University of New South Wales Law Journal
VOTING WITH AN ‘UNSOUND MIND’? A COMPARATIVE STUDY OF THE VOTING RIGHTS OF PERSONS WITH MENTAL DISABILITIES
TREVOR RYAN[*], ANDREW HENDERSON[**] AND WENDY BONYTHON[***]
While Australia has a system of universal franchise, in the sense of a system of voting that is broadly inclusionary, there are some notable exceptions – minors and some convicted criminals are excluded, for example. While the political participation of these excluded groups sometimes attracts attention in the media, there is another group that has been even more marginalised, both within society and within debates over the franchise. This group is persons with mental disability or, more precisely, with actual or assumed impaired decision-making capacity resulting from chronic or acute mental illness, dementia, intellectual disability or brain injury. This appears to be changing as individual nations (including Australia) assess their compliance with the United Nations (‘UN’) Convention on the Rights of Persons with Disabilities (‘CRPD’) on a range of issues and as a greater number of citizens experience dementia in an ageing population. This article seeks to contribute to this reform momentum by comparing the laws relating to this issue across jurisdictions, particularly Japan and Australia, to argue for a political franchise without discrimination against persons with mental disabilities.
The article proceeds as follows. Part II provides a broad overview of domestic and international trends in adult guardianship and the political rights of persons with mental disabilities. Part III describes the Japanese legal framework for guardianship, capacity, and voting, and analyses a recent constitutional judgment that resulted in Japan repealing laws that disenfranchised persons with mental disabilities. Part IV describes the Australian position, including the history of provisions that disqualify from voting persons of (in the language of the statute) ‘unsound mind’, to demonstrate that this position is rooted in exclusionary social policies of the late 19th century and that difficulties in describing the nature of the exclusion have dogged the provision since its inception. Part V draws some comparative lessons to argue that Australia should follow Japan’s lead. Australia should adopt an inclusionary approach to voting that gives vulnerable persons the support and protection necessary to vote rather than an exclusionary approach that alienates persons with mental disabilities from the political community and achieving full citizenship. This Part also
argues that a comparative methodology has value in analysing this issue,
despite differing institutional frameworks. For example, while constitutional restrictions on the Australian Parliament are arguably weaker with regard to regulating the franchise, the fundamental principles are the same and are normative as well as doctrinal. This point is reinforced by the observation in
Part V that that the question of how persons with mental disabilities are integrated into the franchise has deeper significance for the role that electoral law has for constituting a more inclusive citizenry. A comparative approach also provides a model for how international obligations can or should be implemented through domestic law. For these reasons, Part VI concludes by broadly concurring with the recommendations of a recent Australian Law Reform Commission (‘ALRC’) inquiry. In particular, the Australian Parliament should repeal the disqualification from voting and enrolment in section 93(8)(a) of the Commonwealth Electoral Act 1918 (Cth) (‘Electoral Act’) of persons of ‘unsound mind’ and institute a system of exemptions from compulsory voting for persons established under the National Decision-Making Principles to be incapable of expressing a voting preference even with comprehensive support measures.
Persons with mental or intellectual disabilities who have difficulty making decisions across a range of areas may find themselves the subject of an
order appointing a substitute decision-maker. This is a relationship in which an appointed decision-maker or ‘guardian’ is given authority to make decisions and exercise powers over which an adult protected person has been deemed to lack capacity by a court or tribunal. Defined broadly, it encompasses substituted decision-making for economic decisions, for example buying and selling property, and personal decisions, such as where to reside and non-emergency medical decisions.
The concept of substitute decision-making is a very old one, which is at least in part attributable to the hierarchical feudal system of social organisation found in England during medieval times. Based on a system of land tenure, the King was the ultimate owner of all property. Others exercised varying degrees of control over land as a consequence of royal grants and, in return, occupiers of land performed various obligations for the King, such as paying taxes, and providing military support. Those who were granted land rights by the King also had obligations to those lower than them in the hierarchy. Considering that property rights and associated obligations were the foundation of the feudal society, it is unsurprising that historically so few people were recognised as having legal capacity. Similarly, it appears almost inevitable that the concept of the substitute decision-maker would have evolved in the way it did under these circumstances. Indeed, the parens patriae (father of the nation) power of the King was such that, since at least the 1300s, he was able to assume control of land granted to a ‘lunatic’ or ‘idiot’ for the duration of the incapacity, and manage the land on a quasi-trustee basis, providing the means required to support the landowner and his tenants, but potentially retaining the remainder of the proceeds derived from the property for Crown use. In many instances, it was a Crown representative, rather than the King himself, who exercised this power; subsequently it was transferred to the Court of Chancery.
The modern equivalents of these powers are found in the protective jurisdictions, such as guardianship boards and tribunals, and mental health boards and tribunals. As with the Crown in medieval times, these bodies also delegate responsibilities, by way of making orders appointing substitute decision-makers to act on behalf of people who are found to lack functional capacity. Generally, these substitute decision-makers will be authorised to make decisions regarding healthcare, personal or welfare matters, or financial matters. In the event of a private substitute decision-maker being appointed, such as one or more family members or friends, these decision-making powers can be vested separately or comprehensively, and can be exercised in a variety of ways, depending on the circumstances. In the event of no suitable appointee being identified, the court or tribunal can also appoint a public body, such as the Public Advocate or Public Guardian, or the Public Trustee. The appointed body makes decisions regarding health and welfare or financial matters on behalf of a person with impaired decision-making capacity.
There are differences among jurisdictions. Modern regimes feature tailored guardianship orders based on decision-specific capacities. They may place decisions about capacity and the appropriate extent of a guardian’s powers in
the hands of a tribunal with layperson input. A recent trend has been the introduction of the concept of ‘supported’ decision-making in which the emphasis is on assisting the person subject to the order to come to their own decisions by removing physical or social barriers to participation and exhausting untried means of communication.
Adult guardianship law has increasingly become the concern of international and comparative law. A key development was the entry into force of the CRPD in 2008. This was the first multilateral treaty to impose binding obligations upon member states specifically relating to disabilities. The CRPD states as its purpose to ‘promote, protect and ensure the full and equal enjoyment of all human rights and fundamental freedoms by all persons with disabilities, and to promote respect for their inherent dignity’.
Of particular relevance to people with intellectual or mental disabilities with actual or presumed impaired capacity is article 12. This article (‘Equal recognition before the law’) affirms the right of persons with disabilities to legal capacity ‘on an equal basis with others in all aspects of life’ and requires states to ensure that people with disabilities are afforded the same recognition of legal capacity as those without disabilities. Much of the discourse about the impact of implementing the CRPD has focused on article 12 and its implications for involuntary treatment of people with mental illness, and substituted decision-making arrangements.
Article 12 refers extensively to ‘legal capacity’, but does not clearly define the concept. In a subsequent General Comment, the Committee on the Rights of Persons with Disabilities clarified the distinction between ‘legal capacity’ and ‘mental capacity’. Legal capacity requires that the person be recognised as a legal person who holds legal rights (legal standing), and that he or she has the legal agency to act on those rights, and have their actions recognised at law. Mental capacity was identified by the Committee as frequently conflated with legal capacity, relying on controversial and socially contextualised criteria to deprive people of legal capacity on the basis of ‘status’ (diagnosed impairment), ‘negative consequences’ (making decisions with bad outcomes), or ‘functional approach’ (based on an assessment of the person’s ability to understand and process relevant information). The functional approach was criticised by the Committee as flawed in two respects: first, because it is ‘discriminatorily’ applied to people with disabilities; and secondly, because ‘it presumes to be able to accurately assess the inner-workings of the human mind’.
The General Comment also makes it clear that states party to the CRPD
are required to replace all forms of substitute decision-making, including ‘plenary guardianship, judicial interdiction and partial guardianship’, with supported decision-making frameworks. Significantly, the General Comment does not provide a prescribed model of acceptable ‘supported’ decision-making, reflecting the fact that there is no single agreed upon definition of the concept. Yet the following description provides a useful starting point:
In supported decision-making, the individual is always the primary decision maker, but it is acknowledged that autonomy can be communicated in a number of ways, thus provision of support in different forms and intervals can assist in the expression of autonomous decisions. Supported decision-making enables the individual to retain legal capacity regardless of the level of support needed.
There are therefore many ways in which a decision may be supported, including formal instruments such as advance directives and enduring powers
of attorney, and informal support networks of family and friends. At times, these informal arrangements may be formalised, which in turn can reflect broader processes of contractualisation and juridification in social spheres. An example of this is the introduction in Japan and Canada of enduring powers relationships mediated by contract.
It is therefore up to states to develop their own frameworks, itself a
complex process requiring adaptation to local context. A number of bodies and commentators have taken up the challenge of conceptualising Australia’s adoption of a supported decision-making framework. For example, the Victorian Law Reform Commission has employed supported decision-making as an underlying principle for extensive proposed reforms to guardianship and related areas, including the introduction of ‘supporters’. Carney partially resists the tendency toward juridification, suggesting that these ‘[c]ircles of support, micro-boards, or friendship networks, may better be cultivated purely within civil society’, albeit subject to the regulatory framework of ‘informal community, self-help, non-government and government human services agencies’. In part, this is due to the ‘chilling effect’ that formal mechanisms may have in the construction of these support networks and the role that education may have as a form of fostering ‘clarity and accountability’ in the provision of support. Chesterman makes the related point that one ramification of a fully supported framework may mean that many orders for guardianship are redundant due to the impact of support on the ability to make decisions. Like Carney, Chesterman proposes making the transition needed in formal mechanisms such as guardianship through small discrete steps, such as narrowing the scope of decisions that can be made by a guardian. This might create ‘more space for the recognition of supported decision-making as a viable alternative to substituted decisions’.
But the ambit of the CRPD does not solely encompass financial or personal decisions and extends to the active participation of the individual in all aspects of the community. As a result, a redesign of the Australian guardianship framework to comply with a supported decision-making framework will also have to reconsider laws governing the franchise for people with mental disabilities and how acceptable models of supported decision-making might apply in the context of voting. Article 29 of the CRPD lists comprehensive rights of political participation, including the right to vote and to be adequately supported in exercising that right. Persons with disabilities protected by the Convention include those with ‘long-term physical, mental, intellectual or sensory impairments’. The General Comment specifically states that the right to vote is not to be limited as a consequence of need for support in decision-making. Those protections encompass all aspects of political and public life, from participation in elections as a voter, through to nomination and election as an office holder or candidate. Specifically, article 29 of the CRPD identifies that persons with disabilities are entitled to participate ‘directly or through freely chosen representatives’, and that ‘voting procedures, facilities and materials [should be] appropriate, accessible and easy to understand and use’. Secret ballots are also protected.
While the CRPD and parallel developments in European human rights law have been key recent catalysts, the movement to liberalise guardianship laws and promote the civil and political rights of people with disabilities dates back at
least to the 1960s. For example, in 1967, the United States (‘US’) State of Minnesota legislated to clarify that an order of guardianship was not
necessarily tantamount to an assessment of total incapacity and that therefore certain rights could coexist with an order of guardianship, for example to engage in transactions with legal effect, drive a car or vote. France abandoned its traditional regime of incompetency and allowed a person under guardianship to exercise certain rights in 1968. Québec did the same in 1991, Austria in 1983, and Germany in 1994, each attempting to balance protection of the individual with new concepts of self-determination and normalisation.
