The Lawfulness of Citizenship Deprivation: Comparing Australia and the UK

Abstract

The rise in international terrorism has led to a rise in citizenship deprivation. Depriving a person of citizenship represents a harsh national security measure. Although both Australia and the UK have citizenship deprivation legislation, the judicial response has differed. In Australia, two laws providing for deprivation of citizenship have been found unconstitutional. In the UK, significant challenges to citizenship deprivation decisions have failed, including those relating to Shamima Begum, deprived of UK citizenship in 2019, whose request for permission to appeal in respect of the decision was rejected by the UK Supreme Court in August 2024. In this context, it is striking that despite the lesser degree of human rights protection under the Australian Constitution and federal statutes compared with the UK, the Australian courts may have arrived at a significantly rights protective approach to citizenship deprivation, leading to an important procedural safeguard by requiring courts to make decisions on citizenship deprivation. This underlines interesting features of the Australian system, in which the development of doctrines under a written constitution that limits legislative power, such as through the separation of powers, can sometimes lead to significant (if uneven) rights protective outcomes. Short of a shift in UK constitutional law doctrine around the separation of powers (which is unlikely), the Australian decisions cannot be mirrored in the UK. However, they may point towards the possibility of stronger procedural safeguards in the context of citizenship deprivation, as well as some potential human rights law implications.

1. Introduction

The rise in international terrorism has led to a rise in citizenship deprivation (Anderson 2013; Gibney 2013; Pillai and Williams 2017a, 2017b). Depriving a person of citizenship represents a harsh national security measure. In the words of Gordon J of the High Court of Australia, it effects a “permanent rupture in the relationship between the individual and the State, involving loss of fundamental rights including by exposure to detention and deportation from the territory” (High Court of Australia 2023b, para. 76). Although both Australia and the UK have citizenship deprivation legislation, the judicial response has differed. In Australia, two laws providing for deprivation of citizenship have been found unconstitutional in Alexander v Minister for Home Affairs in 2022 and Benbrika v Minister for Home Affairs in 2023 (High Court of Australia 2022a, 2023a). However, another law providing for loss of citizenship due to criminal conviction relating to conduct that took place prior to the grant of citizenship was upheld in Jones v Commonwealth (High Court of Australia 2023b). Although the cases also involve analysis of the heads of legislative power granted to the Australian (Commonwealth) Parliament under the Australian Constitution, the result of constitutional invalidity was achieved on the basis of the separation of powers that has been inferred from the structure of the Constitution (specifically, Chapter III). The High Court reasoned that the impugned citizenship deprivation provisions empowered the executive to impose punishments that could only be imposed judicially, applying and extending what is known as the Lim principle, after the 1992 case of Chu Kheng Lim v Minister for Immigration.

In the UK, significant challenges to citizenship deprivation decisions have failed, including those relating to Shamima Begum, deprived of UK citizenship in 2019 (UK Supreme Court 2021), whose request for permission to appeal in respect of the decision was rejected by the UK Supreme Court in August 2024 (UK Supreme Court 2024). This reflects well known differences between Australian and UK public law. The plenary power of the UK Parliament, as distinct from the limited legislative powers of the Commonwealth Parliament, means that the sometimes heated debates in Australia over the scope of heads of legislative power do not carry over to the UK context. Furthermore, the strong separation of judicial power in the Australian system and the various implications drawn from this separation are not mirrored in the UK system. Nonetheless, the UK does have human rights protection in the form of the Human Rights Act 1998, whereas it is well known that Australia has limited protection for human rights at the federal level, including under the Australian Constitution, though there is some state and territory level human rights legislation, and the possibility of a federal human rights law is currently under consideration (Parliament of Australia 2024).

In this context, it is striking that despite the lesser degree of human rights protection under the Australian Constitution and federal statutes compared with the UK, the Australian courts may have arrived at a significantly rights protective approach to citizenship deprivation, leading to an important procedural safeguard by requiring courts to make decisions on citizenship deprivation. This underlines interesting features of the Australian system, in which the development of doctrines under a written constitution that limits legislative power, such as through the separation of powers, can sometimes lead to significant (if uneven) rights protective outcomes. However, although the doctrinal importance of the separation of powers in the UK is less than in Australia, it may be that there is some relevance of this point to statutory interpretation. If citizenship deprivation is punitive, as the High Court found, then there is at least a possible argument that the Home Secretary needs to exercise the power to deprive a person of citizenship in a way analogous to a court, which could offer modest additional procedural protections under UK law. Furthermore, the analogy of citizenship deprivation with criminal punishment may have possible relevance to human rights law in the UK, especially in respect of Article 4 of the European Convention on Human Rights.

The paper has three sections after this introduction. In Section 2, I describe the power of citizenship deprivation in Australia, outlining the legislation at the time of the High Court’s consideration of the issue, and then the relevant decisions of that Court. In Section 3, I analyse some aspects of the Australian decisions—focusing on the naturalisation and aliens heads of Commonwealth legislative power, and the relevance of separation of powers considerations—to argue that in the context of Australian law, the decisions are persuasive, and impose welcome procedural protections. In Section 4, I turn to a comparison with the UK. Short of a shift in UK constitutional law doctrine around the separation of powers (which is unlikely), the Australian decisions cannot be mirrored in the UK. However, they may point towards the possibility of stronger procedural safeguards in the context of citizenship deprivation, as well as some potential human rights law implications. Section 5 concludes the discussion.

2. Citizenship Deprivation in Australia

In Alexander and Benbrika, the High Court found that two of the provisions of the Australian Citizenship Act 2007 (Cth), which enabled the deprivation of citizenship of suspected terrorists, were unconstitutional. Applying what is called the Lim principle, which limits the circumstances under which the executive can impose punishments on the basis that punishment is generally a judicial function, the Court held that deprivation of citizenship as provided for under these provisions constituted a punishment that could not be imposed by the executive. However, in Jones, it was found that deprivation of citizenship in different circumstances—conviction for an offence committed prior to the grant of citizenship—was constitutionally permissible. In this section, I set out the citizenship deprivation laws and the Australian cases, to lay the groundwork for analysis in Section 3.

2.1. Legislation

Australian citizenship did not exist at the time of creation of the state in Federation in 1901; the most legally relevant status was British subjecthood (High Court of Australia 1906a, 1906b; 1936, p. 650). The concept of Australian citizenship was introduced by the Australian Citizenship Act 1948 (Cth) (originally titled the Nationality and Citizenship Act 1948 (Cth)); more recently, it has been provided for by the Australian Citizenship Act 2007 (Cth). One basis for loss of citizenship—revocation of citizenship in circumstances of making a knowingly false or misleading representation or statement or concealing or causing to be concealed a material circumstance for a purpose of or in relation to the Citizenship Act—was originally provided for by amendments made by the Nationality and Citizenship Act 1958 (Cth). In 1984, an additional ground of deprivation was added addressing criminal convictions relating to conduct before the grant of citizenship. Section 34(2) of the 2007 Act provides that the Minister may, by writing, revoke a person’s Australian citizenship if the person is an Australian citizen by conferral, and (inter alia) “the person has been convicted of an offence against Section 50 of this Act [relating to false statements or representations], or section 137.1 or 137.2 of the Criminal Code [relating to false or misleading information or documents], in relation to the person’s application to become an Australian citizen” or “the person has, at any time after making the application to become an Australian citizen, been convicted of a serious offence” as defined in the Act, and “the Minister is satisfied that it would be contrary to the public interest for the person to remain an Australian citizen”. However, Section 34(3) creates an exception to revocation for conviction of a serious offence if the revocation of citizenship would result in statelessness.

