be invalid ([220]). Instead, his Honour considered that it was only necessary to determine whether ss 56 and 67 of the EM Act were valid 'in their application to circumstances that encompass a general type of direction which includes the [Directions]' ([220]). (Indeed, his Honour cautioned that, in the case of open textured provisions, it would 'rarely be appropriate for a court to speculate upon whether the provisions are valid in all their applications, including hypothetical circumstances that are not before the court' ([219]).)

Justice Edelman acknowledged the 'severe discriminatory effects' that ss 56 and 67 may have in their application to a state of emergency constituted by the occurrence of a hazard in the nature of a plague or an epidemic ([282]). However, his Honour held that the provisions did not, in that application, exceed the threshold of reasonable necessity because of the 'significant restrictions' in the EM Act on their exercise ([283]-[286]). On the question of adequacy in balance, Edelman J held that the purpose of ss 56 and 67 is 'plainly sufficient to justify even the deep and wide burden that the application of those provisions can place upon the freedom prescribed by s 92' ([291]).


The Court thus rejected the plaintiffs' challenge to the constitutional validity of ss 56 and 67 of the EM Act. As the plaintiffs did not seek to argue that the state of emergency declaration, or the Directions, were otherwise outside the powers conferred by the EM Act, the Court made no finding as to their validity.

The Commonwealth's legal team

The AGS Constitutional Litigation Unit (Andrew Buckland, Danielle Gatehouse, David Rowe and Arlette Regan) acted for the Attorney‑General, with the Solicitor‑General, Dr Stephen Donaghu KC, Perry Herzfeld SC, Mark Hosking and Sarah Zeleznikow, as counsel.

The text of the decision is available at:

  Back to top


Buckland, Andrew

Chief General Counsel

Constitutional Litigation

Continuing detention orders upheld

By a 5:2 majority, the High Court (Kiefel CJ, Keane, Bell, Steward and Edelman JJ; Gageler and Gordon JJ dissenting) upheld the validity of the continuing detention order (CDO) scheme in Div 105A of the Criminal Code(Cth) (the Criminal Code).

Minister for Home Affairs v Benbrika
High Court of Australia, 10 February 2021
[2021] HCA 4; 95 ALJR 166; 388 ALR 1; 286 A Crim R 64


The plaintiff, Abdul Nacer Benbrika, was convicted of 2 terrorist offences under ss 102.2(1) and 102.3(1) of the Criminal Code and sentenced to a period of 15 years imprisonment, concluding in November 2020.

In September 2020, the Minister for Home Affairs applied to the Victorian Supreme Court for a CDO in respect of Mr Benbrika to be in force from the date of its making until 5 November 2023 and an interim detention order to be in force from 5 November 2020. On 27 October 2020, Tinney J made an interim detention order.

On 2 October 2020, Mr Benbrika applied for an order reserving the following question for the consideration of the Court of Appeal of the Supreme Court of Victoria:

Is all or any part of Division 105A of the Criminal Code (Cth) and, if so, which part, invalid because the power to make a continuing detention order under section 105A.7 of the Code is not within the judicial power of the Commonwealth and has been conferred, inter alia, on the Supreme Court of Victoria contrary to Chapter III of the Commonwealth Constitution?

On the application of the Attorney‑General of the Commonwealth, that question was removed into the High Court under s 40 of the Judiciary Act 1903 (Cth).

Continuing detention orders

The object of Div 105A of the Criminal Code is to ensure the safety and protection of the community by providing for the continuing detention of terrorist offenders who post an unacceptable risk of committing serious offences under Pt 5.3 of the Criminal Code. A serious Pt 5.3 offence is an offence against Pt 5.3, the maximum penalty for which is 7 or more years of imprisonment.

Division 105A empowers the Supreme Court of a state or territory, on the application of the Minister, to order that a person who has been convicted of a terrorist offence be detained in prison for a further period after the expiration of his or her sentence of imprisonment. A Supreme Court can make a CDO if it is 'satisfied to a high degree of probability, on the basis of admissible evidence, that the offender poses an unacceptable risk of committing a serious Part 5.3 offence if the offender is released into the community' and if it is 'satisfied that there is no other less restrictive measure that would be effective in preventing the unacceptable risk'.

A CDO can be made in respect of a 'terrorist offender', being a person who:

  • has been convicted of a terrorism offence referred to in s 105A.3(1)(a)
  • is in custody, having been continuously in custody since being convicted of the terrorist offence, or is under a continuing or interim detention order
  • will be at least 18 years old at the expiration of their sentence.

The power to make a CDO is only enlivened in the last 12 months of the offender's sentence.

When determining whether to make a CDO, a court must have regard to:

  • the safety and protection of the community
  • any report received from a relevant expert in relation to the offender; and the level of the offender's participation in the assessment by the expert
  • the results of any other assessment conducted by a relevant expert of the risk of the offender committing a serious Pt 5.3 offence; and the level of the offender's participation in any such assessment
  • any report, relating to the extent to which the offender can reasonably and practicably be managed in the community, that has been prepared by the relevant State or Territory corrective services or any other person or body who is competent to assess that extent
  • any treatment or rehabilitation programs in which the offender has had an opportunity to participate; and the level of the offender's participation in any such programs.

The High Court's decision

The High Court delivered 4 judgments: a joint judgment of Kiefel CJ, Bell, Keane and Steward J; a judgment of Edelman J concurring in the result but for different reasons; and separate dissenting judgments of Gageler and Gordon JJ.

Those judgments disclosed broader disagreement between some members of the Court about fundamental Ch III principles, in 2 respects.

The Constitution limits governmental power to detain a person

Although Gageler and Gordon JJ dissented in the result, their Honours seemingly agree with the plurality that the decision of the High Court in Chu Kheng Lim v Minister for Immigration, Local Government and Ethnic Affairs and Anor (1992) 176 CLR 1 (Lim) established a constitutional principle (the Lim principle) protective of individual liberty. That principle operates as a constraint, generally, on government power to detain a person otherwise than as punishment for a criminal offence (see in particular Kiefel CJ and Bell, Keane and Steward JJ at [18], [19] and [28]; Gageler J at [71]; Gordon J at [138]).

In contrast, Edelman J denied this proposition, saying that there was 'insufficient constitutional foundation to expand the Lim principle from one which is concerned with the separation of powers' - that is, with which branch of government can exercise a power of detention - 'to one which is also founded upon the liberty of the individual and is a substantive constraint upon all legislative, executive and judicial power' ([215]).