The right to political participation for people with disabilities has also undergone substantial reform since the 1980s. The United Kingdom abolished any common law rule that had disenfranchised persons on the basis of their mental state. The Federal Court of Canada struck down legislation excluding persons with a ‘mental disease’ from voting in 1988. Canada went on to repeal provisions in their electoral laws that disenfranchised persons on the basis of adult guardianship in 1993. Sweden and the Netherlands have removed all constitutional exclusions of voters with intellectual disabilities. Austria has removed any similar exclusion in its Code. France retains the exclusion but its application must be considered in respect of each potentially excluded voter by a judge. In 2001, the US District Court for the District of Maine declared invalid a provision of the Maine Constitution that disenfranchised those under guardianship because it infringed due process guarantees under the US Constitution.
These developments show that, while many states retain a capacity requirement to vote, there is a trend toward severing the link, particularly where a guardianship order is taken as a marker of a lack of capacity. Unlike section 93(8)(a) of the Australian Electoral Act that excludes voters based on a ‘clinical’ diagnosis of being of ‘unsound mind’, reforms overseas have focused on the adoption of functional tests of capacity or, the approach of the Committee on the Rights of Persons with Disabilities, which has advocated moving beyond such tests altogether, as described above.
It might be objected that abandoning any test is a step too far. First, it could be argued that allowing persons who cannot pass a functional test of voting capacity to vote brings into disrepute or trivialises the entire system of voting, democratic participation, and the notion of individual personality in a liberal democratic state. In other words, the right to vote is premised on obtaining a sound understanding of competing candidates and policies and the significance of voting, or at least the potential to achieve this level of understanding. On this view, voting is a citizen’s duty as well as a right and it must be performed adequately.
One weakness in this argument is that there is considerable vagueness inherent in the concept of ‘capacity to vote’. While vagueness is inherent in
many legal concepts, where fundamental rights such as voting are concerned, greater vigilance should be exercised against vague standards. Long-established yardsticks for capacity for private transactions are not necessarily helpful here. While a key aspect of such tests is whether a person has the ability to ‘understand the nature of the transaction when it is explained’, it is questionable whether the act of voting is a ‘transaction’ at all. If these tests are linked to protectionist interventions in transactions for the benefit of the individual, such a protective function is not clearly apparent when it comes to voting. The context-dependent nature of decision-making ability also makes it difficult to devise a coherent test for capacity. A disability may only become an obstacle where society fails to provide assistance to harmonise that characteristic with surrounding social systems. Accordingly, advances in our understanding of mental and intellectual disabilities have necessitated a fundamental rethinking of the ability of persons with diminished capacity, given the necessary support, to arrive at and express opinions about matters such as voting. In the context of voting, these supports could include programs developed by electoral authorities and enabling technology and also, with adjustments to the law, voting through a representative or adult guardian (as explored below).
A second argument for retaining a link between voting and mental capacity is the ‘slippery slope’ contention that dropping the capacity threshold logically entails granting voting rights to children or intelligent animals. Yet this misses the point that voting rights are fundamentally related to citizenship, which in its ‘normative’ sense is ‘a progressive project which is not just concerned with legal citizens, but with persons and the way persons should act and be treated as members of [the political] community’. Delayed conferral of voting rights upon reaching majority as a symbolic moment of full membership in the political community is one thing. Excluding adults with mental disabilities from joining or remaining in this community by denying them voting rights is another.
A third argument is that persons without capacity to vote are vulnerable to being coerced by fraudulent third parties to vote in a certain way, which allows such parties to have a political impact beyond that of their fellow citizens. The Australian Electoral Commission gave evidence to the Joint Standing Committee on Electoral Matters that between 2008–09 and 2011–12 it had removed 28 603 voters nationally under section 93(8)(a). However, in large-scale elections and individual electorates of tens of thousands, the practical effect is likely to be marginal. A constitutional and regulatory question also arises as to how the problem could be countered using less restrictive means than disenfranchisement, for example voting with a trusted helper or representative, with additional witnessing requirements (as explored below in Part III).
In summary, the arguments for abandoning any test for capacity to vote are stronger than those for retaining such a test (or an equivalent marker such as an adult guardianship order or institutionalisation). Such tests reflect an exclusionary model that denies full citizenship for members of the political community and discriminates on the basis of a mental disability. The following Part demonstrates that a similar conclusion has been reached in Japan, a jurisdiction with a different constitutional framework, yet fundamental similarities such as a stable liberal democracy with representative, responsible government, and with much to offer as a hybrid jurisdiction often curious to learn from other jurisdictions.
Section 11 of Japan’s Public Offices Election Act lists disqualifying criteria for both voting and standing for election and, until amended in May 2013, included ‘persons subject to an adult guardianship order’ (‘seinenhikoukennin’). Japan reformed its adult guardianship regime in 1999 to move beyond its traditional role of property protection toward a regime of broader significance in a society experiencing rapid ageing and welfare marketisation. Accordingly, the conceptual underpinnings of the regime have changed explicitly from paternalism to autonomy, normalisation and usage of remaining capacities. The Japanese regime defines guardianship broadly to cover personal and property decisions and, since the reforms, enduring powers and other like consensus-based arrangements. At the time of the reforms, the Ministry of Justice mooted deleting the provision in Japan’s electoral laws that disqualified adult wards from voting. This was resisted by the ministry responsible for elections (the Ministry of Internal Affairs and Communications) on the grounds that nothing in the reforms mandated such a change and a separate system for monitoring the capacity to vote would be impracticable.
This position became less tenable when Japan signed the CRPD in 2007 and began revising domestic laws for ratification. In December 2009, the Headquarters for the Promotion of Reform of the Disability System was established with the Prime Minister at its head. An expert advisory body under this framework made some recommendations regarding how disqualifications with regard to voting and candidacy rights of people under adult guardianship might be abolished, given their discriminatory effect on persons with disabilities. It also noted knock-on effects for other forms of political participation, such as through deliberative bodies at various levels of government. This momentum was revived when, upon the passage of further ratifying legislation in 2012, both houses of the Diet resolved that the issue of political participation of persons under guardianship orders should be subject to further inquiry. In 2013, the issue was brought to a head by a rare court decision declaring unconstitutional the provisions of the Public Offices Election Act that disenfranchised adult wards. The following section describes this judgment and its consequences.
Takumi Nagoya is a Japanese citizen with Down syndrome who, upon reaching majority (20 years) in 1982, consistently voted in elections until an order of adult guardianship was issued by a Japanese family court in 2007. This had the effect of having Nagoya struck from the electoral roll automatically under section 11 of the Public Offices Election Act. With the assistance of her guardian, Nagoya lodged an application to the Tokyo District Court for a declaration that this provision was unconstitutional. The arguments of the Japanese Government against Nagoya mirrored those considered above.
The first argument of the Government was that a ‘voting right’ (‘senkyoken’) is not merely a right but also a public duty and that guardianship as a criterion of disqualification to vote is a rational means of guaranteeing that this solemn duty is performed adequately. The standard in Japan’s Civil Code required for an order of guardianship in Japan is ‘normally lacking the ability to understand the reason of things through a mental disability’. The Government contended that a ruling that this standard has been met in guardianship proceedings is sufficient to indicate a lack of the requisite ability to form a personal view on the suitability of candidates through information about their political views and policy platform. Put simply, an order of guardianship is a marker of incapacity to vote. This also has the benefit, according to the Government, of practicality, in contrast to specific assessments of capacity to be conducted before each election. Second, the Government argued that if wards were able to vote, there would be a risk not merely of uninformed or informal votes occurring, but also of votes being made under the direction of a third party. For these reasons, the Government argued that disenfranchising adult wards was not a measure that lacks a basis in reason and was therefore within the legitimate bounds of legislative discretion.
The Court rejected this argument. It conceded that voting has aspects of a right and a duty, and that the legislature must have a degree of discretion (albeit subject to judicial review) over the electoral system, which reflects the legitimate interest of the broader citizenry who are given voting rights. The Court also conceded the abstract point that there may be occasions where it is necessary to harness the machinery of one system (ie, guardianship) to achieve the purpose of a ‘parallel’ system (ie, elections), which may in turn have incidental disadvantages for a certain class of persons. Yet the Court underscored the rights aspect of voting and its fundamental importance in a liberal democratic society. It followed a 2005 judgment of the (apex) Japanese Supreme Court, which applied a strict test to restrictions on the right to vote in a case where the vote for single-seat constituencies for the National Diet had been denied to Japanese expatriates. The test for constitutional validity formulated in that case continued a line of cases that borrowed heavily from US constitutional jurisprudence. Like the US Constitution, the Japanese Constitution includes express commitments to popular sovereignty, representative government and universal suffrage, including anti-discrimination rights both generally and specifically in exercising the vote. The Supreme Court held that, given the central importance of the vote as a means of materialising popular sovereignty through representative government, the test to be applied is that constraints upon universal franchise for adults are unconstitutional unless it can be demonstrated that there are ‘unavoidable circumstances’.
The District Court rejected that a guardianship order was such an unavoidable circumstance. First, the Court first looked to the text of the Civil Code, which uses the term ‘normally’ lacking capacity as a requirement for an order of full statutory guardianship. Related provisions in the Civil Code lend support to the view that an order of guardianship is not synonymous with a general declaration of incapacity. For instance, an adult ward does not require the consent of a guardian to marry, divorce, acknowledge a child or create a will; all acts that are premised upon requisite capacity. Therefore the revised Civil Code presumes that, even for a person under full statutory guardianship, incapacity is a decision-specific and fluctuating concept rather than a medically diagnosed condition.
Second, as an empirical matter, the Court accepted evidence that showed that it was possible for a person to form a personal view on various policies of state even without capacity to manage financial affairs. For example, in this case the plaintiff did not have basic numeracy, but could engage in everyday conversation and, given the requisite support, employment. Moreover, as stated above, the plaintiff had in fact voted since reaching majority.
Third, the nature of modern guardianship also militated against conflating guardianship orders with incapacity to vote. Whatever the case in the past, the sole focus of the new guardianship system is on promoting the broader interests of the protected person. This is achieved (at least in theory) through supporting the remaining decision-making capacity of the ward in relation to his or her economic and personal affairs and, where necessary, providing a mechanism for the revocation by the guardian of decisions of a legal nature that are harmful to the individual. This is all the more reason why obstructing the exercise of political rights is alien to modern guardianship. As evidence for this, the Court found that when making an order of guardianship, a family court makes no assessment about the specific capacity of the person to exercise a right to vote. The Court noted that the medical evaluation form provided by the family court as a guideline for doctors to express their views on capacity (to which the courts typically defer) overwhelmingly focuses on capacity to manage financial affairs.
The Court accepted that the broader historical context is relevant to an informed perspective on the evolving conceptual underpinnings of the new adult guardianship system, namely respect for autonomy, normalisation, and making use of remaining faculties. The Court noted that the social status of persons with mental illnesses and intellectual disabilities (and the aged) had evolved both domestically and overseas relative to the 19th century Meiji period in which the ‘incompetent’ (‘kinchisan’) and ‘quasi-incompetent’ (‘junkinchisan’) formulation of adult guardianship was established. The reforms to adult guardianship in 1999 reflected this shift of paradigm through concrete measures such as allowing adult wards to freely purchase everyday items, transactions that could have been revoked by a guardian in the past. Furthermore, a change was made to the Civil Code to the effect that an adult guardian must respect the wishes of an adult ward in both personal and property matters.
Fundamentally, the Court stressed the new ethos of inclusion in the new guardianship regime. It noted that while many Japanese citizens experience a variety of disabilities including low levels of decision-making capacity
through natural or acquired causes including ageing, illness or accident, these ‘by no means rendered them unsuitable as members of a democratic state in
which sovereignty is held by a self-governing populace and should therefore should have a voice in Japan’s governance through voting rights’. Prior to the inception of the new system there were multiple provisions in a range of laws that imposed disqualifications upon persons under guardianship, ranging from licences to practise law to riding a racehorse. Many of these were removed precisely in recognition that there had been a common misconception that
the guardianship system had as its function removing individuals from
the community. The Court held that these changes reflect the fundamentally different concepts underpinning the reformed adult guardianship system. In other words, whereas the former system was more readily characterised as a ‘civil death’ sentence, the reformed adult guardianship system is no longer capable of operating as a yardstick for the capacity required for voting (among other things).