A different set of provisions addressed the issue of deprivation of citizenship due to suspected terrorism. The Australian Citizenship Amendment (Allegiance to Australia) Act 2015 (Cth) inserted into the Citizenship Act “self-executing” procedures for the automatic cessation of citizenship of dual citizens for engaging in terrorism related conduct or fighting in or being in the service of a declared terrorist organisation, in addition to creating a ministerial discretion to revoke citizenship as a result of conviction of a prescribed offence. Sangeetha Pillai and George Williams observe that “[t]he automatic citizenship stripping provisions … [went] further than even the UK legislation, by imposing citizenship deprivation upon all dual citizens that meet the designated criteria, irrespective of the level of threat they pose” (Pillai and Williams 2017b, p. 549). These provisions were also subject to some criticism by the Independent National Security Legislation Monitor in 2019, who pointed out that under them, “citizenship may cease even in a case of relatively low-level offending, with considerable mitigating factors, where the threat to the Australian community may be negligible and where it may be contrary to both Australia’s international relations and the public interest” (Independent National Security Legislation Monitor 2019, pp. 48–50).

The Australian Citizenship Amendment (Citizenship Cessation) Act 2020 (Cth) amended the Australian Citizenship Act and introduced a new scheme for citizenship deprivation. As amended by this Act, Section 36B of the Australian Citizenship Act created a ministerial discretion to revoke citizenship on the basis of conduct. Prior to the repeal of these provisions in 2023 (following the decisions in Alexander and Benbrika), Section 36B provided that the Minister could determine in writing that a person aged 14 or older ceased to be an Australian citizen if the Minister was satisfied that the person engaged in specified conduct while outside Australia, or engaged in specified conduct while in Australia, has left Australia and has not been tried for an offence in relation to the conduct, and the conduct “demonstrates that the person has repudiated their allegiance to Australia”, and “it would be contrary to the public interest for the person to remain an Australian citizen”.

Section 36D provided an alternative basis for citizenship deprivation by the Minister upon conviction of a person of certain offences, though the other requirements were similar to Section 36B. Section 36D relevantly provided that the Minister could determine in writing that a person ceased to be an Australian citizen if the person had been convicted of an offence or offences against one or more specified provisions; the person had been sentenced to a period of imprisonment of at least three years or periods of imprisonment that totalled at least three years in respect of the conviction or convictions; the Minister was satisfied that the conduct to which the conviction or convictions related “demonstrates that the person has repudiated their allegiance to Australia”; and the Minister was satisfied that it would be “contrary to the public interest for the person to remain an Australian citizen”.

Sections 36B(2) and 36D(2) made clear that a minister must not make a determination if the Minister is satisfied that the person would, if the Minister were to make the determination, become stateless (that is, a person who is not a national or citizen of any country). However, pursuant to Sections 36B(4) and 36D(4), the power applied regardless of how the person became an Australian citizen. In assessing the public interest for the purpose of Section 36B or Section 36D, Section 36E(2) provided for the minister to have regard to various matters, including the degree of threat posed by the person to the Australian community, the age of the person, the person’s connection to the other country of which the person is a national or citizen, and Australia’s international relations. Cessation of citizenship meant in practice that a person would be subject to potential immigration detention and removal from Australia. As the plurality judgment explained in Alexander (para. 26),

[t]he practical effect of the exercise by the Minister of the power conferred by s 36B(1) is that the person affected has no right to return to Australia and be at liberty in this country, because the Migration Act 1958 (Cth) would require that (save in the unlikely event that the person were granted a visa) the person be taken into detention for the purposes of deportation.

2.2. High Court Decisions

In Alexander, the High Court held by a 6–1 majority that Section 36B was unconstitutional and accordingly invalid. The plaintiff in Alexander, Delil Alexander, born in Australia in 1986, acquired Australian citizenship at birth, as well as Turkish citizenship by descent because his parents were Turkish citizens. In 2013, Mr. Alexander went to Syria via Turkey. According to the Australian Security Intelligence Organisation, it was “likely” that Mr. Alexander had joined Islamic State by August 2013, and he had “likely engaged” in foreign incursions and recruitment in al-Raqqa Province in Syria on or after 5 December 2014. In November 2017, Mr. Alexander was apprehended by Kurdish militia, and in March 2018, he was transferred to the custody of Syrian authorities. In January 2019, Mr. Alexander was convicted and sentenced by a Syrian court to imprisonment for 15 years, later reduced to five. He was later pardoned but remained in detention in Syria. His citizenship was purportedly cancelled by the Minister of Home Affairs pursuant to Section 36B of the Citizenship Act in July 2021.

The first basis of challenge, relating to Section 51(xix) (granting the power to legislate with respect to naturalisation and aliens), failed due to what the Court considered to be the repudiation of allegiance to Australia inherent in terrorism. The plurality, Kiefel CJ, Keane and Gleeson JJ, wrote that “it is to understand the status of citizenship in an incoherently one-sided way to say that Section 51(xix) supports a law that specifies the criteria by which a citizen may voluntarily renounce Australian citizenship—as Mr. Alexander accepted—but does not support a law that treats voluntary conduct demonstrating a repudiation of allegiance to Australia as an implied renunciation of citizenship” (para. 50). The second basis of challenge, relating to the separation of powers under Chapter III, succeeded. The plurality noted that “detention by way of deprivation of liberty in retribution for reprehensible conduct by an individual is a familiar example of punishment by the State”, but added that “[h]istorically, of course, English law sanctioned criminal conduct by punishments other than detention: corporal and capital punishment come readily to mind”, and moreover, “exile has long been regarded as punishment” (para. 72).

The plurality considered that although Section 36B and other provisions “may be said to pursue a purpose of protecting the Australian community from the risks to peace and security posed by returning foreign fighters”, the “principal purpose” was “retribution for conduct deemed to be so reprehensible as to be ‘incompatible with the shared values of the Australian community’”, quoting Section 36A (para. 75). Since whether a law provided for the adjudication and punishment of criminal conduct was a “matter of substance, not form”, it did not matter that the law used the “emollient language of ‘citizenship cessation’” (para. 79). Moreover, it was significant that a “lower factual threshold” was required under Section 36B than Section 36D, the latter of which required conviction of offences after a trial; Section 36B involved a process of ministerial fact finding in which the state was not required to carry the burden of proof; and the Minister was not even required to observe procedural fairness (para. 80–81, para. 86–87).

The other members of the majority adopted similar reasoning. Gageler J said that he agreed with the conclusions and the “substance” of the plurality judgment (para. 98). Gordon J agreed that Section 36B was invalid on separation of powers grounds, since citizenship cessation, in the context of Section 36B, was “a measure ‘taken in the name of society to exact just retribution on those who have offended against the laws of society’ by engaging in past conduct that is ‘identified and articulated wrongdoing’”, a position that she considered was also consistent with the understanding of banishment throughout history (para. 163, para. 167). Along with the other members of the majority, Edelman J concluded that Section 36B “has a punitive character with a purpose to sanction particular conduct” (para. 247). Steward J dissented on the basis that the power to cancel the plaintiff’s citizenship was not judicial power because “it has never been an essentially judicial function to make orders which denationalise a person”, “the task of the Minister here is not to determine whether the conduct … constitutes the commission of any crime”, and “the purpose of s 36B is not to punish” (para. 332, para. 336–337).