Detention to protect the community from harm

The 6 judges who conceived of the Lim principle as rights‑protective diverged on the question of whether preventative detention of the kind in issue here was a permissible exception to that principle.

a. The plurality held that it was, on the basis that the detention had as its object the protection of the community from harm ([36], [47]), stating that the power in Div 105A was of the same character as the historically established power to detain mentally ill persons who posed a threat to the community ([36]).
b. Justice Gageler also recognised 'exceptional cases' in which non‑punitive detention would be valid ([75]-[76]), including to protect the community from harm, 'at least where the harm is grave and specific' ([79]). Mere prevention of commission of a criminal offence is not a legitimate non‑punitive objective, however. While protecting the community from terrorist acts would be a legitimate protective purpose, Gageler J held that the risk of commission of a serious Pt 5.3 offence could not be treated as a proxy for the risk of a terrorist act, because the category of serious Pt 5.3 offences includes conduct many steps removed from doing or supporting or facilitating any terrorist act.
c. Justice Gordon declined to decide whether there was a new exception to the principle in Lim for judicially ordered preventative detention to protect the public from serious harm of the kind described in the definition of a 'terrorist act' ([177]).

The judgments

Joint judgment (Chief Justice Kiefel and Justices Bell, Keane and Steward)

There were 2 key strands to Mr Benbrika's case. He argued that:

  • The power in Div 105A to make a CDO is not judicial in nature and therefore cannot be conferred on a court by the Commonwealth Parliament.
  • The imposition of detention under Div 105A is contrary to the Lim principle.

The joint judgment dismissed the former argument without much difficulty. The power to make a CDO is subject to the ordinary incidents of the exercise of judicial power ([11]). Chief Justice Kiefel and Bell, Keane and Steward JJ were not persuaded that the provision in Div 105A for annual review of a CDO at the instigation of the Minister undermined the finality required of a judicial order ([13]) or that the fact that a CDO determines new rights and obligations rather than resolving a controversy as to existing rights and obligations meant that the CDO power was non‑judicial in nature ([14]).

As regards the latter argument, the joint judgment drew out (at [18]) 2 distinct propositions set out in Lim, both of which are subject to exceptions:

  • first, the 'general proposition' (Lim at 28) that 'the power to order that a citizen be involuntarily confined in custody is, under the doctrine of the separation of judicial from executive and legislative powers enshrined in our Constitution, part of the judicial power of the Commonwealth entrusted exclusively to Ch III courts'
  • second, the 'principle' (Lim at 27) that 'the involuntary detention of a citizen in custody by the State is penal or punitive in character and, under our system of government, exists only as an incident of the exclusively judicial function of adjudging and punishing criminal guilt'.

Their Honours said that the 'general proposition' and exceptions to it did not need to be considered in answering the referred question because the CDO power was conferred on Ch III courts ([28]). What was crucial in this case was the operation of the Lim 'principle', and the joint judgment concluded that the kind of detention authorised by Div 105A had a 'protective purpose' rather than being penal or punitive and so fell within an exception to that principle ([36]). Notably, the joint judgment:

  • emphasised that it was a necessary but not sufficient condition of validity that Div 105A is non‑punitive: to be valid the power must 'as a matter of substance, … have as its object the protection of the community from harm' ([36]; see also [27])
  • was critical of the notion that the exceptions to the Lim principle might be confined by history, noting that '[t]errorism poses a singular threat to civil society' ([36])
  • observed that Mr Benbrika's argument would entail that either no arm of the Commonwealth government could authorise preventative detention of a person to protect the community from terrorist offending; or that only the executive (and not a court) could do so - a 'distinctly unattractive' argument ([30])
  • found that the protective purpose of the CDO scheme was not denied by the fact that its operation was confined to persons previously convicted and imprisoned for terrorist offences or by the absence of special provision for the treatment and rehabilitation of detainees ([39])
  • accepted that conduct which would constitute a 'serious Pt 5.3 offence' for the purposes of Div 105A would almost invariably pose a real threat of harm to the community ([46]).

Separate concurring judgment of Justice Edelman

Justice Edelman concurred in the result - that Div 105A is not invalid on Ch III grounds - but reached that result by a very different route.

Justice Edelman first explained (at [181]) that the 'essential point' established in Lim was that 'subject to limited historical exceptions … the involuntary detention of a person as an incident of adjudging and punishing criminal guilt is exclusively a judicial power that is subject to Ch III of the Constitution' - that is, the Lim principle is limited to establishing that the power to punish is exclusively judicial.

In that respect his Honour found that Div 105A provided for a form of 'protective punishment' within the 'broad sense' of punishment he considered to be contemplated by Lim ([182]). It was therefore a form of judicial power that could validly be conferred on a Ch III court.

However, Edelman J did recognise that a power to impose punitive detention could not be conferred on a court if the power could not be exercised judicially. That would be the case if the exercise of the power 'would always cause injustice and if the power lacks justification' ([224]) (with the question of justification being determined through a form of structured proportionality analysis ([226]; cf [15] above)). That was not the case with Div 105A ([234]-[237]).

It therefore followed that, because the power to make a CDO was conferred on a court, was judicial in character and was to be exercised judicially, Div 105A did not contravene Ch III of the Constitution ([239]).

Justice Edelman was critical of arguments that some 'serious Part 5.3 offences' did not necessarily involve a risk of harm to the community. His Honour doubted that a stable distinction could be drawn between conduct that was wrong because it was necessarily harmful and conduct that might be wrong only because it was prohibited by the criminal law ([228]–[231]).

Partial dissenting judgment of Justice Gageler

Justice Gageler reiterated statements from earlier cases that detention in custody otherwise than as punishment of criminal guilt is only permissible if it is 'reasonably capable of being seen as necessary for a legitimate non‑punitive objective' (at [78]; cf [21] below). That was the case with the preventative powers in issue in Fardon v Attorney‑General (Qld) (2004) 223 CLR 575 and Thomas v Mowbray (2007) 233 CLR 307, which were appropriately targeted at the protection of public safety. It was on this basis that each could validly be exercised as part of the Commonwealth's judicial power. In contrast, his Honour considered the power in Div 105A to be 'designedly different' because the breadth of the 'serious Part 5.3 offences' with which that Division is concerned includes conduct 'many steps removed from doing or supporting or facilitating any terrorist act' ([92]–[93]). To that extent only, Gageler J held that Div 105A contravenes the Lim principle ([102]) but otherwise has a valid, distributive, operation ([101]).