The Court also rejected the Government’s second argument, namely the danger of voter fraud, by applying the strict test for restrictions on voting rights developed in the 2005 Supreme Court case. It held that this test requires evidence that, without a particular restriction, it would be in practice impossible or extremely difficult to guarantee the fairness of an election. Ultimately, the District Court found that the Japanese Government had not adduced sufficient evidence to show this. Without commenting on their merits, the Court noted overseas examples (including Australia) to demonstrate that it was not necessary to use a guardianship order as a marker of the capacity to vote. The Court acknowledged evidence that, at the level of local government in Japan, there were proven cases of coercion of persons with mental disabilities to vote for a specific candidate. The Court also acknowledged that technical mistakes could be made by persons without capacity at the ballot box. Nevertheless, the Court held that this evidence did not indicate that the frequency of either problem was significant enough to endanger the overall fairness of any election or even affect the outcome. Indeed, the Court noted that, given the prevalence of dementia and other mental disabilities in the community, a considerable number of voters in Japan were statistically likely to lack the capacity to vote, but the results of elections could not be impugned on that basis. For the various reasons above, the Court found section 11(1)(a) of the Public Offices Election Act to be unconstitutional and declared that the plaintiff was entitled to vote.
The aftermath of the case was marked by divisions within party, coalition, and bureaucracy. The Minister for Internal Affairs (responsible for elections) gave the following reason for appealing the decision: legislative reform was desirable but would take time. Therefore, allowing the judgment to stand would merely cause disruption in upcoming elections. Presumably, this reflected the Ministry’s earlier opposition to reform due to the burden of implementing a responsive system of capacity assessments for the tens of thousands of persons under adult guardianship. The Ministry of Justice held that the appeals process should play out to bring unity to the law (there were three pending judgments considering the same issue in other district courts).
Perhaps indicative of Japan’s departure from the ‘developmental state’ of the post-war order, the political class overrode the Ministries’ concerns. Members of the political class were surely aware of the political capital to be lost by appealing. Nagoya’s case garnered heavy media interest in part due to the efforts of her lawyers and a non-profit organisation that, according to its website, collected over 410 000 signatures on a petition. Historically responsive to the politics of social policy, the junior coalition partner of Abe’s conservative administration (New Komeito) strongly opposed the decision to appeal. As former Chief Cabinet Secretary, Prime Minister Abe was certainly aware of the high levels of public approval for Prime Minister Koizumi’s decision to overrule a decision of the bureaucracy to appeal a District Court decision finding the Government liable for unconstitutional isolation of persons with leprosy.
In this context, the case was quickly settled and reform was fast-tracked in time for the House of Councillors election in July 2013. The amendment to the Public Offices Election Act repealed the offending provision. It was decided not to emulate Australia’s approach and substitute for guardianship an alternative marker of ‘incapacity’. Instead, the amendment introduced a supported decision-making approach with safeguards against coercion. First, reformers harnessed a pre-existing mechanism in the Act. This allows for one representative (an adult of the elector’s choice) to cast the vote on the elector’s behalf and another to witness that the instructions are followed faithfully. While this had only applied in the case of physical disability, it now includes mental disability. Importantly, it is a form of support, rather than a substituted ‘proxy’ or ‘delegated’ vote. The elector may give instructions remotely, but in this case the representative must be a member of electoral staff registered at the particular polling booth. This system applies also to early votes and votes made in polling places beyond one’s ‘home’ polling station (known as ‘absentee’ voting, which does not include voting by mail).
Second, the Ministry of Internal Affairs constructed a hierarchical regulatory regime for the supervision of votes cast in care institutions and hospitals. Prefectural Electoral Committees may designate such an institution as an absentee polling place. The local returning officer then compiles a list of (essentially volunteer) candidates to undertake ‘external’ witnessing. Where the voter cannot indicate his or her preference on the ballot sheet, the normal rules for voting by a representative apply (though representatives must be staff of the institution deputised as electoral officials).
Conspicuously absent from this regime is the notion of capacity to vote. The repeated stress in the law on guaranteeing the fairness of elections makes it clear that the emphasis is instead on ensuring that voters in institutions are not coerced. While a witness might be in a position to assess capacity, there is no specific guidance on what amounts to a faithful transmission of preference based on capacity to vote. Thus, it seems that, consistent with an inclusionary approach, it is simply presumed that any elector has the capacity to express his or her preference to the representative.
In summary, the salient features of Japan’s approach to the relationship between voting and capacity are as follows. First, movement from a protectionist guardianship model to a supportive one has been catalysed by changing social values inseparable from international trends. This movement toward a more inclusive notion of citizenship was a key plank in a District Court decision that the nexus between guardianship and disenfranchisement is arbitrary and thus unconstitutional given the fundamental importance of the right to vote in a representative, liberal democracy. Second, the solution adopted eschews any problematical attempt to codify or assess an abstract standard of capacity to vote. Instead, it is an inclusionary approach that is supportive and regulatory in nature. The next Part turns to Australian law to argue that the disenfranchisement of persons of ‘unsound mind’ is equally arbitrary and should be consigned to history for similar reasons.
In the area of adult guardianship, Australia has been a pioneer in replacing courts with tribunals, which has tended to produce fewer and less invasive orders. The Australian Electoral Commission has also been active in providing a range of measures to improve access for voters with physical disabilities. However, it has struggled with establishing an appropriate framework to protect and promote the political rights of persons with mental disabilities.
Section 93 of the Electoral Act, which defines the franchise for Commonwealth Parliamentary elections, currently provides that a person is disqualified from voting if he or she is ‘[a] person who ... by reason of being of unsound mind, is incapable of understanding the nature and significance of enrolment and voting’. An objection to a person’s enrolment, which can be made by any person on the electoral roll for that subdivision, must be accompanied by a certificate from a registered medical practitioner certifying that this criterion is met. The person in question is given notice and has 20 days to respond. The Divisional Returning Officer then determines the matter.
The High Court found the exclusion in its current form constitutionally valid in obiter in Roach v Electoral Commissioner. In that case, the issue was the constitutional validity of amendments to the Electoral Act that disenfranchised prisoners (regardless of the term of imprisonment). The majority held that they were not valid, but that the period of three years term of imprisonment as the threshold for disqualification in the pre-existing legislation was acceptable. This is because the pre-existing legislation embodied a proportionate connection between the sanction of exclusion from political life and a sufficiently serious offence, which was required by the commitment to representative government implied in the text of the Constitution. Members of the Court were far less critical in obiter of the exclusion by way of disenfranchisement of persons of unsound mind. Justices Gummow, Kirby and Crennan, albeit in contrasting the operation of the proportionality test to blanket disenfranchisement of prisoners, stated that the provision ‘plainly is valid. It limits the exercise of the franchise, but does so for an end apt to protect the integrity of the electoral process. That end, plainly enough, is consistent and compatible with the maintenance of
the system of representative government’. Justice Hayne, in dissenting to the majority’s view that certain gains in the franchise are irreversible, employed the ‘absurd’ illustration of the Parliament being prevented from reversing its bestowal of voting rights on a person of ‘unsound mind’. Chief Justice Gleeson stated that the rationale is ‘obvious’ and ‘related to the capacity to exercise choice’, yet conceded that ‘the application of the criterion of exclusion may be imprecise, and could be contentious in some cases’.
While the constitutional validity of section 93(8) of the Electoral Act may be obvious to these former members of the High Court, the notion that it is ‘plainly’ obvious has questionable premises, including the factual premise critiqued above relating to the integrity of an election. As demonstrated in the next section, an uncritical acceptance of this notion belies the contingent roots of the exclusion in the highly contested terrain of the 19th century franchise.
The first appearance of the exclusion in Commonwealth law was in section 4 of the Commonwealth Franchise Act 1902 (Cth): ‘No person who is of unsound mind ... shall be entitled to vote at any election of Members of the Senate or the House of Representatives’. While the exclusion is now more than 100 years old, concerns with how it operates in practice and difficulties in establishing criteria for whether a voter is ‘of unsound mind’ have arguably existed since its first appearance.
When introduced, the Commonwealth Franchise Bill 1902 (Cth) made no reference to voters of unsound mind. It disqualified convicted criminals (mirroring the wording of that used in the Constitution regulating membership of the two houses of Parliament) and ‘Aboriginal natives’ (within the limits of the Constitution), but was silent on capacity, as is the Constitution itself. On the second reading of the Bill in the Senate, Protectionist New South Wales Senator Richard O’Connor moved that the provision be redrafted to ‘remedy what is undoubtedly an omission in the Bill’. This was to add ‘no person who is of unsound mind, or in the receipt of charitable relief as an inmate of a public charitable institution’.
O’Connor argued that the need for an amendment to exclude voters of unsound mind was obvious: ‘I do not think there can be any doubt as to the necessity of ... the amendment, if we do not provide that persons of unsound mind shall not have a vote, there is no reason why lunatics should not be put upon the electoral roll and they might claim the right’.
In response to the amendment, Western Australian Free Trader Senator Norman Ewing expressed concern that the basis on which a voter might be excluded was simply unclear:
Who is to say whether a person is ‘of unsound mind’? Any number of persons of unsound mind go to the ballet-box [sic]. The clause should say – ‘no person who is adjudged of unsound mind’. ... It cannot be left to the returning officer to say whether a man is mad or not. ... The whole clause follows the Western Australian Act, and I think it is bad. It should say – ‘a person adjudged to be of unsound mind,’ which would mean adjudged by the Court.
While Ewing appears to advocate for an amendment, none was moved and there was no further debate or discussion on this aspect of the amendment.
Although the exclusion of voters of unsound mind was given cursory attention, the inclusion of voters ‘in receipt of charitable relief as an inmate of a public charitable institution’ was debated at length. O’Connor agreed to abandon it, ostensibly because of constitutional difficulties (such ‘inmates’ were given the vote in South Australia and were therefore guaranteed the Commonwealth franchise under section 41 of the Australian Constitution). Interestingly, the debate suggests that the inclusion in O’Connor’s amendment of both voters of unsound mind and inmates of charitable institutions in the same amendment may have been an attempt to exclude a class of voters who were perceived to lack the capacity to vote. For example, expressing his support for the blanket nature of the exclusion, Senator Henry Dobson referred to the ‘farce’ of the need to convene ‘a kind of conference ... between the doctor, the matron, and the manager [of the institution] to determine, whether a man or woman should ... go to the poll’.
As Ewing’s observations note, O’Connor’s amendment seems to have drawn inspiration from Western Australian law. Initially the Western Australian franchise extended to voters ‘of full age and not subject to any legal incapacity’, but the franchise was relatively narrow due to gender and property criteria (and plural voting). As elsewhere, franchise reform in Western Australia over the next decade was bi-directional. While amendments to the franchise in 1899 extended the franchise to the unpropertied and women, it also made provision to exclude ‘[e]very person who ... [i]s of unsound mind or in the receipt of relief from Government or from any charitable institution’. The Western Australian position excluded a much larger class of voters than the amendment proposed by O’Connor since it was not limited to inmates of charitable institutions. The distinction was not made in Western Australia, perhaps related to the enthusiastic and marginalising approach to institutionalisation in that State at the time. As the debates reveal, it was partly due to a failure to resolve the conceptual difficulties in this distinction that the proposal to import this provision to the federal franchise failed.