Benbrika was described in the plurality judgment of Kiefel CJ, Gageler, Gleeson and Jagot JJ as a “sequel” to Alexander (para. 2). The question was the validity of Section 36D (which provided for the cessation of citizenship as a result of conviction for certain offences), rather than Section 36B. The Court held by a 6–1 majority that Section 36D was invalid as well. Abdul Benbrika, an Algerian citizen who took Australian citizenship in 1998, was convicted of terrorism offences in 2008. In 2020, the Minister for Home Affairs determined under Section 36D(1) that Mr. Benbrika would cease to be an Australian citizen. The plurality explained that in Alexander, the Lim principle was found to be “applicable to a Commonwealth law which purports to empower the Commonwealth Executive to strip a person of Australian citizenship” because of the “nature and severity of the consequences” of a purported exercise of Section 36B(1) and the “purpose of the power as identified in s 36A” (para. 20–21). The plurality said that these considerations relied upon to characterise Section 36B(1) as punitive applied “with equal force” to Section 36D(1) (para. 26).

The Commonwealth argued that “although the reasoning of the majority in Alexander supports the characterisation of s 36D(1) as authorising the Minister to punish criminal guilt, s 36D(1) materially differs from s 36B(1) in that s 36D(1) does not authorise the Minister to adjudge criminal guilt. Instead, s 36D(1) relies for the adjudgment of criminal guilt on prior conviction by a court” (para. 28). The plurality found that this argument was “founded on an incomplete reading of Lim and an incomplete appreciation of the principle there expounded” (para. 32). That was because the Lim dictum stated that “Ch III of the Constitution precludes the enactment, in purported pursuance of any of the sub-sections of s 51 of the Constitution, of any law purporting to vest any part of that function in the Commonwealth Executive” (para. 33, emphasis added). Furthermore, the concern of the Constitution was with substance, not form (para. 34). The plurality considered that “[s]ection 36D(1) purports to vest … a power to impose additional or further punishment in the Minister” (para. 41).

An alternative argument advanced by the Commonwealth was that citizenship deprivation “should not be seen to be a form of punishment which Ch III of the Constitution commits exclusively to the judicial power” (para. 29). The plurality rejected this argument because of “the absence of any precedent for a court being empowered to order this kind of punishment is not to the point”, “the paucity and inconstancy of the precedents for the legislative empowerment of executive deprivation of citizenship or nationality consequent upon a finding of criminal guilt by a court on which the argument relied”, and the fact that “legislative precedents for a person’s citizenship or nationality being automatically terminated by operation of law upon the person engaging in conduct constituting an offence” could be “relevantly distinguished from judicial power” (para. 45–47). Any functional considerations for involving the executive could be “accommodated … by the simple and common legislative expedient of requiring executive application or certification as a precondition to a court making an order for cessation of citizenship as a component of the punishment” (para. 48).

Gordon J “generally agree[d]” with the plurality’s reasons, but wrote separately to emphasise the rationale for the Lim principle as lying in “first, the historical judicial protection of liberty against incursions by the legislature or the executive; and second, the protection of the independence and impartiality of the judiciary so as to ensure the judiciary can operate effectively as a check on legislative and executive power” (para. 51). Edelman J also agreed with the plurality’s answers to the special case questions (para. 76, para. 115). However, he considered that “the Executive can be empowered by a non-punitive law to determine that a person ceases to be an Australian citizen and also loses their ability to remain in or return to Australia, in circumstances in which that person acts in such an extreme way as effectively to repudiate their allegiance to Australia” (para. 77). Steward J again dissented. He accepted that citizenship deprivation had “harsh” consequences (para. 142–144). However, he rejected the claim that this constituted punishment (para. 158).

In Jones, handed down the same day as Benbrika, a challenge was brought to Section 34(2)(b)(ii) of the Australian Citizenship Act. As noted above, this section provides a power for a minister to revoke a person’s Australian citizenship where the person had received it by conferral, and the person is convicted and sentenced to imprisonment for more than 12 months in respect of an offence that was committed before the person became an Australian citizen. An additional requirement is for the Minister to be satisfied that it would be contrary to the public interest for the person to remain an Australian citizen. Unlike the other two laws relating to citizenship deprivation, this provision was upheld by the High Court because it was not considered to be punitive. Rather, the Court considered that the law merely supported the integrity of the naturalisation process.

The challenge was brought by Phyllip Jones, who was born in the UK and remains a British citizen. He migrated to Australia in 1966 and was granted Australian citizenship in 1988. In 2003, Mr. Jones was convicted of indecent dealing and indecent assault, committed between 1980 and 2001, with two of the counts corresponding to conduct before he became an Australian citizen. For each count, he was sentenced to a term of imprisonment of 2.5 years, to be served concurrently. The Minister for Home Affairs, Immigration and Border Protection revoked his citizenship under Section 34(2) in 2018. The question was whether the provision was invalid because it was not supported by the head of legislative power under Section 51(xix) of the Constitution or because it was in violation of the separation of powers established by Chapter III of the Constitution. The provision was upheld by a 6–1 majority. In respect of Section 51(xix), the plurality, comprising Kiefel CJ, Gageler, Gleeson and Jagot JJ, wrote that the provision was valid under Section 51(xix) because it “provides for ‘an act or process of denaturalization’ which is supported by the ‘naturalization’ limb of s 51(xix) of the Constitution” (para. 4).

In assessing whether the law was punitive under Lim, the plurality applied a test of reasonable necessity. The question was whether the law was “reasonably capable of being seen as necessary for a legitimate non-punitive purpose” (para. 44). The law here was considered to be “reasonably capable of being seen as necessary to protect the integrity of the naturalisation process” (para. 50). That much was apparent in considering the “good character” requirement of citizenship, the relevance of prior serious offending, the fact that “any later conviction of a prior serious offence was likely to call into question the correctness and completeness of the evidentiary basis on which the ministerial assessment was made”, and the continuity with the provisions of the 1948 Act (para. 51). The purpose of protecting the integrity of the naturalisation process was supported by the public interest requirement (para. 54).

Edelman J was also in the majority and concluded that the power in s 34(2)(b)(ii) was “reasonably capable of being seen as necessary to protect the naturalisation process”, but added that a “lengthy and unexplained delay between conviction and the Minister’s decision” could support an inference of improper purpose (para. 182). Steward J, who had dissented in Alexander and Benbrika, “substantially” agreed with the reasons of the plurality, and with the conclusion of the plurality on the basis that Section 34(2)(b)(ii) was “reasonably capable of being seen as necessary for the legitimate non-punitive purpose of protecting the integrity of the naturalisation process” (para. 188, 197). Gordon J, who had been part of the majority in the other cases, dissented. She accepted that the statutory purpose was to protect the integrity of the naturalisation process (para. 72). However, she considered that there was a default characterisation that revocation of naturalisation was punitive and that default characterisation was reinforced by considering “the period of time during which the power may be exercised” (para. 89). As Gordon J put it, the provision “leaves the class of naturalised citizens to whom it applies—persons who became citizens by conferral and who have since been convicted of a ‘serious offence’ within the meaning of Section 34(5)—at permanent risk of the revocation of their nationality and citizenship” (para. 90).