Justice Gageler distinguished between terrorist offences where the risk of harm to the community arose straightforwardly from the prohibited conduct (for example, the offence of engaging in a terrorist act) and 'prophylactic' offences where the risk to the community would crystallise only if further steps were taken beyond the prohibited conduct (for example, the offence of taking steps to become a member of a terrorist organisation) ([55]–[56], [93]).

Dissenting judgment of Justice Gordon

Justice Gordon would have found Div 105A wholly invalid ([177]). Her Honour was concerned with the broad application of the Division ([163] and [175]), which she did not consider was ameliorated by the application of the 'no less restrictive measure' test ([176]). One aspect of this breadth was that 'serious Part 5.3 offences' need not involve an 'immediate harm' ([169]), although Gordon J did not apparently doubt that the commission of any of those offences occasioned some level of harm to society ([167]). Her Honour was also critical (as she had been in Vella v Commissioner of Police (NSW) (2019) 93 ALJR 1236 at [188]) of arguments which referred to what had been said in previous Ch III cases without what her Honour considered sufficient attention to the 'constitutional root' of relevant principles and their application to the specific circumstances of the present case.

The Commonwealth's legal team

The AGS Constitutional Litigation Unit (Andrew Buckland and David Rowe) acted for the Commonwealth Attorney‑General (intervening), with the Solicitor‑General, Dr Stephen Donaghue KC, and Mark Hosking as counsel.

The AGS Law Enforcement team (Rachel Deane, Elle Addams and Rebecca Verdon) acted for the Minister for Home Affairs, with Andrew Berger KC of AGS as counsel.

The text of the decision is available at:

  Back to top


Buckland, Andrew

Chief General Counsel

Constitutional Litigation

Some Aboriginal and Torres Strait Islander non-citizens are not 'aliens'

By a 4:3 majority, the High Court (Bell, Nettle, Gordon and Edelman JJ; Kiefel CJ, Gageler and Keane JJ dissenting) has held that 'Aboriginal Australians', understood according to the tripartite test in Mabo v Queensland [No 2] (1992) 175 CLR 1 at 70, are not 'aliens' within the meaning of that word in s 51(xix) of the Constitution. The conclusion of the majority was summarised by Bell J in these words ([81]):

I am authorised by the other members of the majority to say that although we express our reasoning differently, we agree that Aboriginal Australians (understood according to the tripartite test in Mabo [No 2]) are not within the reach of the 'aliens' power conferred by s 51(xix) of the Constitution.

Love v Commonwealth; Thoms v Commonwealth
High Court of Australia, 11 February 2020
[2020] HCA 3


Mr Love was born in Papua New Guinea and is a citizen of that country. Mr Thoms was born in New Zealand and is a citizen of that country. Neither are Australian citizens. Both have lived in Australia for substantial periods since childhood but did not seek to become Australian citizens. Both have, amongst their biological ancestors, Aboriginal Australians.

Further, on the facts agreed by the parties to the litigation:

  • Mr Thoms identifies as a member of the Gunggari People of Queensland and is accepted by other Gunggari People as a member of the Gunggari People
  • the Gunggari People (Mr Thoms included) are common law holders of native title (recognised by determinations made by the Federal Court of Australia)
  • Mr Love identifies as a descendant of the Kamilaroi group and is recognised as such by at least one Elder of that group.

Mr Love and Mr Thoms (the plaintiffs in this litigation) were each convicted of an offence for which a sentence of imprisonment of 12 months or more is provided. Their visas were therefore cancelled by a delegate of the Minister for Home Affairs under s 501(3A) of the Migration Act 1958 (Cth) (Migration Act). Upon cancellation of their visas the plaintiffs both became 'unlawful non‑citizens' and liable to be removed from Australia under the provisions of the Migration Act. These provisions are supported by s 51(xix) of the Constitution, which confers on Parliament the power to make laws with respect to 'naturalization and aliens'. The issue was whether, in light of the plaintiffs' claims that they are not 'aliens' by virtue of being Aboriginal Australians, the provisions of the Migration Act applying to 'unlawful non‑citizens' extend to them.

Although the Court was divided as to the result in the cases, there was one fundamental point of agreement between all 7 justices. All 7 justices endorsed the observation of Gibbs CJ in Pochi v Macphee (1982) 151 CLR 101 at 109 that:

[T]he Parliament cannot, simply by giving its own definition of 'alien', expand the power under s 51(xix) to include persons who could not possibly answer the description of 'aliens' in the ordinary understanding of the word.

The case turned on the application of that dictum to the position of Aboriginal Australians.

The High Court's decision

Majority (Justices Bell, Nettle, Gordon and Edelman)

Justice Bell

For Bell J, whose reasons were fairly brief ([51]):

The issue in these special cases is whether, as the plaintiffs assert, Aboriginal Australians are persons who cannot possibly answer the description of 'aliens' in the ordinary understanding of the word.

In her Honour's view, the issue was novel ([63]). She noted that the plaintiffs' argument depended on the 'incongruity of the recognition by the common law of Australia of the unique connection between Aboriginal Australians and their traditional lands, with finding that an Aboriginal Australian can be described as an alien within the ordinary meaning of that word' ([71]). Her Honour considered this argument had force. She concluded (at [74], emphasis added):

The position of Aboriginal Australians is sui generis. Notwithstanding the amplitude of the power conferred by s 51(xix) it does not extend to treating an Aboriginal Australian as an alien because, despite the circumstance of birth in another country, an Aboriginal Australian cannot be said to belong to another place.

As to whether each of the plaintiffs in these 2 cases is an 'Aboriginal Australian', her Honour considered that to be a question of fact ([75]).

In Mabo v Queensland [No 2] (1992) 175 CLR 1 (Mabo [No 2]) at 70, Brennan J (Mason CJ and McHugh J agreeing) said, 'Membership of the indigenous people depends on biological descent from the indigenous people and on mutual recognition of a particular person's membership by that person and by the elders or other persons enjoying traditional authority among those people'.

Her Honour held that, if a person meets the tripartite test in Mabo [No 2], that person will be an Aboriginal Australian for the purposes of the aliens power. That said, her Honour does not appear to rule out the possibly that 'a person who is not within the Mabo [No 2] test may nonetheless establish that he or she is an Aboriginal Australian' ([80]). It follows from her Honour's reasoning that Mr Thoms, whom the parties in effect agreed meets the tripartite test, is an Aboriginal Australian and outside the reach of s 51(xix).