In contrast, the ‘unsound mind’ clause was adopted, perhaps because it was recognised as a distinct category of ‘legal incapacity’ long before Western Australia used the term upon self-government. As with the other states, Western Australia received the English Lunacy Regulation Act 1853 (Imp) 16 & 17 Vict, c 70 and related statutes that employed the term ‘unsound mind’. This was codified for the State through the Lunacy Act 1871 (WA), which regulated primarily the committal of lunatics – defined as ‘every person of unsound mind, and every person being an idiot’ – to mental institutions and the management of property upon commitment. The concept itself has deeper roots: ‘the intent of the provision can be traced to the old common law ... under which “idiots” could not vote and “lunatics” could only do so in their “lucid intervals”’.
Perhaps underlying Senator Ewing’s misgivings about the lack of clarity
or due process in assessing capacity to vote was the impression that the Lunacy Act 1871 (WA) had been abused to target politically undesirable groups
in Western Australia including Asians, Indigenous Australians, and women. Without procedural safeguards, the same sentiment might be used to disenfranchise politically undesirable groups at the Commonwealth level. Indeed, the association of being of ‘unsound mind’ and being in ‘receipt of welfare’ suggests that, rather than a ‘natural’ step, codifying the disenfranchisement of those with mental impairment seems to have been connected to prejudice
and anxieties about extending the franchise to a broader segment of society, including the working class, women, Indigenous Australians, criminals, and the morally ‘undeserving’ such as ‘habitual drunkard[s]’ and those receiving social assistance. While these anxieties faded with the democratic development of the franchise, it was ‘criminals’ and ‘lunatics’ and ‘idiots’ (ie, those of ‘unsound mind’) that continued to be excluded at both Commonwealth and state levels. As discussed below, disenfranchisement of criminals has been constitutionally constrained at a federal level and voluntarily abandoned in some states. This is not the case for the last remaining status-based disqualification other than citizenship and age, namely a conception of mental ‘unsoundness’ with roots in the systematic institutionalisation and social control of marginalised individuals in the 19th century.
The issue of whether to revise the ‘unsound mind’ provision in the Commonwealth Electoral Act has been considered at numerous forums. For example, the Joint Select Committee on Electoral Reform, discussing the provision in 1983, noted that the provision was unduly vague and recommended that it should be amended ‘with a view to excluding on the ground only
those persons who are incapable of making any meaningful vote’. Though the Committee suggested that the Commonwealth look to modern state and
territory legislation in the area, it was actually the states that followed the Commonwealth Parliament’s lead after it refined the phrasing to read: ‘by
reason of being of unsound mind, is incapable of understanding the nature
and significance of enrolment and voting’. This was evidently an attempt to transform the provision from one based on status to a functional, decision-specific notion of capacity drawn from the common law.
The Committee’s recommendation to refine the exclusion was adopted and introduced as part of the Commonwealth Electoral Legislation Amendment Act 1983 (Cth). While amendments made by the Act concerning the qualification of voters based on criminal history and age were the subject of debate, the amendment concerning mental capacity was untouched in the House of Representatives. The lack of criteria to determine whether a voter might be of ‘unsound mind’ was the subject of more detailed consideration in the Senate. During the Committee stage of the Bill in the Senate, Senator Michael Macklin raised concerns with the amendment. He noted that there was no effective test proposed in the Act for how to determine the level of an individual’s understanding. In particular, he linked the provision with the questions asked of a voter when they presented themselves to vote, that is, they were only required to provide their name, their address and whether they had voted before – questions ineffective at determining a voter’s capacity. In response, Senator Gareth Evans noted that the phrase had been the subject of judicial interpretation. This was evidently an attempt to transform the provision from one based on status to a functional, decision-specific notion of capacity drawn from the common law. There was no further debate or discussion of the amendment.
In 1989, the exclusion was again amended, this time to introduce the requirement that objections to another person’s enrolment on the basis of unsoundness of mind be ‘supported by a certificate of a medical practitioner’. The amendment was based on a recommendation by the Joint Select Committee on Electoral Reform which in turn appears to have been based on a recommendation made by the Australian Electoral Commission.
However, in its discussion of the amendment, the difficulties apparent in the exclusion and its application were again raised. The Committee asked a number of questions concerning how the exclusion might operate including how a voter subject to an exclusion might be able to object, what other effects a decision to exclude the voter may have and how they might seek re-enrolment. Highlighting the difficulties that the exclusion as a whole presented, the Committee recommended that the provision should be ‘thoroughly reviewed in light of the various issues [the Committee] has cited’. Regrettably, there would appear to have been no review and the subsequent amendment was not subject to any further consideration at the time of its introduction.
As discussed above, Australia’s ratification of the CRPD has catalysed fresh debate over the exclusion in section 93(8)(a) and a range of other laws affecting persons with disabilities. Some organisations have criticised the exclusion on the grounds that it sends a signal that ‘the existence of a cognitive impairment permits a limitation on the exercise of legal agency and thus recognition of legal capacity as a whole’ or that it is ‘vague, stigmatising and overly broad, and does not reflect the true capacity of people with disabilities to make decisions about voting’.
Yet vagueness, anachronistic language, and the retention of a status-based medical model of capacity are no longer the only perceived defects of the exclusion. In particular, the Committee on the Rights of Persons with Disabilities is now concerned that Australia ‘restore presumption of the capacity of persons with disabilities to vote’ in light of its appraisal that ‘persons with disabilities, in particular those with intellectual or psychosocial disabilities, are automatically excluded from the electoral roll’.
A potential solution to the problem is to simply remove the reference to ‘unsound mind’. In 2012, the Joint Standing Committee on Electoral Matters was charged with considering the merits of a Bill that, among other things, proposed to delete this term and instead disqualify a person that ‘in the opinion a qualified person, is incapable of understanding the nature and significance of enrolment and voting’. The Bill defined ‘qualified person’ as:
a person who carries on, and is entitled to carry on, an occupation that involves the provision of care for the physical or mental health of people or for their well-being, and includes any of the following:
(a) a medical practitioner;
(b) a psychiatrist;
(c) a psychologist;
(d) a social worker.
Though the explanatory memorandum claimed that this group of amendments were ‘largely technical in nature, the majority of which raise no human rights issues’, if this Bill were passed, the right to vote could be removed more readily. This is suggestive that the focus of the amendment was not on rights at all, but on paternalistically relieving persons with mental disabilities from the burden of compulsory voting.
Ultimately, on the recommendation of the Joint Standing Committee, the proposed amendment was dropped from the Bill to be given further consideration at a later date. As explained in the Supplementary Explanatory Memorandum, the reasons (certainty and regulatory compliance) given by the Joint Standing Committee seem distant from the focus of the CRPD:
Firstly, while acknowledging that the term ‘unsound mind’ is out-dated, it is of longstanding use in legislation and has an established meaning which provides some certainty. Secondly, one of the amendments increased the range of health professionals who could make an assessment that a person is incapable of understanding the nature and significance of enrolment and voting. [The Joint Standing Committee on Electoral Matters] considered that could result in some persons using the change to circumvent the compulsory voting obligations contained in section 245 of the Electoral Act.
A bolder approach was recommended by the 2014 ALRC Inquiry into Equality, Capacity and Disability in Commonwealth Laws. Initially, in its Discussion Paper, the Commission was in favour of retaining the disqualification, accepting the premise that the provision is ‘for an end apt to protect the integrity of the electoral process’. The Commission even initially suggested that such a ‘controversial proposal’ as deleting the provision ‘would change the nature of voting and voter exclusion in Australia with implications beyond this Inquiry’. Over the course of the Inquiry, this view changed and the Commission recommended in its Final Report that section 93(8)(a) be repealed and not replaced by any new threshold test for capacity.
The reasons for this recommendation were as follows. The Commission accepted that the policy objective of the provision is protective in excusing some persons with disability from compulsory voting. It also noted that by retaining some form of capacity-related qualifications for voting, Australia was not out of step with other democratic countries. The Commission noted the approval given to the disqualification by the High Court and the Joint Standing Committee on Electoral Matters. Yet the Commission observed that these views predate the following statement of the Committee on the Rights of Persons with Disabilities:
Article 29 does not provide for any reasonable restriction or exception for any group of persons with disabilities. Therefore, an exclusion of the right to vote on the basis of a perceived or actual psychosocial or intellectual disability, including a restriction pursuant to an individualized assessment, constitutes discrimination on the basis of disability, within the meaning of article 2 of the Convention.
The Commission acknowledged this view and the general movement
away from a medical paternalist view of disabilities to a rights-based approach. It also noted criticisms made by a number of organisations and agencies that legal capacity is often conflated with mental capacity and that mental disability can operate illegitimately as a bar to a person’s legal agency to exercise their rights. It reiterated criticisms that the exclusion is vague and stigmatising. Furthermore, the Commission found no evidence for the contention that removing the provision ‘would cause any new problems with regard to the integrity of the electoral system, undue influence or fraud’. Indeed, the Commission drew a parallel with women who were denied certain property
rights in the 19th century on the presumed risk that they may be coerced by
their husbands. The Commission also took into account the consensus of stakeholders that there should be no new threshold test.
The Commission therefore proposed an inclusionary model, namely one that retains the right to vote but excuses persons from compulsory voting on the basis of a ‘functional exemption’. A person will be exempt if they demonstrate to a Divisional Returning Officer an inability to
(a) understand information relevant to voting at the particular election;
(b) retain that information for a sufficient period to make a voting decision;
(c) use or weigh that information as part of the process of voting; or
(d) communicate their vote in some way.
These criteria draw from the National Decision-Making Principles, which ‘provide a conceptual overlay, consistent with the CRPD, that is applied
in a Commonwealth decision-making model and provides the basis for review
of relevant Commonwealth, state and territory laws’. They are in the main underpinned by a supported decision-making framework that has developed alongside modern adult guardianship law.
In light of the positive obligations under the CRPD (especially article 29), the Commission also made a number of recommendations to encourage further support for voters with mental disabilities and (in Recommendation 9-7) explore ways of reconciling supported voting with the secret ballot:
Recommendation 9-5: Section 234(1) of the Commonwealth Electoral Act 1918 (Cth) should be amended to provide that if any voter satisfies the presiding officer that he or she is unable to vote without assistance, the presiding officer shall permit a person chosen by the voter to assist them with voting.
Recommendation 9-6: The Australian Electoral Commission should provide its officers with guidance and training, consistent with the National Decision-Making Principles, to improve support in enrolment and voting for persons who require support to vote.
In summary, the salient points of Australia’s position on disenfranchisement on the basis of mental disability are as follows. Rather than being a rule of time immemorial, the origins of this exclusionary provision in Australia are associated with late 19th century prejudice and anxieties about the franchise, which have since been abandoned in other areas such as gender, race, and class. The notion conveyed by some authoritative voices within Parliament, the executive and the courts that it is necessary, unobjectionable or uncontested, belies the concerns that have been expressed with the exclusion historically and in the present. In contrast, the most recent reform proposal has recommended an inclusionary approach that does not subject persons with mental disabilities to any burden of proving capacity that no other citizens are required to bear.
Indeed, this supportive and inclusionary approach recommended by the ALRC is fundamentally similar to the new Japanese model, albeit adapted to the Australian context. This context includes compulsory voting, where the National Decision-Making Principles are employed to excuse rather than exclude. Australia also lacks any pre-existing provision for voting through a supervised representative. There are existing provisions for ‘assistance’, but only where the voter is present and there are no checks as to whether the assistance is faithful to directions. Yet there is no reason why Australia could not also introduce representatives and witnesses, capitalising on the considerable volunteer workforce that mobilises for Australian elections and the extensive network of mobile polling locations including hospitals and nursing homes. The ALRC is wary of departing from the principle of a secret ballot due to conformity with the CRPD and a local election ‘culture’ embraced strongly by electoral officials. In the same vein, there is a paradoxical risk that persons once overlooked in the Australian Electoral Commission’s broad discretion in enforcing compulsory voting will now bear an added burden of seeking to be excused through these new formal channels. However, in light of the perceived requirements of the CRPD, the Commission is clear that the exemption should not be made on the basis of status (ie, mental disability) or through a medical certificate.
