Judicial agreements and disagreements in Alexander v Minister for Home Affairs
Sangeetha Pillai
21.09.2022
Since 2015, Australia has had controversial citizenship-stripping laws as a part of its national security toolkit. These laws apply to dual citizens deemed to have repudiated their allegiance to Australia by virtue of their activities, and were first introduced in response to an increase in citizens travelling overseas to serve as ‘foreign fighters’ for organisations like Islamic State.
Since its introduction, Australia’s citizenship stripping legislation has provided for a dual citizen to lose their Australian citizenship in two ways. If they were convicted of prescribed Criminal Code offences relating to terrorism or foreign incursions and recruitment, a ministerial discretion to strip citizenship was activated. The prescribed offences ranged from the very serious (e.g. committing a terrorist attack) to things that might involve no hostile intent (e.g. entering an area declared to be a no-go zone by the government). However, in some circumstances – where the citizen had acted in a foreign country, or had acted in Australia but since left the country – agencies were concerned that obtaining a conviction was impractical. To account for this, the legislation included a conduct-based citizenship-stripping mechanism. This allowed dual citizens in these circumstances to lose citizenship on the basis of having engaged in the conduct elements of the offences included within the conviction-based ground, without any conviction being necessary.
The first iteration of the citizenship-stripping laws provided that conduct-based citizenship deprivation would operate automatically. In 2020, following critique by the Independent National Security Legislation Monitor and various government departments, the laws were amended. The critical change in the revised scheme was that s 36B – the new conduct-based provision – no longer deprived people of their citizenship automatically. While it was still triggered by the conduct elements of the same offences that gave rise to conviction-based citizenship-stripping, s 36B instead created a ministerial discretion to revoke Australian citizenship where the Minister determined this was in the public interest.
In Alexander v Minister for Home Affairs [2022], the High Court found s 36B to be invalid in its entirety by a 6:1 majority (Steward J dissenting). For many, this was not an unexpected outcome: since the earliest days of Australia’s citizenship-stripping laws, multiple experts have warned that there was a likelihood that legislating for conduct-based denationalisation without conviction carried a serious risk of constitutional invalidity.
Indeed, the most remarkable aspect of the case may be the fact that it made it to the High Court at all. Alexander was stripped of his Australian citizenship on 2 July 2021. Litigation was initiated remarkably quickly, a mere eleven days later. Any further delay may have changed everything. Two days after proceedings were commenced, Alexander, who was in detention in Syria in a government-controlled prison associated with serious human rights violations, completely lost contact with his lawyers and family. The case proceeded with his sister acting as litigation guardian.
In this post, I unpack key aspects of the High Court’s decision in Alexander. I focus on the lines of agreement and disagreement amongst members of the Court with respect to the two issues that attracted the most consideration: whether s 36B infringed the separation of judicial power, and whether it fell within the scope of the naturalization and aliens power in s 51(xix) of the Constitution.
The separation of judicial power
All six majority judges found that s 36B was invalid because it infringed the separation of judicial power. While there were four separate majority judgments (a joint judgment by Kiefel CJ, Keane and Gleeson JJ, a judgment by Gageler J that largely concurred with the joint judgment, and separate judgments by Gordon and Edelman JJ), the entire majority adopted relatively consistent reasoning on this point.
There were two key issues at play: the scope of the principle in Chu Kheng Lim v Minister for Immigration, Local Government & Ethnic Affairs [1992] (Lim), and the question of whether s 36B infringed the Lim principle.
In Lim, the joint judgment of Brennan, Deane and Dawson JJ stated at [22] that:
There are some functions which, by reason of their nature or because of historical considerations, have become established as essentially and exclusively judicial in character. The most important of them is the adjudgment and punishment of criminal guilt under a law of the Commonwealth.
Leaving aside exceptional cases, Brennan, Deane and Dawson JJ said that subjecting a citizen to involuntary detention in custody is penal or punitive in character, and can only be done as part of the exclusively judicial function of punishing criminal guilt. They went on to say that citizens enjoy, at least in times of peace, a constitutional immunity against imprisonment other than by order of a court exercising judicial power (at [24]).