As to Mr Love, the facts before the Court fell short of clearly establishing that Mr Love meets the third part of the tripartite test. The facts (relevantly) stated only that one 'elder' of the 'Kamilaroi tribe' recognises him as a member of the Kamilaroi group. However, having regard to the facts alleged by Mr Love in his pleaded case and apparently on the basis that the Commonwealth had not contested that Mr Love is an Aboriginal Australian, Bell J considered that she should proceed on the basis that Mr Love is an Aboriginal Australian within the meaning of the tripartite test ([78]-[79] and her Honour's answers at [82] to the questions in the special case).

Justice Gordon

Like Bell J, Gordon J considered that the position of Aboriginal Australians with regard to the aliens power in s 51(xix) was not covered by existing authority ([294]).

In considering the correct interpretation of s 51(xix), Gordon J firmly rejected the notion that 'aliens', which is a constitutional term, is an antonym of 'citizen', which is merely a statutory term ([295], [300]). For her Honour, the constitutional term 'aliens' has a basal sense which is independent of any statute. Indeed, the basal sense of the term 'aliens' is logically anterior to, and controlling of, what a statute about aliens can validly provide. ([327]-[330]).

Justice Gordon described the constitutional sense of the term 'aliens' this way ([296], [301]-[302], original emphasis):

The constitutional term 'aliens' conveys otherness, being an 'outsider', foreignness ...

… It describes a person's 'lack of relationship with a country'… These concepts were, and remain, intrinsic to the constitutional word 'aliens'.

Her Honour accepted that '[i]t may be that, in most cases, someone who does not hold Australian citizenship is within the reach of the aliens power' ([304], emphasis in original). But not all cases: the previous High Court decisions which might be read as suggesting otherwise 'are to be understood as laying down a guiding principle, not an absolute rule' ([311]).

Applying the constitutional sense of aliens as 'others' or persons who 'belong to another place', her Honour was of the view that 'Aboriginal Australians occupy a unique or sui generis position in this country' ([333], specifically [335]):

Aboriginal Australians are not outsiders or foreigners – they are the descendants of the first peoples of this country, the original inhabitants, and they are recognised as such. None of the events of settlement, Federation or the advent of citizenship in the period since Federation have displaced the unique position of Aboriginal Australians.

There are several other passages in her Honour's reasons which emphasise the spiritual connection of Aboriginal Australians to the land and waters that now make up the territory of Australia ([289]-[290], [298] and [340]).

As to the position of the 2 plaintiffs in the case, Mr Thoms clearly met the tripartite test, but the position of Mr Love was 'more complex'. Relying on the position of the Commonwealth in not challenging the status of Mr Love as an Aboriginal Australian, her Honour, like Bell J, determined the case on the basis that both plaintiffs were Aboriginal Australian ([387]-[388], [390]).

Justice Edelman

Justice Edelman readily ascribed an 'essential meaning' to the word 'aliens' in s 51(xix) of the Constitution: 'a foreigner to the Australian political community' ([393]; see also [404]). It followed that, for his Honour, a person cannot be an alien if she or she is 'a "belonger" to the political community' ([394]).

As with Gordon J, Edelman J firmly rejected the proposition that aliens and citizens are the obverse of each other ([433]-[437]). True it is that being a statutory 'citizen' is sufficient to make one a member of the political community and therefore not an alien ([432], [437]). However, citizenship is not the only or exhaustive test. The ultimate test for alienage is whether one falls within or without the 'essential meaning' of the word 'alien' & that is, whether or not one is a 'belonger to the Australian political community' ([437]-[438]). In applying this test, the concept of 'allegiance' provides little assistance. Allegiance is a 'consequence' or 'characteristic' of being a member of the political community, not a test for membership ([429]).

To his Honour, ties to physical territory can be a pertinent consideration: 'the legal and political considerations of political community have always been heavily influenced by metaphysical ties to physical territory' ([439]). This is because ([438]):

[A political community] is not a thing that exists in space. It is a metaphysical construct that describes a group of people who belong to a defined place or territory, here the land of the Australian state …

Ultimately, like Bell and Gordon JJ, Edelman J placed great weight on the 'powerful personal attachment to land' ([447]) of Aboriginal persons ([450], [451]) (emphasis in original):

Just as the attachment to country that arises from citizenship of parents and birth in the defined territory can be an underlying basis for membership of political community independent of citizenship legislation, so too are the powerful spiritual and cultural connections between Aboriginal people and the defined territory of Australia.

The very words 'Aboriginal' and 'indigenous', ab origine or 'from the beginning', enunciate a historical, and original, connection with the land of Australia generally. The sense of identity that ties Aboriginal people to Australia is an underlying fundamental truth that cannot be altered or deemed not to exist by legislation in the same way that changing legislative definitions of citizenship cannot alter the fundamental truth underlying identity that is shaped by the core combined norms that metaphorically tie a child to Australia by birth and parentage.

While his Honour considered that the tripartite test was an acceptable means of identifying members of a particular sub‑group of Indigenous people who enjoy continuing connection with particular land (458]):

it is not set in stone, particularly as an approach to determining Aboriginality as the basis for those fundamental ties of political community in Australia which are not dependent upon membership of a particular sub‑group.

His Honour determined that it was not necessary to consider issues that might surround the application of the tripartite test, as there was an absence of contest on this point. It was, in his Honour's view, 'plainly open to treat Mr Love as Aboriginal'. In the case of Mr Thoms, being an agreed fact, it is 'plain' that he is an Aboriginal person ([462]).

Justice Nettle

The reasoning of Nettle J followed a significantly different path from that of the other 3 majority justices. Unlike them, Nettle J conceived of 'alien' status in terms of the traditional common law concepts of allegiance and protection.

After a detailed and scholarly outline of the legal and historical development of the concepts of 'alienage', 'allegiance' and 'protection' at common law, Nettle J summarised the state of affairs at the time of federation as follows ([249]):

the ordinary understanding of alienage - and thus the connotation of the word 'aliens' in s 51(xix) of that Constitution - then [that is, at Federation] depended on the want of a permanent allegiance to and protection by the sovereign of the Commonwealth, formerly regarded as an undivided Crown but now identified as the Crown in right of Australia.