There is thus considerable discretion as to how a Returning Officer might assess the validity of an exemption. There is no mention of the role that enabling technology may play in expanding opportunities for participation specific to mental disability and voting. This is relevant to discretion because an official may accept the word of a carer or guardian who too easily opts for an exemption without a sound understanding of the potential role technology can play in facilitating communication. Nor is the role of an ‘assistant’ specified with precision. This vagueness is directed at support and exemptions rather than exclusion and is therefore less concerning from a rights perspective. Even still, after explaining the comparative methodology of the article, the next Part seeks to better conceptualise the assistant’s role and the impact of values on such discretions.
The purpose here of comparing the Japanese and Australian approaches to reform in the area of voting rights for persons with mental disabilities is twofold. The first is a normative project of determining ‘best practice’ in the area applying the criteria of international human rights and liberal democratic principles. The second is to ‘enhance capacity for self-reflection’ by generating a meta-perspective that enables us to better understand the nature and significance of voting and the various processes by which electoral reform proceeds. This methodology faces a number of challenges. The universalism assumed by the first purpose rests on the ability to identify a normative basis and either discount or account for the distinctive cultural and historical contexts that exist across jurisdictions. Related to this latter problem is the complication for comparative analysis posed by the expressive role played by constitutional law (defined broadly to include electoral law) to express, help to constitute, or influence national identity. This expressive role is most evident in the language of constitutional preambles, but issues such as defining the franchise also speak to how a nation sees itself. Despite these challenges, matters such as the rights of persons with disabilities recognised in international human rights law are by their nature suited to a universalist, normative methodology. Furthermore, this is consistent with the distinctively expressive role that constitutional law may have in a particular jurisdiction. Indeed, this expressive effect may be an integral part of self-reflection in aid of understanding what we mean by the right to vote in Australia.
With regard to the normative purpose, it is the authors’ view that Japan has adopted a more CRPD-compliant approach by leapfrogging the Australian position and moving directly from disenfranchisement through an order of adult guardianship to abandoning capacity tests altogether in relation to the franchise. It may be retorted that the Australian legal system differs from that of Japan in its constitutional framework (for example, no bill of rights), different electoral laws (that, for example, adopt compulsory voting), and adult guardianship systems (largely administered by tribunals rather than courts). The private ordering
(by voter objection) approach adopted by Australia disenfranchises a smaller number of individuals than was the case in Japan. Section 93(8) seems to be enforced much more selectively than more ‘automatic’ mechanisms such
as guardianship or criminal incarceration. Accordingly, a direct comparison between the decision of a Japanese district court on the constitutional validity of its disenfranchisement through court-ordered adult guardianship and the equivalent question in Australian law is flawed. However, this article does not seek to make a comparison of this direct nature.
Certainly, the Japanese case and the experiences of other jurisdictions do have clues for how the obiter of the Australian High Court could be challenged. A fundamental commonality is that Australian constitutional jurisprudence recognises limitations (albeit implied) on Parliament’s capacity to define the franchise. This is rooted in a principle of non-discrimination that has evolved alongside concepts of popular sovereignty and representative democracy. It is true that the constitutional tests differ across jurisdictions. The Japanese court applied a test of ‘unavoidable circumstances’ as the sole justification to depart from the universal franchise. This is closer to the test of ‘clear and present danger’ in US constitutional law than the weaker Australian test of ‘reasonably appropriate and adapted to serve an end which is consistent or compatible with the maintenance of the constitutionally prescribed system of representative government’. However, the significance of Australian cases such as Roach and Rowe is in part the introduction of this proportionality test in a way that strengthens judicial oversight over the Australian Parliament’s ability to define the franchise. Recent case law on constitutional protections has also indicated a greater willingness on the part of the High Court to review hard data related to a law’s implementation. The upshot is potentially closer oversight of matters such as evidence for claims that disenfranchisement on the basis of mental incapacity protects the integrity of elections in a material way, or in a way that is unachievable through more proportionate alternative means. Proportionality also limits the means available to achieve other goals such as relieving the burden of voting upon persons with mental disability.
A proportionality test might also look at the surrounding provisions and the way in which they are implemented. For example, the Electoral Act and other legislation provide avenues for administrative and judicial review. This could ensure that common law presumptions of capacity are not rebutted without reasonable evidence. It might also provide greater clarity to the test along the lines of separating out a doctor’s assessment of ‘mental unsoundness’ from the question of whether the voter understands the nature and significance of enrolment and voting. Nevertheless, the existence of these channels and their capacity for utilisation are separate matters. As noted earlier, between 2008–09 and 2011–12, 28 603 voters had been removed under this provision nationally. Despite its ongoing use, the operation of the provision would appear to have never been the subject of either administrative or judicial review. This is unsurprising given that the persons in the best position to assist in an application for review are precisely those most likely to have lodged the objection in the first place. In other words, the objecting party is most likely to be in a carer or guardianship relationship with the person and has the benevolent, though paternalistic, motive of ensuring that the person is not fined for failing to vote.
Japan is of course only one jurisdiction with potential comparative lessons. New Zealand, for instance, equates incapacity to vote with compulsory institutionalisation under the Mental Health (Compulsory Assessment and Treatment) Act 1992 (NZ). Orr argues that, if adapted to Australia’s system of compulsory voting, this could amount to a compelling switch from excluding to excusing persons with mental disabilities from voting. This is true, yet the criterion of institutionalisation reopens a vexed distinction identified over 100 years ago when the Australian franchise was initially being debated. Another possibility is to introduce greater procedural safeguards such as automatic court or tribunal oversight over all decisions about disenfranchisement on the basis of mental incapacity. This does not resolve the problems of conceptualising capacity to vote, and apparently would not satisfy the United Nations Committee on the Rights of Persons with Disabilities. Schriner proposes that if there is to remain a test of capacity it should be universally applied, rather than applied
only to those who, by virtue of the disability, may be placed in the position of
bearing the onus of demonstrating capacity to vote. This might take the form of providing basic personal information to be enrolled, which would be enabled by support for those with disabilities. Whether this amounts to a test at all is questionable, and if it were extended beyond such basic information it would veer dangerously toward the types of ‘educational qualification’ declared by one High Court Justice (albeit in dissenting obiter) to be inconsistent with the contemporary franchise and thus unconstitutional.
Therefore, the better option is to follow Japan’s lead and abandon any test or disqualification on the basis of mental incapacity (however phrased or conceptualised) and instead focus on support and protection enabling marginalised individuals to vote. The ALRC has provided a compelling proposal for adapting this approach to the Australian context, namely excusing individuals’ obligation to vote by applying the National Decision-Making Principles. As noted above, there remains a degree of uncertainty in this approach and the application in this context of principles originating in guardianship regimes devised to protect vulnerable individuals (and their estates) from harmful transactions is problematic. Nevertheless, the modern formulation employs a functional test that, in the form of an excuse rather than an exclusion, is consistent with the treatment afforded to all other citizens who, for a variety of reasons, are excused from voting on the grounds of fairness rather than capacity.
As noted above, Japan’s approach does not answer the question of what is the precise role of an assistant or representative under a ‘supported voting’ model. It could merely be an extension of the assistance hitherto given to persons with physical disabilities and blindness, namely standing in for the physical act of voting. In the case of mental disabilities, however, the role may be more of an interpretative one, including identifying a particular voting intention. The decision here may be less a unidirectional signal and more a process of dialogue and support. Because of this, even if it is not employed as a marker for capacity to vote, there remains a role for the theory of modern guardianship (as embodied by the CRPD) in guiding the practice of supported voting. Indeed, for persons who need support in the form of a guardian in other areas of their lives, the guardian would be a logical choice for voting support.
A guardian’s authority to vote is specifically barred by legislation in a number of jurisdictions. Yet one of the largely unexplored ramifications of the shift from a substitute to supported decision-making framework may be that this is permissible, along with other non-substitutable decisions such as consenting to marriage or making a will. Certainly, the argument that these decisions are too personal to make is no longer tenable if the underlying principle is that the decision is in no way a substitution. The authors are reluctant to propose that this approach be introduced without further discussion given the criticisms of some interpretations of the CRPD as to what supported decision-making actually entails. For example, a question arises as to how far along the spectrum a guardian may travel from identifying a voting intention to actively influencing that decision. Where the person in question is comatose or catatonic and enabling technology has been exhausted to no avail, it is difficult to see that any voting decision could be anything but a substitution. It might be argued that the pre-expressed intentions of the person may remedy the problem that, in such cases, the life authorship expressed through voting is absent. It could even assuage the regulatory concerns detailed in Part II over manipulation of vulnerable voters. This would also overcome the problem that with some mental disabilities, such as dementia, capacity fluctuates and discerning an intent may take time. Yet, perhaps more so for voting than other decisions, information about a particular election cannot be truly anticipated at any great interval before election day, especially before the close of nominations. It is therefore the authors’ view that, while decision-making assistance for an elector might be supported by guardianship theory, providing ongoing authority for guardians to vote on behalf of another is unwarranted. As Carney and others have observed above, there are good reasons to have faith in the capacity of civil society and informal support networks to function in the spirit of the CRPD. These networks can themselves be supported by educative programs and a web of accountability spanning government and non-government sectors.
Beyond the normative conclusion that Australia should follow Japan’s lead, there are comparative lessons to be drawn in enabling us to reflect on how the franchise is conceived within a society, and the ‘constitutive’ characteristics of voting identified in the American context by Winkler:
First, voting may be an expression of the individual’s identity: the voter, by the act of participation, expresses the ideas and sentiments of belonging and membership. ... Second, American society uses the institution of elections – and the voting inherent in them – to express its values to the members of the community, transmitting and regenerating the ideals that form the core of our cultural identity.
It is precisely at the margins of the franchise where these identity characteristics of voting are most important and most threatened. While attitudes toward the inclusion of persons with disabilities have shifted over time, new threats of an existential nature have emerged alongside reproductive technologies designed to screen genetic disabilities such as Takumi Nagoya’s disability (Down syndrome) out of existence. There is therefore a perennial danger that the constitutive characteristics of voting will be lost where decisions over how the franchise is defined and implemented are relegated to a bureaucracy driven by a legitimate concern for the independent and professional administration of elections. There are peculiar risks in Australia, where there is a tendency to emphasise the duty aspect of voting through the system of compulsory voting.
These extra-legal and attitudinal factors are just as important as the ‘hard’ law. Ultimately, under any model that departs from automatic exclusions, the materialisation of the right to vote may turn on the attitudes of officials, carers, and volunteers. These may be driven by paternalistic concerns rather than a deeper reflection of the significance of voting for self-determination and citizenship for marginalised individuals. It is also necessary therefore to look beyond the law to programs aimed at fostering this reflection on the part of the wider community. Of course, given the expressive effects of law described above, law reform itself is part of this educative process.
Voting is an essential component of a representative democracy, but also a fundamental act of citizenship and a symbol of belonging to a political community. ‘Unsoundness of mind’ is one of the few exceptions to the universal franchise to have survived to the present. Despite the necessary caveats, a comparison across jurisdictions, Australia and Japan in particular, allows us to conclude that this exclusion is non-compliant with the CRPD and that the Australian Parliament should therefore repeal the ‘unsound mind’ disqualification upon voting and enrolment. It should instead institute exemptions from compulsory voting on the grounds of impaired decision-making capacity. These exemptions should operate on the basis that exhaustive support – enabled by technology and informed by modern guardianship principles – is provided to all persons with mental disabilities. To ensure that latent paternalistic values do not prevent carers, guardians, and officials from adequately providing or seeking these supports, all Australian governments should actively disseminate information about these supports and the values and principles that underpin them.