In the thirty years since the Lim principle was established, questions about its scope have overwhelmingly surfaced in cases relating to executive detention. The principle has been interpreted restrictively, and has developed somewhat of a reputation for being the kind of constitutional guard dog that has grown old but that rarely barks.
In Alexander, the Minister suggested that Lim should continue to be applied restrictively. While accepting that adjudging and punishing criminal guilt is exclusively judicial, the Minister argued that the Lim principle had no application to laws that did not impose detention in custody, such as s 36B.
Notably, all seven judges dismissed this restrictive reading. The entire Court found that the Lim principle is not confined to laws imposing detention in custody, and confirmed that there are at least some circumstances in which citizenship-stripping is punitive in nature, and therefore something that can only be imposed by a court.
The disagreement between the majority judges and Steward J related to whether s 36B itself was punitive. The majority judges, adopting relatively united reasoning, found that it was. They emphasised that citizenship-stripping is one of the harshest consequences that could be imposed on a person, and that historically, since as far back as ancient Roman times, denationalisation, banishment and exile had been recognised as punitive.
In the majority’s view, while s 36B had a stated purpose of community protection, it was correctly construed as a law that imposed denationalisation as retribution for conduct, and that in doing so it improperly employed an exclusively judicial process. The majority judges identified a range of things that were said to support this reading:
the fact that the statutory scheme identified the conduct giving rise to denationalisation under s 36B as conduct ‘incompatible with the shared values of the Australian community’ (Kiefel CJ, Keane and Gleeson JJ with Gageler J agreeing; Edelman J);
the fact that this conduct was defined by reference to past criminal conduct (Gageler J; Gordon J); and
the fact that the power to determine that particular conduct is extreme and reprehensible and to strip citizenship on the basis of that conduct both lay with the Minister (Edelman J), and that the Minister was not required to adhere to procedural fairness requirements when exercising these powers (joint judgment with Gageler J agreeing).
By contrast, Steward J found that denationalisation has never been an exclusively judicial function. He noted that denationalisation – as distinct from exile – was not originally possible under the common law, and was first introduced via legislation. This, he said, first occurred in circumstances that did not involve the exercise of judicial power or involve punishment: for example, early UK legislation stated that a person would become an alien if they became naturalised in a foreign state.
Moreover, Steward J disagreed with the majority’s view that s 36B was punitive. He found that the purpose of s 36B was not retribution, but rather to recognise a person’s repudiation of allegiance to Australia, and attach a consequence – denationalisation – to this repudiation. He noted that, properly applied, s 36B would only operate to denationalise people who had already rejected the Australian body politic, and who would be unlikely to view the loss of Australian citizenship as a punishment.
The aliens power
A threshold argument advanced by the plaintiff in Alexander was that s 36B lacked the support of a constitutional head of power. Because the Minister ultimately decided to rely only on the naturalization and aliens power in s 51(xix) for support, discussion was confined to whether this power supported s 36B.
In light of the consistency between the majority judgments on the separation of judicial power question, extensive discussion of the aliens power question was not strictly necessary. Indeed, Gordon J directly observed that it was ‘neither necessary nor appropriate to determine whether s 36B [was] supported by the "naturalization and aliens" power’.
Nonetheless, all members of the Court except for Gageler J (who signalled his agreement with the remarks made in the joint judgment) engaged extensively with the question of whether s 51(xix) was capable of supporting denationalisation laws. This is a surprising departure from the approach the High Court has usually taken. Historically, the Court has answered questions about the scope of the aliens power on a case by case basis and only to the minimum extent necessary, with the consequence that the scope of the power, or even its core content, have remained notoriously unresolved.
In Pochi v Macphee [1982] (Pochi), a man who did not hold Australian citizenship but who had lived in Australia for 20 years, and was otherwise said to be ‘in every way an Australian’, was found to be an alien within the scope of s 51(xix) of the Constitution. Chief Justice Gibbs noted that the aliens power was broad but not unlimited, holding that ‘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’ (at [9]). This boundary has come to be known as the ‘Pochi limit’.