The relevant content or 'indicia' of the reciprocal concepts of 'allegiance' and 'protection' have always been 'inherently contestable' ([251]). So it was, his Honour acknowledged, at the time of federation. Accordingly, 'Parliament was given power to select the relevant indicia'. Since 1948 Parliament has exercised that power in citizenship statutes ([251]). The indicia usually selected by Parliament have been place of birth, or nationality of a parent, or some mixture of those. Those persons who possess the statutory indicia owe allegiance to the sovereign and can at all times claim the sovereign's protection; those who do not are 'aliens' and can be dealt with by Parliament as such.

There are limits to how far Parliament can go in this respect. In his Honour's view, it is possible for some persons to have so strong a claim to the protection of the sovereign (and therefore to owe allegiance to that sovereign) that it is beyond Parliament's power to treat them as aliens ([252]; see also [260]).

His Honour acknowledged that past decisions of the Court appear to indicate that Parliament may classify any person who is born abroad or is a foreign citizen - including a person meeting the just mentioned definition of Aboriginality - as an 'alien'. However, that conclusion struck his Honour as 'intuitively … at odds with the growing recognition of Aboriginal peoples as "the original inhabitants of Australia" and the ubiquity of Australian dual citizens' ([263]). This (to his Honour) counter‑intuitive conclusion was a sign to stop and reconsider the matter from first principles.

In his reconsideration, Nettle J focused on the holding in Mabo [No 2] that the common law recognises, and has always recognised, native title rights: rights and interests in land that arise from the traditional law and custom of Australia's Indigenous peoples ([268]). Going deeper, his Honour found that there is something 'more fundamental' than the common law's recognition of native title and, indeed, 'logically anterior' to it. It is that the common law gave 'recognition' to the Aboriginal societies from which those traditional laws and customs emerged ([269]).

Justice Nettle placed great weight on one particular aspect of traditional laws and customs ([271]):

for present purposes, the most significant of the traditional laws and customs of an Aboriginal society are those which allocate authority to elders and other persons to decide questions of membership. Acceptance by persons having that authority, together with descent (an objective criterion long familiar to the common law of status) and self‑identification (a protection of individual autonomy), constitutes membership of an Aboriginal society: a status recognised at the 'intersection of traditional laws and customs with the common law'.

'That status', said his Honour, 'is necessarily inconsistent with alienage' ([272]). The common law cannot on the one hand recognise Aboriginal societies and in particular their autonomy under traditional law and custom to determine questions concerning membership and on the other hand confer on the members of those Aboriginal societies a legal status ('alien') that allows the sovereign to treat that recognition at naught and render members of those Aboriginal societies liable to deportation ([272]):

Permanent exclusion from the territory of Australia would have abrogated the common law's acknowledgment of traditional laws and prevented the observance of traditional laws and customs by persons excluded (and, depending on their positions in society, also by others). To classify any member of such an Aboriginal society as an alien would have been to recognise that the Crown had power to tear the organic whole of the society asunder …

The Crown must therefore be taken, at common law, to owe a 'unique obligation of protection' to members of Aboriginal societies ([272], [278]). Reciprocal to that duty of protection on the part of the Crown is a duty of permanent allegiance to the Crown by those to whom the duty of protection is owed ([279]). Thus members of traditional Aboriginal societies possess the indicia of allegiance and protection. It follows that they are not aliens.

It further follows that, while some Aboriginal people cannot be constitutional 'aliens', some Aboriginal people (that is, Aboriginal people who are not members of a traditional Aboriginal society) are capable of being constitutional 'aliens' (see, for example, [282]). That is, 'traditional' in sense articulated in Members of the Yorta Yorta Aboriginal Community v Victoria (2002) 214 CLR 422, of belonging to an Aboriginal society that has, since the acquisition of sovereignty by the Crown, remained continuously united in and by its acknowledgement and observance of traditional laws and customs ([278]).

Mr Thoms is a native title holder. He must therefore be, inferred Nettle J, a member of an Aboriginal society whose laws and customs have relevantly maintained a continuous existence and vitality since the Crown's acquisition of sovereignty. He is therefore not an 'alien'. In relation to Mr Love, however, Nettle J considered it necessary for the Federal Court of Australia to find the relevant facts and, on that basis, to determine the matter according to law ([287]-[288]).

Minority (Chief Justice Kiefel and Justices Gageler and Keane)

Common to the reasoning of each member of the minority (and thus a principal point of divergence between the majority and the minority) is an acceptance that 'alien' and 'citizens' are, at least generally speaking, antonyms. Every person is either one or the other. It followed that if any given Aboriginal Australian does not meet the statutory criteria for citizenship then, so long as those criteria are constitutionally valid, he or she must be an 'alien'.

Chief Justice Kiefel

Her Honour could see no basis in the text of s 51(xix) to conclude that Parliament's power to make laws with respect to 'aliens' operates any differently for Aboriginal Australians than it does for any other persons. The question was therefore whether any constitutional implication could be found to that effect ([8]). Such an implication is not to be lightly inferred (14):

[i]t is a serious matter to deny a power which is fundamental to the structure of the Constitution and the governance of Australia. The basis for an implication having this effect must be pellucidly clear.

Chief Justice Kiefel noted that on the plaintiffs' argument Aboriginality turns on (inter alia) 'recognition by the Elders or other persons having traditional authority amongst those people' ([23]). Her Honour was not prepared to extend this principle into the law of alien‑status. It would have the effect of attributing to a given Aboriginal group 'the kind of sovereignty which was implicitly rejected by Mabo [No 2]' ([25]; see also [37]).

As to the point that was so influential for Bell, Gordon and Edelman JJ - the special connections of various groups of Aboriginal peoples to particular areas of land and waters in Australia - her Honour considered it erroneous to import notions of connection to land 'from an entirely different area of law' (that is, the law of native title) 'to answer questions of a constitutional kind about the relationship between an Aboriginal group and its members and the Australian body politic'. Her Honour considered this to be 'wrong as a matter of law and of logic' ([31]).

First, such connections to land that are of importance in native title law are connections to particular places within the territory of Australia, not to 'the territory of the whole of Australia' ([30]). Second, the 'belonging' to the land that is relevant to the law of native title 'is not the "belonging" spoken of in the constitutional sense' ([32]-[33]):

In the constitutional context 'belonging' refers to the formal legal relationship between a person and the community or body politic in question. In Australia it is apt to describe the connection between a citizen and the body politic. It reflects a conclusion reached about that relationship rather than a premise upon which the relationship may be founded.

Nor could her Honour find a constitutional implication pointing to a special constitutional category of persons to whom s 51(xix) does not apply ([39]-[46]).