[*] Assistant Professor, University of Canberra.
[**] Sessional Lecturer, University of Canberra.
[***] Assistant Professor, University of Canberra.
 Opened for signature 30 March 2007, 2515 UNTS 3 (entered into force 3 May 2008).
 Australian Law Reform Commission, Equality, Capacity and Disability in Commonwealth Laws, Report No 124 (2014) (‘ALRC Report’).
 Terminology varies: traditional common law nomenclature referred to substitute decision-makers as ‘guardians’; however current legislation includes terms such as ‘administrator’, ‘guardian’ and ‘manager’: see, eg, Guardianship and Management of Property Act 1991 (ACT) pt 2; Guardianship and Administration Act 2000 (Qld) ch 3; Guardianship and Administration Act 1990 (WA) s 3.
 Note also that certain decisions remain exclusively the province of the relevant protective body, commonly the guardianship tribunal, of each respective jurisdiction. Such decisions commonly include special medical care decisions. Other limitations on a substitute decision-maker’s powers include issues of adoption, marriage and testamentary disposition, as well as voting: see, eg, Guardianship and Management of Property Act 1991 (ACT) s 7B.
 Nick O’Neill and Carmelle Peisah, Capacity and the Law (Sydney University Press, 2011) [5.2.1] <http://worldlii.austlii.edu.au/au/journals/SydUPLawBk/2011/7.html#Heading22> .
 John Seymour, ‘Parens Patriae and Wardship Powers: Their Nature and Origins’ (1994) 14 Oxford Journal of Legal Studies 159, 167, quoting Falkland v Bertie  EngR 8; (1696) 23 ER 814, 818.
 Ibid 162.
 See Israel Doron, ‘Elder Guardianship Kaleidoscope – A Comparative Perspective’ (2002) 16 International Journal of Law, Policy and the Family 368.
 Noting that Australian tribunals have the power to attach specific conditions to guardianship orders under legislation, particularly in the event that the person has capacity to make some decisions relating to a particular aspect of their life, but doesn’t demonstrate full capacity: see, eg, Guardianship and Management of Property Act 1991 (ACT) s 7(2).
 For an excellent treatment of the shift from courts to tribunals in Australia, see Terry Carney and David Tait, The Adult Guardianship Experiment: Tribunals & Popular Justice (Federation Press, 1997).
 Kristin Booth Glen, ‘Changing Paradigms: Mental Capacity, Legal Capacity, Guardianship, and Beyond’ (2012) 44 Columbia Human Rights Law Review 93, 124.
 Ibid 167–8.
 See, eg, 2010 World Congress on Adult Guardianship Committee, ‘Proceedings of the First World Congress on Adult Guardianship Law 2010 (2010 nen seinen kouken hou sekai houkokushoshuu)’ (Yokohama, 2011); Glen, above n 14; Robyn Carroll and Anita Smith, ‘Mediation in Guardianship Proceedings for the Elderly: An Australian Perspective’ (2010) 28 Windsor Yearbook of Access to Justice 53; Robert D Dinerstein, ‘Implementing Legal Capacity under Article 12 of the UN Convention on the Rights of Persons with Disabilities: The Difficult Road from Guardianship to Supported Decision-Making’ (2012) 19(2) Human Rights Brief 8; Phillip B Tor and Bruce D Sales, ‘A Social Science Perspective on the Law of Guardianship: Directions for Improving the Process and Practice’ (1994) 18 Law and Psychology Review 1.
 CRPD art 1.
 See Annegret Kämpf, ‘The Disabilities Convention and Its Consequences for Mental Health Laws in Australia’ in Bernadette McSherry (ed), International Trends in Mental Health Laws (Federation Press, 2008) 10; Rosemary Kayess and Phillip French, ‘Out of Darkness into Light? Introducing the Convention on the Rights of Persons with Disabilities’ (2008) 8 Human Rights Law Review 1.
 See Shih-Ning Then, ‘Evolution and Innovation in Guardianship Laws: Assisted Decision-Making’  SydLawRw 6; (2013) 35 Sydney Law Review 133.
 Committee on the Rights of Persons with Disabilities, General Comment No 1: Article 12: Equal Recognition before the Law, 11th sess, UN Doc CRPD/C/GC/1 (19 May 2014) .
 Ibid .
 Ibid –.
 Nina A Kohn, Jeremy A Blumenthal and Amy T Campbell, ‘Supported Decision-Making: A Viable Alternative to Guardianship?’ (2013) 117 Penn State Law Review 1111, 1121.
 Soumitra Pathare and Laura S Shields, ‘Supported Decision-Making for Persons with Mental Illness: A Review’ (2012) 34(2) Public Health Reviews 1, 4.
 Kohn, Blumenthal and Campbell, above n 24, 1121.
 See Henriette Sinding Aasen et al (eds), Juridification and Social Citizenship in the Welfare State (Edward Elgar, 2014).
 See Trevor Ryan, ‘Is Japan Ready for Enduring Powers? A Comparative Analysis of Enduring Powers Reform’ (2014) 9 Asian Journal of Comparative Law 241.
 Kohn, Blumenthal and Campbell, above n 24, 1121–2.
 Michael Bach and Lana Kerzner, ‘A New Paradigm for Protecting Autonomy and the Right to Legal Capacity: Advancing Substantive Equality for Persons with Disabilities through Law, Policy and Practice’ (Commissioned Paper, Law Commission of Ontario, October 2010) 83–94.
 Victorian Law Reform Commission, Guardianship, Report No 24 (2012) 89, 125–50.
 Terry Carney, ‘Supported Decision-Making for People with Cognitive Impairments: An Australian Perspective?’ (2015) 4 Laws 37, 53.
 John Chesterman, ‘Capacity in Victorian Guardianship Law: Options for Reform’  MonashULawRw 27; (2010) 36(3) Monash University Law Review 84, 92.
 Carney, ‘Supported Decision-Making’, above n 33, 53.
 Chesterman, above n 35, 100.
 Article 29 of the CRPD provides that:
States Parties shall guarantee to persons with disabilities political rights and the opportunity to enjoy them on an equal basis with others, and shall undertake:
(a) To ensure that persons with disabilities can effectively and fully participate in political and public life on an equal basis with others, directly or through freely chosen representatives, including the right and opportunity for persons with disabilities to vote and be elected, inter alia, by:
(i) Ensuring that voting procedures, facilities and materials are appropriate, accessible and easy to understand and use;
(ii) Protecting the right of persons with disabilities to vote by secret ballot in elections and public referendums without intimidation, and to stand for elections, to effectively hold office and perform all public functions at all levels of government, facilitating the use of assistive and new technologies where appropriate;
(iii) Guaranteeing the free expression of the will of persons with disabilities as electors and to this end, where necessary, at their request, allowing assistance in voting by a person of their own choice;
(b) To promote actively an environment in which persons with disabilities can effectively and fully participate in the conduct of public affairs, without discrimination and on an equal basis with others, and encourage their participation in public affairs, including:
(i) Participation in non-governmental organizations and associations concerned with the public and political life of the country, and in the activities and administration of political parties;
(ii) Forming and joining organizations of persons with disabilities to represent persons with disabilities at international, national, regional and local levels.
 CRPD art 1.
 Committee on the Rights of Persons with Disabilities, General Comment No 1: Article 12: Equal Recognition before the Law, 11th sess, UN Doc CRPD/C/GC/1 (19 May 2014) [29(f)].
 See Kiss v Hungary (European Court of Human Rights, Chamber, Application No 38832/06, 20 May 2010), in which the European Court of Human Rights held that disenfranchisement on the basis of partial guardianship breached art 3 of Protocol to the Convention for the Protection of Human Rights and Fundamental Freedoms, opened for signature 20 March 1952, 213 UNTS 262 (entered into force 18 May 1954) (the right to free elections).
 Terry Carney, ‘Civil and Social Guardianship for Intellectually Handicapped People’  MonashULawRw 5; (1982) 8 Monash University Law Review 199, 215.
 Loi nº 68-5 du 3 janvier 1968 portant réforme du droit des incapables majeurs [Law No 68-5 of 3 January 1968 on the Reform of the Right of Incapacitated Adults] (France) JO, 4 January 1968. See also George Szmukler, ‘Fifty Years of Mental Health Legislation: Paternalism, Bound and Unbound’ in Sidney Bloch, Stephen A Green and Jeremy Holmes (eds), Psychiatry: Past, Present, and Prospect (Oxford University Press, 2014) 133, 135‑7, 147; see also the discussion of the changing approach to the ‘classification’ of disabilities in Myriam Winance, Isabelle Ville and Jean-François Ravaud, ‘Disability Policies in France: Changes and Tensions between the Category-based, Universalist and Personalized Approaches’ (2007) 9 Scandinavian Journal of Disability Research 160.
 Code civil du Québec [Civil Code of Quebec], SQ 1991, c 64, arts 256–9.
 Judgment of the Tokyo District Court (2013) 45 Jissen Seinen Kouken 92.
 Electoral Administration Act 2006 (UK) c 22, s 73(1): ‘Any rule of the common law which provides that a person is subject to a legal incapacity to vote by reason of his mental state is abolished’.
 Canadian Disability Rights Council v Canada  3 FC 622.
 Judgment of the Tokyo District Court (2013) 45 Jissen Seinen Kouken 92.
 Regeringsformen [The Instrument of Government] (Sweden) No 1974: 152, ch 3, art 2; European Union Agency for Fundamental Rights, ‘The Right to Political Participation of Persons with Mental Health Problems and Persons with Intellectual Disabilities’ (Report, October 2010) 19.
 Bundesgesetz über die Wahl des Nationalrates (Nationalrats-Wahlordnung 1992 – NRWO) [Federal Law on National Council Elections (Regulations on National Council Elections 1992 – NRWO)] (Austria), BGBl 471/1992, as amended by BGBl II 147/2008 (7 May 2008).
 Code électoral [Electoral Code] (France) art L5.
 Doe v Rowe, 156 F Supp 2d 35 (D Me, 2001). The decision was noted by the US Supreme Court in Tennessee v Lane,  USSC 1762; 541 US 509 (2004) concerning the application of Title II of the Americans with Disabilities Act of 1990, 42 USC §§ 12 131–65 (1990), but the application of the section to voting was not considered.
 Jonathon Savery, ‘Voting Rights and Intellectual Disability in Australia: An Illegal and Unjustified Denial of Rights’  SydLawRw 14; (2015) 37 Sydney Law Review 287, 297.
 The common law position applicable in a number of Australian states is stated in Gibbons v Wright  HCA 17; (1954) 91 CLR 423, 438 (Dixon CJ, Kitto and Taylor JJ): ‘the mental capacity required by the law in respect of any instrument is relative to the particular transaction which is being effected by means of the instrument, and may be described as the capacity to understand the nature of that transaction when it is explained’.
 See, eg, Human Rights Law Centre, Submission No 54 to Australian Law Reform Commission, Equality, Capacity and Disability in Commonwealth Laws, January 2014, 9.
 Kohn, Blumenthal and Campbell, above n 24, 1151.
 Kim Rubenstein, ‘Citizenship and the Centenary – Inclusion and Exclusion in 20th Century Australia’  MelbULawRw 24; (2000) 24 Melbourne University Law Review 576, 578.