In the 40 years since, the High Court has repeatedly affirmed that the ‘Pochi limit’ exists, but has only identified one category of people who fall beyond the reach of the aliens power: Aboriginal and Torres Strait Islander Australians who satisfy the tripartite test in Mabo v Queensland, in Love v Commonwealth. By contrast, various categories of people have been found to fall within the potential reach of the aliens power, including people who arrived in Australia as infants (MIMIA v Nystrom [2006]), people born in Australia to non-Australian parents (Singh v Commonwealth [2004]), British subjects who had been full members of the Australian community prior to the Australia Acts (Shaw v MIMA [2003]) and Papua New Guinea residents who had held Australian citizenship prior to Papua New Guinea’s independence (MIMIA v Ame [2005]).
Alexander raised a new question: in the absence of any change in territorial boundaries, could Parliament draw on s 51(xix) to deprive specific individuals of citizenship and then treat them as aliens?
Interestingly, as with the separation of judicial power question, there was a baseline on which the Court reached unanimous agreement. All seven judges accepted that, to some extent, s 51(xix) is indeed capable of supporting citizenship deprivation legislation, in contexts where a person has renounced their allegiance by acting in a manner that is inconsistent with continued membership of the Australian body politic. This is technically obiter, but the strongest kind. It provides a clear signal as to the starting point from which the Court is likely to proceed if, for instance, there is a future challenge to s 36D of the Australian Citizenship Act, which continues to allow for citizenship-stripping on conviction-based grounds.
The Alexander judgments reveal that the current members of the Court may differ in the threshold at which they find that s 51(xix) supports denationalisation on this basis. As previously outlined, s 36B – the impugned provision – allowed for citizen-stripping on the basis of conduct even where a conviction had not been secured. Section 36D, which remains in place, allows for citizenship-stripping where a conviction has been secured for offences with largely the same conduct elements.
Kiefel CJ, Keane and Gleeson JJ, Gageler J and Steward J were of the opinion that the naturalization and aliens power supported s 36B in its entirety. Gordon J and Edelman J likewise acknowledged that some of the more serious conduct captured by s 36B (such as committing a terrorist act) qualifies as conduct so extreme that s 51(xix) would support a law providing for denationalisation on its basis. However, both commented that some of the less extreme prescribed conduct in s 36B (such as entering an area declared to be a no-go zone by the government) might fall short of the standard required for denationalisation on these grounds to be supported by s 51(xix).
This suggests that if there is a future constitutional challenge to s 36D (which allows for citizenship revocation where a person commits any of the conduct captured by s 36B and is also convicted), Kiefel CJ and Keane, Gleeson, Gageler and Steward JJ would find that s 51(xix) supports s 36D in its entirety, whereas for Gordon and Edelman JJ, parts of s 36D might fall outside the scope of what is supported by this head of power.
A similar division between Kiefel CJ, Keane, Gleeson, Gageler and Steward JJ on the one hand, and Gordon and Edelman JJ on the other, emerged on the question of how to approach the last 40 years of aliens power jurisprudence. Kiefel CJ, Keane, Gleeson, Gageler and Steward JJ accepted the Pochi limit, as well as the authority of the long line of aliens power cases that had succeeded Pochi.
By contrast, both Edelman and Gordon JJ made comments in obiter indicating some appetite for revisiting this jurisprudence. Gordon J noted that national and international circumstances can cause the application of the constitutional term ‘alien’ to change, and that it may be the case that Pochi would be differently decided today. Edelman J more stridently suggested that the historical line of aliens power cases had extended the scope of the power ‘far beyond any ordinary understanding, capturing more and more members of the permanent population of the Commonwealth of Australia’, such that at some point it would ‘become necessary to confront the correctness of those decisions rather than tip toeing around them’.