3. Analysis of the Australian Decisions

In this section, I analyse the High Court’s approach to citizenship deprivation. I consider the separation of judicial power in Australia, and the limited Commonwealth legislative power under the Australian Constitution in respect of naturalisation and aliens. I argue that the analysis of these issues by the Australian High Court in the recent decisions is credible, building on and improving some aspects of the earlier case law. The extension of the Lim principle is persuasive. Furthermore, although there remains lingering confusion around the scope of the head of legislative power in Section 51(xix) of the Australian Constitution, the recent cases represent a positive shift, particularly in relation to the understanding of the foundation of citizenship legislation.

3.1. Separation of Judicial Power

The powers of the Commonwealth Parliament are subject to certain limitations, transgression of which will render its legislation unconstitutional and invalid. Some of these limitations are express, and some of them have been found by the High Court to be implied by the text or structure of the Australian Constitution. The separation of powers between the judiciary and other branches is one of the implied limitations. This separation has long been recognised as an important principle under the Australian Constitution, Chapter III of which sets out provisions relating to the judiciary. At the Commonwealth level, what is often called the Boilermakers principle “go[es] a long way towards protecting the decisional independence of federal courts by strictly separating the judicial and non-judicial powers of government” (Ananian-Welsh and Williams 2014, p. 621). This principle stipulates that

Chap[ter] III does not allow the exercise of a jurisdiction which of its very nature belongs to the judicial power of the Commonwealth by a body established for purposes foreign to the judicial power, notwithstanding that it is organized as a court and in a manner which might otherwise satisfy ss 71 and 72, and that Chap[ter] III does not allow a combination with judicial power of functions which are not ancillary or incidental to its exercise but are foreign to it ((High Court of Australia 1956, p. 296), affirmed (Privy Council 1957); see also (High Court of Australia 1918); but compare (High Court of Australia 1995)).

Two further principles of particular significance apply in respect of judicial power in Australia. First, although the same strict separation of powers does not apply at the state level within Australia’s federal system, the High Court has developed a body of law addressing the position of state courts (High Court of Australia 1996, 2006b, 2010a). What is called the Kable principle provides that if a state law purports to confer a function on a state court that is “repugnant to or incompatible with” its role as a repository of Commonwealth jurisdiction (High Court of Australia 1996, p. 103), by substantially impairing the institutional integrity of the state court (High Court of Australia 2014b, para. 40), the law is unconstitutional (the principle also applies to territory courts (High Court of Australia 2004a, 2014b)). Kable v Director of Public Prosecutions (NSW) itself concerned an ad hominem state law directing a state court to order that a particular person be detained, which the High Court found invalid. The Kable principle has led to the Court striking down further state laws, which involved the enlistment of the courts to implement decisions of the executive or other disregard for ordinary judicial processes (High Court of Australia 2009, 2010b, 2011). Among commentators, the principle is controversial (Winterton 2002; Goldsworthy 2011; Goldsworthy 2014) but also has defenders (Emerton 2019).

Second, the Lim principle sets out certain limits upon government action at the Commonwealth level that is taken otherwise than by a court. Chu Kheng Lim v Minister for Immigration, Local Government and Ethnic Affairs concerned the constitutionality of certain immigration detention arrangements under the Migration Act 1958 (Cth). It involved a challenge by Cambodian nationals who had arrived in Australia by boat without a valid entry permit and been detained in custody since their arrival under legislation providing for compulsory detention of certain non-citizens. This legislation was found to be supported by Section 51(xix). The question remained whether it violated an express or implied constitutional limitation. The laws providing for detention were held to be valid, but Brennan, Deane and Dawson JJ (forming a majority with Mason CJ agreeing at p. 10) made an important observation of constitutional principle deriving from the separation of powers (p. 27):

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. That function appertains exclusively to and “could not be excluded from” the judicial power of the Commonwealth. That being so, Ch III of the Constitution precludes the enactment, in purported pursuance of any of the sub-sections of s 51 of the Constitution, of any law purporting to vest any part of that function in the Commonwealth Executive. … [P]utting to one side the exceptional cases to which reference is made below, 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.

The plurality went on to identify “qualifications” to “the general proposition 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” (at least in peacetime). These were as follows:

• the “arrest and detention in custody, pursuant to executive warrant, of a person accused of crime to ensure that he or she is available to be dealt with by the courts”, since this was not punitive or appertaining exclusively to judicial power, and was subject to the supervisory jurisdiction of the courts;
• the “[i]nvoluntary detention in cases of mental illness or infectious disease”, since this was also non-punitive and did not necessarily involve the exercise of judicial power; and
• “the traditional powers of the Parliament to punish for contempt and of military tribunals to punish for breach of military discipline” (p. 28).

Moreover, the detention of aliens (as distinct from citizens) by the executive “for the purpose of expulsion or deportation” or “in the context and for the purposes of executive powers to receive, investigate and determinate an application by that alien for an entry permit and (after determination) to admit or deport” was found in Lim to be permissible under Chapter III (p. 32). This was so provided that the detention is “limited to what is reasonably capable of being seen as necessary for the purposes of deportation or necessary to enable an application for an entry permit to be made and considered” (p. 33).

Despite facing questioning by some Justices in Kruger v Commonwealth and Al-Kateb v Godwin (High Court of Australia 1997, pp. 109–111; 2004b, para. 258), the Lim principle appears to be in good standing today, especially in the wake of its confirmation in Alexander and Benbrika. In Plaintiff M76, Plaintiff S4, Plaintiff M96A and AJL20, Justices of the High Court considered aspects of the principle in relation to immigration detention (High Court of Australia 2013, para. 140–141; 2014a, para. 26; 2017, para. 21; 2021c, para. 22). In addition, in an earlier Benbrika decision and Garlett (High Court of Australia 2021a, 2022b), the Court found that preventive detention regimes—respectively, a Commonwealth scheme relating to preventive detention of terrorist offenders and a state scheme relating to preventive detention of “serious offenders”—were not in violation of the Lim principle and Kable principle, respectively, because they were protective, rather than punitive. However, in Alexander and the more recent Benbrika, the principle was applied to strike down the Commonwealth’s citizenship deprivation laws. Subsequently, in NZYQ v Minister for Immigration, Citizenship and Multicultural Affairs (High Court of Australia 2023c), indefinite immigration detention was also found to violate the Lim principle—overruling Al-Kateb—as were, in YBFZ v Minister for Immigration, Citizenship and Multicultural Affairs (High Court of Australia 2024), certain monitoring and curfew conditions for those released from immigration detention following the decision in NZYQ.