 Peter Rabins notes that the ‘ethical issues [of guardianship] are different than for adults’: Peter V Rabins, ‘Issues Raised by Research Using Persons Suffering from Dementia Who Have Impaired Decisional Capacity’ (1998) 1 Journal of Health Care Law & Policy 22, 23.
 Joint Standing Committee on Electoral Matters, Parliament of Australia, Advisory Report on the Electoral and Referendum Amendment (Improving Electoral Procedure) Bill 2012 (2012) 29 (table 2.1).
 In a 2000 submission to the Commonwealth Joint Standing Committee on Electoral Matters concerning fraudulent voting, the Australian Electoral Commission noted that the magnitude of irregularities in a single electorate of 70 000 would need to be substantial: Australian Electoral Commission, Submission No 26 to Joint Standing Committee on Electoral Matters, Parliament of Australia, Inquiry into the Integrity of the Electoral Roll, 17 October 2000, [8.9]. See also Savery, above n 54, 296–7.
 Hiroshi Oda, Japanese Law (Oxford University Press, 2nd ed, 1999) 1.
 Kōshoku-senkyo-hō [Public Offices Election Act] (Japan) Act No 100 of 1950, s 11(1)(a) (‘Public Offices Election Act’).
 Mihoko Okamura, ‘The Adult Guardianship System (seinen kouken seido)’ in National Diet Library (ed), Declining Fertility, Ageing and Countermeasures (shoushi koureika to sono taisaku) (2005) 198, 199. A good English-language comparative source is Terry Carney, ‘Aged Capacity and Substitute Decision-Making in Australia and Japan’ [2003–04] LAWASIA Journal 1.
 Okamura, above n 64, 200.
 ‘Disenfranchisement through Guardianship Unconstitutional (seinen kouken de senkyoken soushitsu wa iken)’ (2013) 3 Kousei Fukushi 6, 7.
 See Cabinet Office website: People with Disabilities Reform Promotion Council, People with Disabilities Reform Promotion Headquarters (8 December 2009) <http://www8.cao.go.jp/shougai/suishin/kaikaku/
 Judgment of the Tokyo District Court (2013) 45 Jissen Seinen Kouken 92, 102.
 Act on the Implementation of Disability Health and Welfare Policies toward Inclusive Communities [chiikishakainiikerukyousei no jitsugennimuketearatanashougaihokenfukushishisaku o kouzuru tame no kaneihouritsu no seirinikansuruhouritsu] (Japan) Act No 51 of 2012.
 Judgment of the Tokyo District Court (2013) 45 Jissen Seinen Kouken 92, 102
 Nagoya’s father had taught Nagoya that voting was an important duty and she had diligently kept herself informed of the timing of elections. Although Nagoya had basic literacy, she was weak in numeracy so her father applied for full statutory guardianship because of an anxiety about her future shared by many ageing parents of children with disabilities: ibid 99.
 Ibid 98.
 Minpō [Civil Code] (Japan) Act No 89 of 1896, s 7. The official translation uses the term ‘constantly lacks the capacity to discern right and wrong’, which is one possible translation of ‘riji o benshoku suru nouryoku o kaku joukyou’, but it is inconsistent with both the context of the provision and judicial interpretation (see below).
 Judgment of the Tokyo District Court (2013) 45 Jissen Seinen Kouken 92, 98.
 Ibid 103.
 Ibid 98.
 Ibid 103.
 Ibid 104.
 Judgment of the Supreme Court of Japan (2005) 59(7) Minshu 2087. Rights in proportional national voting blocs under Japanese hybrid system were granted under the Act. While the government attempted to distinguish the 2005 case on the basis that it concerned the inability to exercise a right rather than the existence of the right itself, the District Court rejected the relevance of this distinction in the application of the constitutional test.
 Understandable given the Occupation-era origins of Japan’s Constitution: see John W Dower, Embracing Defeat: Japan in the Wake of World War II (Norton, 1999) 346–99.
 Nihonkoku Kenpō [Constitution] (Japan) 3 November 1946. For example, article 1 states:
The Emperor shall be the symbol of the State and of the unity of the People, deriving his position from the will of the people with whom resides sovereign power.
Article 15 states:
The people have the inalienable right to choose their public officials and to dismiss them. All public officials are servants of the whole community and not of any group thereof. Universal adult suffrage is guaranteed with regard to the election of public officials. In all elections, secrecy of the ballot shall not be violated. A voter shall not be answerable, publicly or privately, for the choice he has made.
Article 14 states:
All of the people are equal under the law and there shall be no discrimination in political, economic or social relations because of race, creed, sex, social status or family origin.
Article 44 states:
The qualifications of members of both Houses and their electors shall be fixed by law. However, there shall be no discrimination because of race, creed, sex, social status, family origin, education, property or income.
 Judgment of the Supreme Court of Japan (2005) 59(7) Minshu 2087.
 Judgment of the Tokyo District Court (2013) 45 Jissen Seinen Kouken 92, 98; Minpō [Civil Code] (Japan) Act No 89 of 1896, s 7.
 Minpō [Civil Code] (Japan) Act No 9 of 1898, ss 738, 764, 780, 966.
 Minpō [Civil Code] (Japan) Act No 9 of 1898, ss 742, 963.
 Judgment of the Tokyo District Court (2013) 45 Jissen Seinen Kouken 92, 99.
 Ibid 101.
 Ibid 98.
 Ibid 99.
 Ibid 101.
 Minpō [Civil Code] (Japan) Act No 9 of 1898, s 858.
 Judgment of the Tokyo District Court (2013) 45 Jissen Seinen Kouken 92, 101.
 Ibid 99–100 (translated by the authors).
 Okamura, above n 64, 200.
 Judgment of the Tokyo District Court (2013) 45 Jissen Seinen Kouken 92, 101.
 A Frank Johns, ‘Ten Years After: Where Is the Constitutional Crisis with Procedural Safeguards and Due Process in Guardianship Adjudication?’ (1999) 7 Elder Law Journal 33, 44–5.
 Judgment of the Tokyo District Court (2013) 45 Jissen Seinen Kouken 92, 102.
 Ibid 100.
 Ibid. This is particularly true of some local elections that require that the voter write the candidate’s name on the ballot paper.
 Ibid 104.
 ‘Disenfranchisement through Guardianship Unconstitutional’, above n 66, 6.
 ‘Judgment Finds Disenfranchisement for Guardianship Unconstitutional, Government Appeals (seinen kouken de senkyokensoushitsu no ikenhanketsu, seifugakouso)’, Nikkei News (Tokyo), 27 March 2013 (‘Judgment Finds Disenfranchisement for Guardianship Unconstitutional’). Note that under Japanese constitutional jurisprudence, the finding applied only to the plaintiff and did not render the provision in its application to others void. The Diet is expected to remedy the defect, but the provision is not struck down as it is in other jurisdictions such as the US and Australia: Oda, above n 62, 42.
 Jeff Kingston, Japan’s Quiet Transformation: Social Change and Civil Society in the Twenty-First Century (RoutledgeCurzon, 2004) 118–19.
 Nagase Osamu, Deprivation of Voting Rights Found Unconstitutional in Japan (14 March 2013) Inclusion International <http://inclusion-international.org/depriviation-of-voting-rights-found-unconstitutional-in-japan> .
 Frances McCall Rosenbluth and Michael F Thies, Japan Transformed: Political Change and Economic Restructuring (Princeton University Press, 2010) 62–3.
 Judgement Finds Disenfranchisement for Guardianship Unconstitutional, above n 117.
 Judgment of the Kumamoto District Court (2001) 1748 Hanrei Jihou 30.
 Seinenhikoukennin no senkyoken no kaifukutou no tame no koushokusenkyohoutou no ichibu o kaiseisuruhouritsu [Act to Reform the Public Offices Election Act to Restore Voting Rights to Persons under Adult Guardianship] (Japan) Act No 21 of 2013. Parallel reforms were made with regard to elections for prefectural assemblies and governors and constitutional referenda.
 Public Offices Election Act s 48. Note that the supervision and administration of even national elections is undertaken by the lowest level of government. The returning officer and electoral staff are appointed by local electoral committees: at ss 37–8.
 Koushoku senkyohoushi kourei [Public Offices Election Act Enforcement Ordinance] (Japan) Ordinance No 89 of 1950, ord 56(4).
 Ministry of Internal Affairs and Communications (Japan), Seinenhikoukuennin no senkyoken no kaifukutou no tame no koushokusenkyohou no ichibu o kaisei suru houristsu no shikou ni tsuite [On the Commencement of the Act to Partially Reform the Public Offices Election Act to Restore the Right to Vote to Persons Under Adult Guardianship], Electoral Circular 46, 31 May 2013.
 Koushoku senkyohoushi kourei [Public Offices Election Act Enforcement Ordinance] (Japan) Ordinance No 89 of 1950, ord 50(5).
 Ministry of Internal Affairs and Communications (Japan), above n 126. By way of regulating the regulators, the law imposes a duty on returning officers at an institution to ‘endeavour to implement voting fairly’ and actively cooperate with prefectural and local government in the new scheme: Public Offices Election Act s 49(9). The national Ministry provides templates for the considerable paperwork attached to this process, including approvals, reporting, and claims from institutions to the prefecture to defray costs, including remunerating witnesses to an amount set by the local election committee based on local regulations.
 Ministry of Internal Affairs and Communications (Japan), above n 126. Because the institution fits within the meaning of an ‘absentee polling place’, for which special rules are provided, just as special rules are provided for voting at overseas embassies: Koushoku senkyohoushi kourei [Public Offices Election Act Enforcement Ordinance] (Japan) Ordinance No 89 of 1950, ord 56(4).
 Carney and Tait, above n 13, 192.
 See Australian Electoral Commission, Report on Performance: Assisting Australians with Special Needs (12 October 2012) 2011–12 Annual Report <http://annualreport.aec.gov.au/2012/report-on-performance/
 Electoral Act s 93 states:
(1) Subject to subsections (7) and (8) and to Part VIII, all persons:
(a) who have attained 18 years of age; and
(b) who are:
(i) Australian citizens; or
(ii) persons (other than Australian citizens) who would, if the relevant citizenship law had continued in force, be British subjects within the meaning of that relevant citizenship law and whose names were, immediately before 26 January 1984:
(A) on the roll for a Division; or
(B) on a roll kept for the purposes of the Australian Capital Territory Representation (House of Representatives) Act 1973 or the Northern Territory Representation Act 1922;
shall be entitled to enrolment.
 Electoral Act s 93(8)(a). The provision continues:
(b) has been convicted of treason or treachery and has not been pardoned; is not entitled to have his or her name placed or retained on any Roll or to vote at any Senate election or House of Representatives election.
 Electoral Act ss 114(1A), 116(4)(b).
 Electoral Act s 118(1).
 Electoral Act s 118.
  HCA 43; (2007) 233 CLR 162 (‘Roach’).
 Ibid 182 – (Gleeson CJ), 204 –,  (Gummow, Kirby and Crennan JJ).
 Ibid 176–82 – (Gleeson CJ), 197–204 – (Gummow, Kirby and Crennan JJ).
 Ibid 200  (Gummow, Kirby and Crennan JJ).
 Ibid 217 .
 Ibid 175 .
 This was later incorporated into Electoral Act. The Commonwealth Electoral Act 1902 (Cth) had regulated parliamentary elections separately to the franchise itself. The Electoral Act carried the ‘unsound mind’ provision over to s 39(4). The provision was then moved to s 93(8) when the sections were later renumbered pursuant to Commonwealth Electoral Legislation Amendment Act 1984 (Cth) s 5. The omitted text is:
and no person attainted of treason, or who has been convicted and is under sentence or subject to be sentenced for any offence punishable under the law of any part of the King’s dominions by imprisonment for one year or longer ...