Her Honour concluded ([46]):

What is the source of this proposed new principle if it is not the common law of native title? Clearly enough it is of such a nature that it may not be altered either by statute or by the Constitution. Because it is immutable it might be understood to bear the characteristics of a higher principle of which natural law might conceive. But such conceptions are generally not regarded as consistent with constitutional theory. And they are regarded by some as antithetical to the judicial function since they involve an appeal to the personal philosophy or preferences of judges.

Justice Gageler

In Gageler J's consideration of the word 'aliens' in s 51(xix), his Honour drew a distinction between 2 different types of words that appear in s 51. There are certain words which have a fixed, objective meaning, 'lighthouse' being the classic example. These are to be contrasted with certain other words that describe a legal status: 'marriage', 'bankruptcy', 'aliens' and so on. Each of these latter words refers to a status whose meaning and content is to a significant extent shaped by legislation. It is inherent in the nature of the legal status of being 'married' that one has met certain formal criteria laid down by Parliament. It is thus an error for lawyers to treat the word 'alien' in the same way as the word 'lighthouse' ([88]):

the nature of the legislative power to determine who has and who does not have the legal status of alienage is wholly inconsistent with the notion that a person's status as an alien or non‑alien falls to be determined independently of the exercise of the power as a question of constitutional fact.

In Gageler J's view, s 51(xix) confers a scope of legislative choice to the Commonwealth Parliament to prescribe criteria for the determination of who is to have from birth the status of an Australian citizen (and thus a member of the body politic of the Commonwealth of Australia) as distinct from a non‑citizen or alien ([94]). It is not the case that the scope of legislative choice is 'entirely unconstrained'. However, such constraints as exist need to be sourced in the Constitution itself.

Justice Gageler observed that 'members of Aboriginal and Torres Strait Islander societies have never been understood to fall outside the standard common law or statutory rules that have from time to time governed the distinction between a British subject or Australian citizen, on the one hand, and an alien, on the other hand' ([103]). By operation of those rules, Aboriginal people living in any part of Australia at the time of the acquisition of sovereignty by the British Crown immediately became British subjects and not aliens. By operation of those same rules, Aboriginal persons born in any part of colonial Australia thereafter were also British subjects, not aliens ([104); see also [110]).

The plaintiffs' argument, however, depended on something different from the application of the standard common law rules about alienage. The plaintiff's argument was said to depend on the 'common law's recognition of the continuing existence of self‑determining indigenous societies maintaining a spiritual and cultural connection with the land within Australia through observance of traditional laws and customs'. This gave rise to a 'belonging' to the land that is 'deep and enduring' and, ultimately 'inconsistent with the treatment of members of those societies as strangers to that land or foreigners to Australia' ([117]).

His Honour was unable to accept this argument ([127]-[128]):

Morally and emotionally engaging as the plaintiffs' argument is, the argument is not legally sustainable. The common law antecedents of the Constitution provide no basis for extrapolating from common law recognition of a cultural or spiritual connection with land and waters within the territory of the Commonwealth to arrive at constitutionally mandated membership of or connection with the political community of the Commonwealth. The considerations which informed the common law development in Mabo cannot be transformed by any conventional process of constitutional interpretation or implication into a constitutional limitation on legislative power.

In order to be accepted, the plaintiffs' argument would require 'supra‑constitutional innovation' ([133]). Further, the argument would 'concede to a non‑constitutional and non‑representative non‑legally‑accountable sub‑national group a constitutional capacity greater than that conferred on any State parliament' ([137]) and would 'inject an element of indeterminacy into the administration of the legal status of alienage' ([138]).

Justice Keane

Justice Keane's principal conclusion was that the aliens power draws no distinctions between Aboriginal and non‑Aboriginal persons ([147]):

Section 51(xix) cannot be read as if it distinguished between persons of Aboriginal descent on the one hand and persons descended from other races on the other, so that the former are excluded from its scope. Each plaintiff is within the scope of s 51(xix) no less than any other child who is born abroad of an Australian parent and does not apply for Australian citizenship.

His Honour could see 'a strong moral case for special recognition of Aboriginal people in the Constitution', but public debate about that is 'necessary precisely because the Constitution, in its current terms, does not have that effect' ([178]). The reading of s 51(xix) advanced by the plaintiffs 'does no little violence' to the Constitutional text ([181]).

The special connection of Aboriginal persons with land and waters within Australia was, in his Honour's view, not relevant to the law of alien status ([195]):

The plaintiffs' argument confuses the body politic that was brought into existence at Federation with lands and waters, parts of which were occupied by particular Aboriginal groups long before that body politic came into being. The plaintiffs' argument also confuses the spiritual connection of an indigenous person to particular lands and waters with a connection to the body politic that is inconsistent with alienage.

After listing a number of difficulties, practical and conceptual, with the tripartite test as a test for alien status, his Honour also found the tripartite test problematic from the point of view of sovereignty ([197]):

to suggest that members of Aboriginal groups have authority to make choices that bind the Commonwealth of Australia is to attribute to those persons a measure of political sovereignty. To assert that the ordinary application of laws made pursuant to s 51(xix) of the Constitution to foreign citizens born outside Australia such as the plaintiffs is displaced as a result of recognition by members of the Aboriginal group from which they claim descent, is to assert an exercise of political sovereignty by those persons. 

Past decisions of the Court in native title cases had rejected the argument that the Aboriginal people of Australia have sovereignty adverse to the Crown ([199]).

Finally, his Honour strongly rejected the argument that 'persons of the Aboriginal race' owe a permanent allegiance to the Crown as the reciprocal of an obligation of special protection owed by the Crown to the Indigenous people of the continent ([214]-[215]). His Honour said of that argument ([217]):

One can readily understand that the plaintiffs would grasp at any straw that may save them from what they might understandably perceive as a harsh overreaction by the executive government to their circumstances; but the absence of a cogent explanation as to how permanent allegiance may lawfully be repudiated invites the query whether other persons of Aboriginal descent not confronted with the same immediate difficulties would so blithely embrace the rank paternalism that suffuses this argument. 

Outcome and subsequent challenge to the correctness of the decision

In the result, a majority of the High Court held that Mr Thoms is an Aboriginal Australian according to the tripartite test. It follows from that Mr Thoms cannot be deported from Australia under the provisions in the Migration Act 1958 (Cth) that provide for the removal from Australia of 'unlawful non‑citizens'.