The development of the Lim principle in Alexander (confirmed in Benbrika) is persuasive. Although this case law extends the Lim principle beyond its original focus on detention, such incremental development by analogy is consistent with longstanding methods of constitutional interpretation in Australia. Furthermore, it seems plausible to consider that citizenship deprivation is a form of punishment given both its severe effects on the individual, and the fact that it is imposed as a consequence of past conduct, supported by a possible analogy with the historical practice of banishment (Macklin 2014; Gibney 2020). Indeed, the plurality considered the deprivation of citizenship as meriting protection under the Lim principle to perhaps an even greater extent than detention (para. 73):

Today, the fundamental value accorded to the liberty of the individual provides the rationale for the strict insistence in the authorities that the liberty of the individual may be forfeited for misconduct by that person only in accordance with the safeguards against injustice that accompany the exercise of the judicial power of the Commonwealth. The case for the strict insistence on these safeguards is, if anything, stronger where the penalty for misconduct involves not only a loss of liberty within the community, but the loss of all entitlement to be both within the community and at liberty.

However, the use of the concept of punitive, rather than protective, purpose to distinguish between forms of executive action that are permissible and those that are not remains, at times, a slippery feature of the Lim case law, including in Alexander and Benbrika. That is because it is apparent that some executive action can be both punitive and protective in nature (indeed, even criminal punishment can be intended to protect the community). In Alexander, Gageler J observed that the “distinction” between a “punitive purpose” and a “protective purpose” remained “analytically useful” in this area “as long as the notion of what amounts to a protective purpose is kept within bounds which make the distinction meaningful” (para. 107). Be that as it may, it shows a point of vulnerability in the decisions in Alexander and Benbrika from reliance on the Lim principle: if Australian citizenship deprivation laws were to be reframed in a less apparently punitive way, it is at least possible that they might escape the principle.

The other possible legislative response to the cases of Alexander and Benbrika would be to provide for citizenship deprivation to be carried out through judicial processes, with the attendant safeguards inherent to those processes. And indeed, after the decision in Benbrika, the Australian Citizenship Amendment (Citizenship Repudiation) Act 2023 was passed by the Commonwealth Parliament, requiring that citizenship cessation orders be made by courts on the application of the executive, rather than by the executive itself. Given the severe consequences of citizenship deprivation for the individual, it can be seen as a positive step to include the safeguards of judicial confirmation of the executive’s assessment, and in this way the result of the Alexander and Benbrika cases has arguably been to strengthen the integrity of citizenship deprivation processes in Australia.

In contrast, the application of the Lim principle to the issue of immigration detention through the decisions of NZYQ and YBFZ—despite addressing the original detention focus of Lim—seems less convincing. The decision in NZYQ is in tension with past statements of the Lim principle, which made clear that detention “reasonably capable of being seen as necessary for the purpose […] of deportation” was permitted (p. 33). Even if prolonged detention results from attempting to arrange deportation because of obstacles, it is difficult to see why such detention ceases to be “necessary” for that purpose. Gleeson CJ (in the minority) considered in Al-Kateb that this detention power is “not punitive in nature, and does not involve an invalid attempt to confer on the Executive a power to punish people who, being in Australia, are subject to, and entitled to the protection of, the law” (para. 4). Rather than Al-Kateb, a more appropriate past decision to reconsider might have been Kruger, since the removal of Indigenous children during the Stolen Generations that was found constitutional in that case presents a potential Lim problem, as well as possible problems under other constitutional principles (Baldwin 2023).

3.2. Scope of Section 51(xix)

In the federal system of government in Australia, the Commonwealth (federal) Parliament has only limited powers to legislate as provided for by the Australian Constitution. Section 51(xix), one of these powers, provides that “[t]he [Commonwealth] Parliament shall, subject to this Constitution, have power to make laws for the peace, order, and good government of the Commonwealth with respect to … naturalization and aliens”. Section 51(xix) has been an important provision for sustaining Australia’s migration laws, especially since the 1980s. Prior to this, the Commonwealth tended to rely on the immigration power in Section 51(xxvii), which provides for a power to make laws with respect to “immigration and emigration”. However, the High Court’s interpretation of Section 51(xxvii) was such that those who were “absorbed into the community” were no longer within the reach of the power under Section 51(xxvii) (High Court of Australia 2022a, para. 205; 1923, 1949, 1975). Subsequently, the Commonwealth began placing emphasis on Section 51(xix) instead.

Decisions of the High Court have since found that the power to legislate with respect to “naturalization and aliens” in Section 51(xix) is to be read differently from that to legislate with respect to “immigration and emigration” in Section 51(xxvii). In Pochi v Macphee in 1982, the High Court held that an alien who is not naturalised does not cease to be an alien by absorption into the community (High Court of Australia 1982, pp. 111, 113). The Court thus held that an Italian migrant who had lived in Australia for decades without taking citizenship, but who committed a crime and had his visa cancelled as a result, could be deported from the country under laws supported by Section 51(xix) of the Constitution. In later judgments, the Court found that British subjects who lived in Australia for decades without taking citizenship were within the reach of Section 51(xix), as were children born in Australia to foreign citizens who were not Australian citizens or permanent residents, and citizens of a territory that gained independence (High Court of Australia 1988, 2003, 2004c, 2005, 2006a; but see High Court of Australia 2001).

An emphasis on the holding of citizenship might be thought to raise a concern that the Commonwealth Parliament could decide who was and who was not an alien within the reach of Section 51(xix) on the basis of its own citizenship laws, a position that would allow the Parliament (rather than the courts) to define the ambit of its own power over “aliens” under Section 51(xix), in tension with the supervisory role of the judiciary under the Australian Constitution. However, in important and much-quoted dicta in Pochi, Gibbs CJ (with whom two other Justices agreed, forming a majority in that case: pp. 112, 116) placed some limits, albeit cautious ones, on the scope of the power to legislate with respect to “aliens” (pp. 109–10):

Clearly the 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. This question was not fully explored in the present case, and it is unnecessary to deal with it. However, the Parliament can in my opinion treat as an alien any person who was born outside Australia, whose parents were not Australians, and who has not been naturalized as an Australian.

Over the years, members of the Court confirmed the limit identified in Pochi that Parliament could not legislate in respect of people who “could not possibly answer the description of ‘aliens’ in the ordinary understanding of the word” (the “Pochi limit”) (High Court of Australia 2002, para. 31; 2003, para. 9; 2004c, para. 4, 124, 151–153). These confirmations served to illustrate that although citizenship laws passed by Parliament in practice determined who was and who was not an alien (High Court of Australia 1988, pp. 183–84; 1992, p. 25; 2003, para. 2; 2004c, para. 4, 122, 205), the power of Parliament to decide alienage was not entirely at large and remained under the Court’s supervision. In Love v Commonwealth, a majority of the Court applied this Pochi limit to find that non-citizen Aboriginal Australians, some of whom the government was seeking to deport from Australia, were not aliens because of their connection to the country as Indigenous inhabitants (High Court of Australia 2020). Love attracted a measure of controversy (Australian Bar Association 2020), but a bid by the centre-right Liberal Party–National Party coalition government to overturn it was dropped once the centre-left Labor Party took office (Karp 2022).