Edelman and Gordon JJ’s judgments throw open the question of how the scope of s 51(xix) ought to be approached, if not in the manner adopted in the cases over the last 40 years. Edelman J suggested that this might be navigated by aligning the scope of the aliens limb of s 51(xix) with the scope of the immigration limb of the immigration and emigration power in s 51(xxvii). This power has historically been held to authorise laws regulating people who enter Australia as immigrants. However, after a period of ‘probation’, immigrants pass beyond the reach of the power through absorption into the Australian community. On this approach, long-term permanent residents, currently regarded as aliens, would typically fall beyond the reach of s 51(xix). This would raise questions about the constitutionality of sweeping character-based visa-cancellation powers contained in the Migration Act 1958, which currently extend to such residents. It would call into question the correctness of the recent decisions in Falzon [2018] and Chetcuti [2021], in which very long-term residents were held to be aliens.
Justice Edelman’s approach leaves many questions open. For example, Edelman J stressed that the constitutional meaning of alienage cannot be defined by statutory choices. How to define what a constitutional alien is, other than by reference to who has been included or excluded from the Australian community via legislation, has been a million-dollar question hanging over s 51(xix) for decades. Edelman J’s suggestion to unify the jurisprudence on the aliens power and the immigration power does not resolve it. This would decouple the constitutional concept of alienage from the statutory concept of citizenship by rendering any person absorbed into the Australian community to be a non-alien. However, in doing so it tethers the meaning of alienage to the question of absorption into the Australian community, which has also been held to be informed by statutory choices. The ultimate question of how to enable Parliament to exercise the power granted to it over naturalization and aliens without permitting the stream to rise above its source remains unanswered.
As currently a comfortable majority of the High Court does not seem minded to overhaul the jurisprudence on the aliens power, such questions do not need immediate answers. But Edelman and Gordon JJ’s comments may foreshadow future change with wide ranging implications for migration law, especially since Edelman J is guaranteed a place on the High Court until 2044.
What’s next
The decision in Alexander means that s 36B is now invalid. This means that anybody who was stripped of their citizenship under that provision is an Australian citizen again. While recent statements suggest that 22 people in total have been stripped of their citizenship, only two people lost their citizenship under s 36B since its enactment in 2020.
However, the provision that formed the basis for most citizenship stripping prior to 2020 (ss 33AA) provided for denationalisation on a nearly identical basis to s 36B: the only significant difference was that it was even broader in scope because it provided for automatic citizenship loss. This suggests that if a new challenge is brought by a person who lost their citizenship under s 33AA, it would have strong prospects of success. Whether this will eventuate in light of the logistical hurdles that make litigation in this area so difficult remains to be seen.
In the meantime, the Commonwealth retains plenty of mechanisms via which citizens who are believed to pose a risk can be excluded from the Australian community. In addition to the ongoing power to strip a person convicted of certain national security offences of citizenship under s 36D, the Commonwealth may avail of over 90 other anti-terror laws, including laws that allow a citizen outside Australia to be ‘locked out’, via temporary exclusion orders and passport cancellation powers. It is possible that some of these mechanisms may face their own constitutional challenges in the future, although once again, similar logistical hurdles to those faced by Alexander may apply.
The extensive discussion of the aliens power throughout the Alexander judgments may have foreshadowed positions likely to be taken by the members of the High Court in Minister for Immigration, Citizenship, Migrant Services and Multicultural Affairs v Montgomery – a case that was heard in April, and expected to be decided later this year. In Montgomery, the Commonwealth was, amongst other things, seeking the overturn of Love v Commonwealth; in which a majority of the High Court found that Aboriginal and Torres Strait Islander people who satisfied the tripartite test in Mabo are not constitutional aliens, notwithstanding their citizenship status. But very recently, following a change in government, the Commonwealth opted to withdraw its challenge in Montgomery, rendering a decision unnecessary. How the jurisprudential battlelines over the scope of s 51(xix) will ultimately play out remains to be seen.
Dr Sangeetha Pillai is a constitutional lawyer and a Senior Research Associate at the Andrew and Renata Kaldor Centre for International Refugee Law.
Suggested citation: Sangeetha Pillai, ‘Judicial agreements and disagreements in Alexander v Minister for Home Affairs’ on AUSPUBLAW (21 September 2022) <https://www.auspublaw.org/blog/2022/09/judicial-agreements-and-disagreements-in-alexander-v-minister-for-home-affairs>