The majority was unable to agree on whether Mr Love is an Aboriginal Australian for the purposes of the tripartite test in Mabo [No 2]. Three justices were sufficiently persuaded that Mr Love should be taken to be an Aboriginal Australian, in large part because the Commonwealth put no submission to the contrary. The fourth member of the majority (Nettle J) considered the facts before the Court to be insufficient to decide the point. In the absence of any agreement on the parties on the issue, Mr Thoms' status as an Aboriginal Australian will need to be resolved by the Federal Court on remittal.

In November 2021, about 21 months after the decision in Love and Thoms, Justice Keane made orders (on the application of the Commonwealth Attorney‑General) removing into the High Court an appeal by the relevant Commonwealth Ministers from the decision of Derrington J in Montgomery v Minister for Immigration, Citizenship, Migrant Services and Multicultural Affairs [2021] FCA 1423 (Montgomery). The case concerned Mr Montgomery, who was born in New Zealand, is a New Zealand citizen and is not an Australian citizen. Following cancellation of Mr Montgomery's visa, he was taken into immigration detention pursuant to provisions of the Migration Act. Mr Montgomery, relying on the decision in Love and Thoms, claimed that his detention was unauthorised because he is not an 'alien' but an Aboriginal man, arising from his cultural adoption by the Munanjali people of southern Queensland. Mr Montgomery claimed that he is a Munanjali man and does not need to have a biological Aboriginal ancestor to be Aboriginal according to the laws and customs of the Munanjali people. The Commonwealth Ministers accepted that Mr Montgomery identifies as an Aboriginal person and that, following a process of cultural adoption, he had been recognised as a Munanjali man by persons enjoying traditional authority amongst those people. However, the Ministers did not accept that Mr Montgomery satisfies the tripartite test endorsed by the majority in Love and Thoms because he is not biologically descended from any Aboriginal or Torres Strait Islander person. The Ministers and the Commonwealth Attorney‑General (intervening ) also submitted that the decision in Love and Thoms was wrong and should be overruled.

Oral argument took place before 7 justices of the High Court on 6 and 7 April 2022. There was considerable argument as to whether the decision in Love and Thoms had a ratio decidendi. The Commonwealth Ministers and the Attorney‑General submitted that, notwithstanding Bell J's statement in Love and Thoms at [81], given the different course of reasoning taken by Nettle J, no ratio decidendi emerges from the reasons of the members of the majority in Love and Thoms. Mr Montgomery (and various persons intervening in the appeal in support of Mr Montgomery) submitted that Bell J's statement in Love and Thoms at [81] constituted the ratio decidendi of the case. The High Court reserved its decision.

On 28 July 2022, before the High Court had handed down its judgment in the matter, the Commonwealth Ministers filed a notice of discontinuance, thereby discontinuing the appeal in Montgomery.

The Commonwealth's legal team

The AGS Constitutional Litigation Unit (Gavin Loughton and Danielle Gatehouse) and the Dispute Resolution Group (Baden Powell) acted for the Commonwealth, with the Solicitor‑General, Dr Stephen Donaghue KC, Nick Wood and Julia Watson as counsel.

The text of the decision is available at:

  Back to top


Loughton, Gavin

Senior Executive Lawyer

Courts cannot make 'common fund orders' in representative proceedings

By a 5:2 majority, the High Court (Kiefel CJ and Bell, Keane, Nettle and Gordon JJ; Gageler and Edelman JJ dissenting) held that the power of the Federal Court and the Supreme Court of New South Wales (NSW Supreme Court) in representative proceedings to make 'any order the Court thinks appropriate or necessary to ensure that justice is done in the proceeding' does not extend to making a 'common fund order' (CFO).

BMW Australia Ltd v Brewster; Westpac Banking Corporation v Lenthall
High Court of Australia, 4 December 2019
[2019] HCA 45; (2019) 269 CLR 574; 374 ALR 627


These appeals, which were heard together in the High Court, were brought by the defendants in 2 unrelated representative proceedings (colloquially known as 'class actions'). One representative proceeding was in the Federal Court against BMW Australia, in relation to defective airbags in cars. The other representative proceeding was in the NSW Supreme Court against Westpac, for alleged breaches of statutory and financial obligations in the course of advising customers to purchase insurance policies. Both proceedings involved rights under Commonwealth statutes and so required each court to exercise federal jurisdiction.

In both proceedings, third‑party litigation funders were engaged to fund the claims. With a view to securing this ongoing financial support, the representative parties in each case asked each court to make a CFO.

A CFO is a particular type of court order that is made in favour of a litigation funder. The order is usually made early in the proceeding - that is, well in advance of the trial. A CFO provides that the litigation funder is entitled to receive a specified proportion of any moneys that are ultimately recovered. For example, the CFO in Westpac provided that the litigation funder was entitled to receive from the group members in the proceeding (assuming it succeeded) the lesser of either:

The CFO in Westpac was made under s 33ZF(1) of the Federal Court of Australia Act 1976 (FCA), which provides:

a. 3 times the litigation funder's expenditure 'on legal costs, disbursements, adverse costs orders and fees of the costs referee' or
b. 25% of 'the net resolution sum'.

In any proceeding (including an appeal) conducted under [Pt IVA of the FCA], the Court may, of its own motion or on application by a party or a group member, make any order the Court thinks appropriate or necessary to ensure that justice is done in the proceeding.

In BMW, the plaintiff applied for a CFO under s 183 of the Civil Procedure Act 2005 (NSW) (CPA) (which is in relevantly identical terms to s 33ZF of the FCA). The NSW Court of Appeal held that the NSW Supreme Court had the power to make an order in the form sought.

The High Court granted leave to appeal against both decisions and heard the appeals together for convenience, as they raised common issues.

The issues

Westpac and BMW (each the appellant before the High Court) attacked each court's power to make a CFO on constitutional and non‑constitutional grounds:

  • First, the appellants argued that, on the proper construction of the legislation regulating representative proceedings in each court (Pt IVA of the FCA - in particular, s 33ZF; and equivalent provisions in the CPA, including s 183), the court did not have power to make a CFO (the statutory construction argument).
  • Second, they argued that the provisions purport to permit the exercise of non‑judicial power and thus infringe Ch III of the Constitution (and the Kable principle (Kable v Director of Public Prosecutions (NSW) (1996) 189 CLR 51)in the case of s 183 of the CPA) (the Ch III argument).
  • Third, they argued that a CFO would result in a litigation funder acquiring the 'fruits of the action' (that is, a percentage of any favourable judgment or settlement sum) from the claimant group without 'just terms'. In this way, the appellants argued that a CFO would infringe the constitutional guarantee of just terms for an acquisition of property in s 51(xxxi) of the Constitution (the acquisition of property argument).