In Chetcuti v Commonwealth, the Court found that a British subject who had arrived in Australia before Australian citizenship legislation was enacted was within the reach of Section 51(xix), having failed to take Australian citizenship over several decades thereafter (High Court of Australia 2021b). The plurality judgment of Kiefel CJ, Keane, Gageler and Gleeson JJ contained unusual dicta that created new uncertainty: the plurality stated that Parliament could define who was an alien and failed to qualify this position by reference to the Pochi limit (para. 12, whereas the other Justices referred to this limit: para. 37, 66, 104). This approach was in conflict with longstanding principles of Australian constitutional law, since without the Pochi limit, the Court would seemingly be abdicating its power of review over the interpretation of Section 51(xix), as Edelman J later pointed out in Alexander (High Court of Australia 2022a, para. 196–198). Furthermore, the plurality judgment suggested that it was the “aliens” power that served as the basis for enacting citizenship legislation, not, as had been suggested previously, Section 51(xix) generally (including both aliens and naturalisation limbs) (High Court of Australia 2006a, para. 48). Among other potentially odd consequences, this seemed to imply that even persons gaining Australian citizenship by naturalisation did not do so under legislation enacted pursuant to the naturalisation power (Baldwin 2021).

Although still seeking to give Parliament scope to define alienage, in its analysis the plurality judgment in Alexander seemed to course correct from Chetcuti (perhaps surprisingly, since the plurality comprised some of the same Justices). The plurality now acknowledged the Pochi limit, concluding that it was not exceeded in respect of someone who had repudiated allegiance (para. 35, 46). Moreover, the plurality seemed to accept that it was Section 51(xix) of the Constitution as a whole (not merely the “aliens” part of that provision) that gave Parliament the power to “create and define the concept of Australian citizenship” (para. 33). This trend continued in Jones, in which all Justices at last turned to focus on the naturalisation limb of Section 51(xix) to support legislation granting Australian citizenship by naturalisation (including legislation to support the integrity of the naturalisation process). This was a welcome step, away from an anomalous reliance on the aliens limb in the context of citizenship legislation.

However, arguably, it does not go far enough, since the naturalisation limb of Section 51(xix) is relevant not just for citizenship by naturalisation, but also supports all citizenship legislation. As I have previously suggested, “‘naturalization’ refers to a process with an endpoint, which may imply the power to enact legislation for that endpoint, currently being citizenship” (Baldwin 2021). Accordingly, the entire citizenship legislation system has a sufficient connection with naturalisation. In Jones, Edelman J wrote that: “Naturalisation itself is the formal recognition of a person’s status as a member of the Australian political community or the conferral of such a status upon a person who was not already such a member. The naturalisation power is the power to recognise, or to create, the status of non-alien” (para. 132). On this reasoning, the naturalisation power can properly be understood as the basis for the entirety of citizenship legislation in Australia.

Notwithstanding the Court’s welcome shift on the naturalisation limb, some uncertainty remains around the aliens limb and its potential support for citizenship deprivation. Although the naturalisation limb might be a plausible basis for the citizenship deprivation power considered in Jones, since there were flaws in the naturalisation process for a particular person, there is complexity around relying on either the naturalisation limb or the aliens limb for the citizenship deprivation laws considered in Alexander and Benbrika. As Edelman J pointed out in Jones, the naturalisation power is a power with respect to naturalisation, not denationalisation (para. 143). He rejected an argument advanced by the Commonwealth that the aliens power in Section 51(xix) authorised legislation to denationalise non-aliens (para. 120). Edelman J stated that the aliens power “cannot be a sufficient source of power to denationalise those people of the Commonwealth who were never aliens or who had ceased to be aliens” (para. 131). After all, “the aliens power is not a power over non-aliens” (para. 132). Generally speaking, it cannot be addressed to those outside the class of aliens, because there would not be a sufficient connection between a law and the subject matter of aliens. Someone who holds citizenship would presumably not be an alien, and therefore not usually subject to the power.

One possible solution to this problem, which has been advocated for by Commonwealth Solicitor-General Stephen Donaghue, is to enable the Commonwealth to treat dual citizens as aliens. However, this approach would be unattractive in turning Section 51(xix) into an essentially unlimited power over a significant class of Australian citizens—comprising approximately half the population—which could potentially subvert constitutionalism in Australia. In this vein, Edelman J in Alexander expressed concerns about the “imperial march” of the aliens power, noting that the Commonwealth had argued that dual citizens were within its reach (para. 183). He also addressed the unusual plurality dicta in Chetcuti, which he cautioned should not be read as conferring an “unconstrained power on the Parliament” under Section 51(xix) (para. 197). For his part, he suggested a return to the “absorption” test that is applied under the immigration power in Section 51(xxvii) (para. 204–211). Edelman J reprised this theme in Jones, objecting that the Commonwealth’s interpretation of the aliens power would treat much of the population (such as those who held dual citizenship) as “second-class” Australians susceptible to the aliens power (para. 103).

A more attractive solution is to accept that those who have repudiated their allegiance to Australia by conduct become within the reach of the aliens power. For example, in Alexander, Edelman J seemed to accept that Section 36B was supported by the aliens power because “the validity of s 36B hinges upon the extreme wrongdoing that is required by, and inherent in, the notion of repudiation of allegiance to Australia to be determined by the Minister” (para. 234). Gordon J found that Section 36B was within the scope of the aliens power “to the extent that s 36B(1)(a) covers the ‘paradigm case of implicit renunciation’ of allegiance by ‘spying or fighting’ for an enemy state declared to be at war with Australia, and the closely related category of conduct involving service in the armed forces of a declared terrorist organization” (para. 154). It was unnecessary to decide whether “to the extent that s 36B captures conduct beyond serving in the armed forces of a country at war with Australia and conduct of foreign fighters”, it might exceed Section 51(xix) (para. 155). The analysis that fighting for an enemy state or terrorist organisation transports a person within the aliens power under Section 51(xix) seems entirely plausible. However, if it were not accepted that repudiatory conduct made a citizen susceptible to the aliens limb of Section 51(xix), an alternative possibility (suggested by Edelman J in Jones) would be to identify an implied constitutional power of denationalisation in limited circumstances that is “incidental to the existence of the Commonwealth as a state” (para. 139). Thus, it does not seem wrong to conclude that the Australian Constitution can sustain citizenship deprivation laws. Instead, the issue is one of what safeguards these laws require to be compatible with the Australian Constitution, a question that was addressed by the High Court through its analysis of the Lim principle.

4. Comparison with the UK

How does the Australian judicial approach compare with that of the UK, the legal system from which Australia’s own legal system, at least in part, descended? In many respects, the two systems have taken fundamentally different approaches to this issue. The limited Commonwealth legislative powers under the Australian Constitution have no parallel in the UK, given the plenary power putatively enjoyed by the UK Parliament under the doctrine of parliamentary sovereignty—subject, at least, to the theoretical possibility of hard limits on the power of the UK Parliament as canvassed in obiter comments in cases such as R (Privacy International) v Investigatory Powers Tribunal (UK Supreme Court 2020). Since, in the UK, the constitutional authorisation of the statute is not in question, citizenship deprivation decisions are usually challenged either on the basis of the Human Rights Act 1998 (HRA), which implements the European Convention on Human Rights domestically, or on the basis of the statutory authorisation of the deprivation decision under Section 40 of the British Nationality Act 1981 (BNA) (House of Commons Library 2024).

Section 40(2) states a broad power that “[t]he Secretary of State may by order deprive a person of a citizenship status if the Secretary of State is satisfied that deprivation is conducive to the public good”. However, Section 40(4) indicates that “[t]he Secretary of State may not make an order under subsection (2) if he is satisfied that the order would make a person stateless”. There is also a power under Section 40(3) to deprive a person of citizenship resulting from registration or naturalisation if it was obtained by fraud, false representation or concealment of a material fact, and a power in Section 40(4A) to deprive a naturalised citizen of citizenship even if it would make the person stateless if “the deprivation is conducive to the public good because the person, while having that citizenship status, has conducted him or herself in a manner which is seriously prejudicial to the vital interests of the United Kingdom, any of the Islands, or any British overseas territory” and “the Secretary of State has reasonable grounds for believing that the person is able, under the law of a country or territory outside the United Kingdom, to become a national of such a country or territory”.

Thus, challenges to citizenship deprivation decisions often focus on whether the person deprived of citizenship would be rendered stateless, or on whether the deprivation would violate Convention rights, or on procedural issues. For example, the issue of statelessness was key to challenges in Al-Jedda v Secretary of State for the Home Department and Pham v Secretary of State for the Home Department (UK Supreme Court 2014, 2015). In R (Begum) v Secretary of State for the Home Department, Shamima Begum had her citizenship cancelled in 2019 under Section 40(2) of the BNA on the basis that she was a “a British/Bangladeshi dual national who it is assessed has previously travelled to Syria and aligned with ISIL”, and her “return to the UK would present a risk to … national security” (UK Supreme Court 2021, para. 1). The UK Supreme Court rejected the argument that Ms. Begum, in a camp in Syria, had to be granted leave to enter to return to the UK to participate in her challenge to the decision (para. 107–111). The Court also held that the appeal to the Special Immigration Appeals Commission (SIAC) was on public law grounds; SIAC was not able to remake the decision (para. 66–72).

Given that the “exercise of the power conferred by Section 40(2) must depend heavily upon a consideration of relevant aspects of the public interest, which may include considerations of national security and public safety, as in the present case”, the Court reasoned that some aspects of the decision might not be justiciable and others might depend “on an evaluative judgment of matters, such as the level and nature of the risk posed by the appellant, the effectiveness of the means available to address it, and the acceptability or otherwise of the consequent danger, which are incapable of objectively verifiable assessment”. Thus, the Court stated that “SIAC has to bear in mind, in relation to matters of this kind, that the Secretary of State’s assessment should be accorded appropriate respect, for reasons both of institutional capacity (notwithstanding the experience of members of SIAC) and democratic accountability” (para. 70). The caution of the UK Supreme Court is notable (Martin 2021), especially compared with the sweeping decisions of the High Court of Australia, which simply cast aside large parts of the system of citizenship deprivation in Australia and have led to the introduction of judicial processes governing deprivation.

In the UK, it is also possible to look to human rights law to challenge citizenship deprivation decisions. However, it has proved to be of only limited assistance. Under the European Convention on Human Rights, deprivation of citizenship could potentially raise issues at least under Article 3 (torture or inhuman or degrading treatment or punishment), Article 4 (slavery, servitude, and forced or compulsory labour), Article 6 (fair trial), Article 8 (private and family life, home and correspondence), Article 14 (discrimination), and Protocol 4 Article 3 (expulsion of nationals), though the UK has not ratified Protocol 4 (and so the rights it states do not appear in Schedule 1 to the HRA). Much of the case law focuses on Article 8 (European Court of Human Rights 2023b). The position of the European Court of Human Rights is that citizenship deprivation under Article 8 may be permissible so long as it is not arbitrary. This requires an assessment of issues including whether the deprivation was in accordance with the law, whether it was accompanied by procedural safeguards, whether authorities acted diligently and swiftly, and whether deprivation would render the applicant stateless.

For example, a violation of Article 8 was found in Usmanov v Russia, where citizenship was annulled due to the applicant’s omission of information about siblings, which was grossly disproportionate (European Court of Human Rights 2021a). A violation was also found in Huseynov v Azerbaijan, where the consequence of the citizenship deprivation was statelessness, in breach of Article 7 of the United Nations Convention on the Reduction of Statelessness (European Court of Human Rights 2023a). Outside such extreme circumstances, it may be difficult to show that citizenship deprivation violates the European Convention on Human Rights. The European Court has correctly recognised that the issue is not a one-way street, since “terrorist violence, in itself, constitutes a grave threat to human rights” (European Court of Human Rights 2022, para. 50). The Court has also said that deprivation of citizenship is not a punishment “in criminal proceedings” that violates the Protocol 7 Article 4 right not to be tried or punished twice in criminal proceedings (European Court of Human Rights 2020, para. 73). Furthermore, in the UK, the issue of extraterritoriality is raised as a barrier to human rights challenges relating to citizenship deprivation in respect of a person outside the UK (Home Office 2014, para. 13).

In this context, it is interesting that in Australia there is a lack of human rights protection at the federal level, and yet a significantly rights protective outcome has arguably been achieved in respect of citizenship deprivation. Nor is such human rights protection through “other” constitutional doctrines, not derived from express human rights protections, entirely surprising for Australia. As shown by the Communist Party Case, in which the banning of the Communist Party in Australia was found not to be supported by a head of Commonwealth legislative power (High Court of Australia 1951), some human rights issues have been addressed in practice through the lens of legislative power, and Edelman J’s concerns about the reach of the aliens power under Section 51(xix) and its “imperial march” carry on that tradition. Moreover, as outlined above, additional human rights issues have been addressed through constitutional implications including those relating to judicial power; it was ultimately the separation of judicial power that proved decisive in the Australian decisions on citizenship deprivation.

Recently, greater attention has been paid to Article 4 of the European Convention on Human Rights as a possible basis of challenge to citizenship deprivation decisions. Article 4 prohibits slavery, servitude, and forced or compulsory labour, and has been found by the European Court of Human Rights to extend to human trafficking as well. However, an argument that the deprivation decision in relation to Ms. Begum was a violation of Article 4 because she was a victim of human trafficking was rejected by Court of Appeal, and permission to appeal from the Court of Appeal’s decision was rejected by the UK Supreme Court (UK Supreme Court 2024). Ms. Begum argued that the failure when taking the deprivation decision in 2019 to “consider whether the Appellant (then in Syria) was a potential victim of trafficking, whether there had been any failures by public authorities to protect her from trafficking, the obligations owed to her under Article 4 at that time as a victim or potential victim of trafficking, and whether deprivation would impede the State’s ability to discharge those obligations” constituted a violation of Article 4 (para. 5).

The argument that Article 4 is violated by depriving Ms. Begum of citizenship in circumstances of possible human trafficking relied on the decision of the European Court of Human Rights in VCL and AN v United Kingdom, in which the Court stated that “as soon as the authorities are aware, or ought to be aware, of circumstances giving rise to a credible suspicion that an individual suspected of having committed a criminal offence may have been trafficked or exploited, he or she should be assessed promptly by individuals trained and qualified to deal with victims of trafficking” (European Court of Human Rights 2021b, para. 160). Furthermore, the Court stated that “the prosecution of victims, or potential victims, of trafficking may, in certain circumstances, be at odds with the State’s duty to take operational measures to protect them where they are aware, or ought to be aware, of circumstances giving rise to a credible suspicion that an individual has been trafficked” (para. 159). In Ms. Begum’s case, SIAC had made a finding of fact that “there is a credible suspicion that Ms. Begum was recruited, transferred, and then harboured for the purpose of sexual exploitation”, while a child at the time (Special Immigration Appeals Commission 2023, para. 219).