In the BMW appeal, there was an additional complication to the constitutional challenges in that the CFO was made under a state law which (the parties assumed) was 'picked up' by s 79 of the Judiciary Act 1903 (Cth).

The Commonwealth Attorney‑General intervened to make submissions in support of a CFO being able to be validly made under the relevant provisions. The Attorneys‑General for Victoria, Queensland and Western Australia also intervened.

The High Court's decision

In a joint judgment, Kiefel CJ and Bell and Keane JJ held that, as a matter of statutory construction (leaving aside the constitutional challenges), neither s 33ZF nor s 183 empower the court to make a CFO. Justices Nettle and Gordon substantially agreed with the joint judgment, but each gave separate reasons. In light of their conclusion on the statutory construction issue, it was not necessary for the majority Justices to consider the constitutional issues.

Justices Gageler and Edelman each delivered separate dissenting judgments, holding that the courts were empowered to make a CFO. Their Honours briefly considered and dismissed the constitutional challenges to the provisions.

Statutory construction issue: do ss 33ZF and 183 empower the court to make a CFO?

Joint judgment of Chief Justice Kiefel and Justices Bell and Keane

The joint judgment concluded that a CFO fell outside the terms of the 'broad' statutory power to make an order 'appropriate or necessary to ensure that justice is done in the proceeding' ([3], [19], [47], [48]). This was because their Honours considered that a CFO is not an order 'ensur[ing] that the proceeding is brought fairly and effectively to a just outcome' but, rather, an order 'in favour of a third party with a view to encouraging it to support the pursuit of the proceeding' ([47]).

The joint judgment concluded that s 33ZF did not support making a CFO for a number of textual and contextual reasons, including:

  • As a matter of 'ordinary and natural meaning', s 33ZF authorised an order 'apt to advance the effective determination by the court of the issues between the parties to the proceeding', but a CFO providing a 'financial inducement to support the proceeding' did not fall within that description ([50]).
  • There was a 'conceptual difficulty of the absence of criteria to guide the exercise of discretion' whether to make a CFO ([59]).
  • The Act contemplated that the solution to the problem of uneconomic litigation was not to make a CFO but to 'halt the representative proceeding' ([65]).
  • Section 33ZF was a supplementary or 'gap filling' provision containing words of limitation and, therefore, it should not be 'relied upon as a source of power to do work beyond that done by the specific provisions which the text and structure of the legislation show it was intended to supplement' ([70]).

The joint judgment analysed the purpose of Pt IVA of the FCA ([82]-[94]) in considering the 2 policy aims said to support the making of a CFO (access to justice and avoiding wasted costs of 'book building'). Their Honours observed that access to justice could be achieved by making a different kind of funding order - namely, a 'funding equalisation order', which takes as its starting point the actual costs incurred in funding the litigation ([88]-[90]).

As their Honours held that s 33ZF of the FCA (and s 183 of the CPA) did not support making a CFO, the joint judgment did not need to address the 2 constitutional arguments ([4]).

Other Justices in the majority (Justices Nettle and Gordon)

Justice Nettle noted that there were 'cogent arguments either way' ([122]) but ultimately agreed with the reasons and orders in the joint judgment ([125]).

Justice Gordon essentially agreed with the joint judgment ([135], [138]-[140], [146], [149], [154], [166]) and joined in the orders their Honours proposed ([170]). In considering the policy justifications for making a CFO ([167]), her Honour noted that asking and answering questions about the economics of a representative proceeding was not 'appropriate or necessary in ensuring that justice is done in a proceeding' ([158]).

Dissenting judgments (Justices Gageler and Edelman)

Justice Gageler held that 'a CFO can be thought appropriate or necessary to ensure that justice is done in the proceeding, and so can be made in the exercise of the power conferred by [s 33ZF of the FCA and s 183 of the CPA]' ([106]). His Honour held that the powers in ss 33ZF and 183 extend to ensuring 'substantive justice' is done as between the representative party and other group members, including by ensuring the expenses of litigation are shared fairly ([109]-[111]). His Honour rejected the proposition that an order which serves to shore up the commercial viability of the proceeding from the perspective of a litigation funder can have nothing to do with enhancing the interests of justice ([110]). Justice Gageler also rejected 'the suggestion that making a CFO requires the Court to embark on an inquiry which is beyond its institutional competence' ([115]).

Justice Edelman took a somewhat different approach, drawing analogies with the principles of unjust enrichment, maritime salvage and other branches of law in which courts make orders for reasonable remuneration in respect of unrequested work ([193], [199]). His Honour held that there was no reason in principle why a CFO could not be made at an interlocutory stage on a pre‑emptive basis in anticipation of a final order ([215]-[222]).

Constitutional issues

Only the dissenting judges (Gageler and Edelman JJ) addressed the appellants' constitutional arguments against the power to make CFOs. Their Honours dealt with those arguments briefly, noting that they were 'not strong' ([174]) and did 'not withstand scrutiny' or 'warrant elaborate responses' ([119]).

In rejecting the appellants' Ch III argument, Gageler J reasoned that s 33ZF of the FCA conferred power 'as an incident of strictly judicial proceedings, to be exercised by reference to the Court's assessment of the interests of justice' ([119]) and that such a power was clearly appropriate to the judicial branch of government ([119]). Justice Edelman gave similar reasons ([225]-[227]), adding that it was entirely within judicial power for courts 'to create new rights in the sense of recognising and giving effect to rights that differ from a previously settled understanding' ([225]).

In rejecting the acquisition of property argument, Gageler and Edelman JJ held that a CFO made under s 33ZF of the FCA (and s 183 of the CPA, as picked up and applied by s 79 of the Judiciary Act) would involve an 'adjustment of competing rights, claims or obligations of persons in a particular relationship' and, therefore, would not involve an 'acquisition of property' engaging s 51(xxxi) of the Constitution ([120], [230]).

The Commonwealth's legal team

The AGS Constitutional Litigation Unit (Gavin Loughton, Emily Kerr and Jamie Blaker) acted for the Commonwealth Attorney‑General, with Stephen Lloyd SC, David Hume and Kim Pham as counsel.

The text of the decision is available at:

  Back to top