However, this argument of an Article 4 violation was rejected by the Court of Appeal, in part because the Court said that there was “no authority either of a domestic court or of the Strasbourg court which has held that the non-punishment principle extends beyond criminal prosecutions (which is the context in which it was discussed in VCL, see para. 160–161) to a decision to deprive an individual of her citizenship, or to a refusal to repatriate her” (Court of Appeal 2024, para. 82). Several UN special rapporteurs have criticised the Court of Appeal’s decision on the basis that Ms. Begum is a possible victim of trafficking (United Nations Office of the High Commissioner for Human Rights 2024). Nonetheless, the UK Supreme Court panel considering permission to appeal (comprising Lords Reed, Hodge and Lloyd-Jones) also rejected the Article 4 argument, stating that “[t]his argument is based on submissions as to the effect of Article 4 which appear to the panel to go well beyond the existing case law of the European Court of Human Rights, or the incremental development of the principles established by that case law” (para. 6). Interestingly, neither the Court of Appeal nor the UK Supreme Court cited or considered the Australian case law that concluded that citizenship deprivation was indeed a form of punishment. If it is accepted that citizenship deprivation is a form of punishment that is analogous to criminal prosecution, then there may be more plausibility to Ms. Begum’s argument that the dicta in VCL apply to these circumstances. Even though this would “extend” these dicta “beyond criminal prosecutions”, as the Court of Appeal put it, this could be viewed as an incremental development. However, at this stage, such a step has not been taken.

Another area of potential significance for the Alexander and Benbrika decisions in the UK is the separation of powers. In the UK, there is no strict separation of powers, nor any strict rule mirroring the Lim principle that punishment can only be a judicial task outside certain exceptions, though the Chapter III case law has been cited when interpreting Article 6 of the European Convention on Human Rights (UK House of Lords 2003, para. 50). Even though the structural independence of the judiciary was increased by the Constitutional Reform Act 2005—by creating the UK Supreme Court and reforming the office of the Lord Chancellor—it would be a bold move for the courts to assert under the UK constitution the kind of separation that exists in Boilermakers and Lim. For one thing, this would involve claiming a right to invalidate statutes that infringed that separation of powers, which, as discussed above, is usually not considered to be a power that the UK Supreme Court holds (though views are sometimes expressed to the contrary).

However, there might be some relevance in statutory interpretation to the issue of the separation of powers and the conception of citizenship deprivation as punishment. In R v Secretary of State for the Home Department, Ex parte Venables (UK House of Lords 1998), it was held that a task exercised by the executive that is “penal” is “judicial” in nature, and this imposes certain restrictions on how it is to be exercised. As such, the Home Secretary was not permitted to take into account petitions from the public when sentencing convicted murderers. (Subsequently, this power was found to be incompatible with Article 6 of the European Convention on Human Rights (UK House of Lords 2003).) In Venables, Lord Steyn explained the position as follows (p. 526):

Parliament must be assumed to have entrusted the power to the Home Secretary on the supposition that, like a sentencing judge, the Home Secretary would not act contrary to fundamental principles governing the administrative of justice. Plainly a sentencing judge must ignore a newspaper campaign designed to encourage him to increase a particular sentence. It would be an abdication of the rule of law for a judge to take into account such matters. The same reasoning must apply to the Home Secretary when he is exercising a sentencing function.

Lord Goff made a similar point (pp. 490–91):

[I]f the Secretary of State implements a policy of fixing a penal element of the sentence of a mandatory life prisoner pursuant to his discretionary power under Section 35, he is to this extent exercising a function which is closely analogous to a sentencing function with the effect that, when so doing, he is under a duty to act within the same constraints as a judge will act when exercising the same function. In particular, should he take into account public clamour directed towards the decision in the particular case which he has under consideration, he will be having regard to an irrelevant consideration which will render the exercise of his discretion unlawful.

Accordingly, if the reasoning in Alexander and Benbrika were considered applicable to the UK context, it could at least be argued that, in making deprivation decisions, the Home Secretary is “under a duty to act within the same constraints as a judge will act when exercising the same function” because of the penal nature of citizenship deprivation. This might offer some procedural protections—for example, it could impose a limitation on the kinds of considerations the Home Secretary could take into account in making decisions about citizenship deprivation, at least ruling out petitions or other public input of that nature into deprivation decisions—and provide a countervailing consideration to the national security considerations that otherwise lead to broad judicial deference, as outlined by the UK Supreme Court in Begum. However, again, this step has not been taken in the UK at the present stage. And, in any event, acknowledging that citizenship deprivation is “penal” and, therefore, potentially “judicial” in the Venables sense may not amount to a significant restraint on the power of the UK executive to deprive people of citizenship. It would not necessarily imply the need to follow the example of the Australian Citizenship Amendment (Citizenship Repudiation) Act 2023 (Cth), and empower courts, rather than the executive, to make decisions on citizenship deprivation—a course that is perhaps well suited to ensuring procedural protections in this area.

5. Conclusions

The High Court of Australia has taken some significant decisions in relation to citizenship deprivation laws, finding key provisions to be unconstitutional. In the UK, such bold strokes have not been attempted, and it is unlikely to be possible for UK courts and tribunals to take especially forceful steps in the current legal system. The limited heads of legislative power in the Australian system, and especially under the powers relating to aliens and naturalisation in Section 51(xix) of the Australian Constitution, although not the operative issue in the citizenship deprivation cases, place constraints on federal power that are relevant in this area; no such limitation exists in respect of the plenary power putatively enjoyed by the UK Parliament. Instead, courts must look to statute, especially the conditions under the BNA and the protections of the HRA.

What proved especially important in the Australian High Court’s decisions was the doctrine of separation of powers, inferred from the structure of the Australian Constitution, and especially the Lim principle. It was found that citizenship deprivation was punitive and therefore its imposition by the executive was unconstitutional. In the UK, there would need to be a much stronger separation of judicial power than currently exists for the courts to apply something resembling the Lim principle. However, if courts in the UK were minded to acknowledge, in line with the Australian cases, that there was something punitive about citizenship deprivation, there might at least be some implications for the procedural safeguards that are discerned through statutory interpretation of the relevant provisions of UK citizenship deprivation legislation. In addition, there could be human rights law implications, including in relation to Article 4 of the European Convention on Human Rights. However, to date, there is no sign of such a shift.

More broadly, the case law on citizenship deprivation in Australia and the UK also indicates the potential impact of a written constitution that limits legislative power. The significance of such a constitution in the context of citizenship deprivation—even of a constitution that does not protect human rights very much—has proved great. In Australia, rights protection is sometimes effected indirectly at the constitutional level through issues of legislative authorisation, as Edelman J discussed in terms of the aliens power and was evident in the Communist Party Case, and doctrines such as the separation of powers, in its Kable and Lim guises. The recent decisions illustrate the unusual, but significant, human rights tradition of the Australian High Court, in which human rights issues can sometimes be addressed in effect by reference to concepts like the separation of powers and legislative power, rather than through express human rights protections. The protection offered in practice by such doctrines is uneven, but no trifle, and, as the example of citizenship deprivation illustrates, at least on occasion may exceed in its practical impact statutory human rights protection along the lines of the HRA.