 There are no disqualifications in the Constitution on voting itself. Section 8 merely states that:
The qualification of electors of senators shall be in each state that which is prescribed by this Constitution, or by the Parliament, as the qualification for electors of members of the House of Representatives; but in the choosing of senators each elector shall vote only once.
Section 30 states:
Until the Parliament otherwise provides, the qualification of electors of members of the House of Representatives shall be in each State that which is prescribed by the law of the State as the qualification of electors of the more numerous House of Parliament of the State; but in the choosing of members each elector shall vote only once.
 The second paragraph of s 4 of the Commonwealth Franchise Act 1902 (Cth) stated:
No aboriginal native of Australia Asia Africa or the Islands of the Pacific except New Zealand shall be entitled to have his name placed on an Electoral Roll unless so entitled under section forty-one of the Constitution.
 Australian Constitution s 44(ii). Interestingly this was not the case in the states. A number of states’ constitutions rendered a member of Parliament’s seat vacant upon the member becoming of ‘unsound mind’: see, eg, Constitution Act 1855 (Tas) ss 13, 24.
 Leader of the Protectionist Government in the Senate and later inaugural High Court Justice.
 Commonwealth, Parliamentary Debates, Senate, 10 April 1902, 11 574 (Richard O’Connor).
 Ibid. Senator Stewart responded wryly that ‘I know a considerable number of madmen who take part in politics’: at 11 576 (James Stewart).
 Ibid 11 576–7 (Norman Ewing).
 Ibid 11 576 (Richard O’Connor).
 Ibid 11 578 (Henry Dobson).
 Ibid 11 577 (Norman Ewing).
 Western Australian Constitution Act 1889 (WA) s 39.
 Constitution Acts Amendment Act 1899 (WA) s 17. Subsection (2) concerned convictions. The Western Australian Constitution in its earlier form did not list persons of unsound mind as being disqualified, though it did mention persons convicted of a crime. However, positive criteria included being of full age ‘and not subject to any legal incapacity’: Constitution Act 1889 (WA) s 39. The Constitution Act Amendment Act 1893 (WA), introduced the disqualification provisions in ss 13 (Legislative Council) and 22 (Legislative Assembly). Section 39 of the Constitution Act 1889 (WA) as enacted listed only having been ‘attainted or convicted of treason, felony, or any infamous offence’ at the end of the positive criteria. The provisions were reproduced in the Electoral Act 1907 (WA) s 18 as ‘(a) is of unsound mind; or (b) is wholly dependent on relief from the State or from any charitable institution subsidised by the State, except as a patient under treatment for accident or disease in a hospital’; and then later amended to remove the relief part (b) by the Electoral Act Amendment Act (No 3) 1964 (WA) s 6(a).
 Philippa Martyr, ‘Equal under the Law? Indigenous People and the Lunacy Acts in Western Australia to 1920’ (2011) 35 University of Western Australia Law Review 317, 323–4.
 Commonwealth, Parliamentary Debates, Senate, 10 April 1902, 11 576 (Richard O’Connor).
 For example, s 2 of the Madhouses Act 1828 (Imp) 9 Geo 4, c 41, amended by the Madhouses Law Amendment Act 1829 (Imp) 10 Geo 4, c 18, s 14 extended to ‘all persons whatsoever who are lunatic, idiot, or of unsound mind’. For more detail regarding this reception in Victoria, see Carney, ‘Civil and Social Guardianship’, above n 43, 206.
 Lunacy Act 1871 (WA) s 2.
 Graeme Orr, The Law of Politics: Elections, Parties and Money in Australia (Federation Press, 2010) 59, citing R Clayton, Parker’s Law and Conduct of Elections (Charles Knight, 1996) [2.7].
 Martyr, above n 157, 320.
 See Parliamentary Electorates and Elections Act 1912 (NSW) s 20(3)(g).
 Regarding the development of the franchise regarding these groups in Australia, see Orr, above n 161, 46–9. Further, Carney speaks of the legacy in Australia (in spirit if not in form) of United Kingdom poor houses and their moral and punitive elements: Terry Carney, Social Security Law and Policy (Federation Press, 2006) 26.
 See Roach  HCA 43; (2007) 233 CLR 162.
 Kay Schriner, Lisa A Ochs and Todd G Shields, ‘The Last Suffrage Movement: Voting Rights for Persons with Cognitive and Emotional Disabilities’ (1997) 27(3) Publius 75, 75.
 In addition to the sources listed below, see also Department of the Prime Minister and Cabinet, Electoral Reform Green Paper: Strengthening Australia’s Democracy (2009). The Joint Standing Committee on Electoral Matters considered the clause in 1983, 1986, and 2012: Joint Select Committee on Electoral Reform, Parliament of Australia, First Report (1983) 105; Joint Select Committee on Electoral Reform, Parliament of Australia, The Operation During The 1984 General Election of the 1983/84 Amendments to Commonwealth Electoral Legislation (1986) 30–3; Joint Standing Committee on Electoral Matters, above n 60, 28–37 [2.63]–[2.94].
 Joint Select Committee on Electoral Reform, First Report, above n 167, 105 [5.23]–[5.25].
 Pursuant to Commonwealth Electoral Legislation Amendment Act 1983 (Cth) s 23. Victoria adopted the same wording in s 48(2)(d) of the Constitution Act 1975 (Vic) through s 5 of the Constitution Act Amendment (Electoral Legislation) Act 1984 (Vic). New South Wales also adopted the new wording pursuant to the Parliamentary Electorates and Elections Amendment Act 2006 (NSW) sch 3 item 3, amending s 25(a) of the Parliamentary Electorates and Elections Act 1912 (NSW). The original NSW provision also excluded habitual drunkards: at s 20(3)(g).
 See especially Banks v Goodfellow (1870) LR 5 QB 549; Gibbons v Wright  HCA 17; (1954) 91 CLR 423, 438 (Dixon CJ, Kitto and Taylor JJ).
 Commonwealth, Parliamentary Debates, House of Representatives, 9 November 1983, 2513 (Steele Hall); Commonwealth, Parliamentary Debates, House of Representatives, 10 November 1983, 2604 (Steele Hall).
 Commonwealth, Parliamentary Debates, Senate, 30 November 1983, 3075 (Michael Macklin).
 Commonwealth, Parliamentary Debates, Senate, 1 December 1983, 3142 (Michael Macklin).
 Ibid 3142–3 (Gareth Evans).
 Electoral Act s 116(4), as inserted by Electoral and Referendum Amendment Act 1989 (Cth) s 41.
 Joint Select Committee on Electoral Reform, 1983/84 Amendments to Commonwealth Electoral Legislation, above n 167, 30 [3.38].
 Ibid 32 [3.46].
 Human Rights Law Centre, above n 56, 2 .
 Committee on the Rights of Persons with Disabilities, Concluding Observations on the Initial Report of Australia, Adopted by the Committee at its Tenth Session, UN Doc CRPD/C/AUS/CO/1 (21 October 2013) .
 Ibid .
 Electoral and Referendum Amendment (Improving Electoral Procedure) Bill 2012 (Cth) sch 3 item 3.
 Electoral and Referendum Amendment (Improving Electoral Procedure) Bill 2012 (Cth) sch 3 item 4.
 Explanatory Memorandum, Electoral and Referendum Amendment (Improving Electoral Procedure) Bill 2012 (Cth) 3. See Freedom of Information Act 1982 (Cth) s 47F(7). Indeed the explanatory memorandum explains that these categories of profession were adopted from the provisions in the Freedom of Information Act 1982 (Cth) on personal privacy, rather than as a result of reflection on who would be most qualified to make such an assessment of capacity to vote.
 Supplementary Explanatory Memorandum, Electoral and Referendum Amendment (Improving Electoral Procedure) Bill 2012 (Cth), 2.
 ALRC Report, above n 2.
 Australian Law Reform Commission, Equality, Capacity and Disability in Commonwealth Laws, ALRC Discussion Paper No 81 (2014) 210–11 [9.21]–[9.23] (‘ALRC Discussion Paper’), quoting Roach  HCA 43; (2007) 233 CLR 162, 200  (Gummow, Kirby and Crennan JJ).
 ALRC Discussion Paper, above n 188, 210–11 [9.23].
 ALRC Report, above n 2, 262.
 Ibid 263.
 Ibid 264.
 Committee on the Rights of Persons with Disabilities, Views: Communication No 4/2011, 10th sess, UN Doc CRPD/C/10/D/4/2011 (16 October 2013) 12 [9.4] (‘Bujdosó v Hungary’).
 ALRC Report, above n 2, 264.
 Ibid 263.
 Ibid 265.
 Ibid 266.
 Ibid 266 n 25.
 Ibid 268.
 ALRC Discussion Paper, above n 188, 54 [3.3].
 ALRC Report, above n 2, 270.
 Electoral Act ss 200DL, 234. See also: at s 234A. Section 200DL uses the term ‘disabled’, but s 234 seems to rule out mental disability (but includes being illiterate). The person can choose anyone to provide this assistance.
 There were 3391 mobile polling locations for the 2013 federal election: Australian Electoral Commission, 2013 Federal Election Downloads and Statistics (1 July 2015) <http://www.aec.gov.au/elections/
 ALRC Report, above n 2, 269.
 Though the report mentions technology as a means of preserving the secret ballot: ibid 272.
 Vicki C Jackson, ‘Comparative Constitutional Law: Methodologies’ in Michel Rosenfeld and András Sajó (eds), The Oxford Handbook of Comparative Constitutional Law (Oxford University Press, 2012) 54, 70.
 Ibid 70–1.
 Ibid 71–2.
 Ibid 60.
 The figures for Australia are presented elsewhere in this article. In Japan, the number of persons
subject to full guardianship orders is over 135 000, a figure growing with an ageing population: ‘Disenfranchisement through Guardianship Unconstitutional’ above n 66, 6.
 Orr, above n 161, 59.
 See McGinty v Western Australia  HCA 48; (1996) 186 CLR 140; Lange v Australian Broadcasting Corporation  HCA 25; (1997) 189 CLR 520; Roach  HCA 43; (2007) 233 CLR 162; Rowe v Electoral Commissioner (2010) 243 CLR 1 (‘Rowe’).
 See Schenck v United States,  USSC 64; 249 US 47, 52 (Holmes J) (1919).
 Roach  HCA 43; (2007) 233 CLR 162, 199  (Gummow, Kirby and Crennan JJ).
  HCA 43; (2007) 233 CLR 162.
 (2010) 243 CLR 1.
 See, eg, Betfair Pty Ltd v Western Australia  HCA 11; (2008) 234 CLR 418.
 Savery, above n 54, 299.
 Appeals may be made to the Australian Electoral Commission and subsequently to the Administrative Appeals Tribunal: Electoral Act pt X. In NSW and Victoria where a similar provision has been adopted, the excluded voter may make an application for review to the NSW Local Court (Parliamentary Electorates and Elections Act 1912 (NSW) s 36) or the Victorian Civil and Administrative Tribunal (Electoral Act 2002 (Vic) s 42). The Federal Court has jurisdiction over matters of law under the Administrative Decisions (Judicial Review) Act 1977 (Cth).
 In its review of the amendment that introduced the requirement for a medical practitioner to certify that the voter falls within the operation of the provision, the Joint Select Committee on Electoral Reform opined that:
[The provision] requires that the elector be of unsound mind and, as a consequence incapable of understanding the nature and significance of voting. A person of unsound mind, yet capable, despite his mental condition of understanding the nature and significance of enrolment and voting is still entitled to the franchise as is, a person who is incapable of understanding the nature and significance of enrolment and voting for some reason other than unsoundness of mind.
Joint Selection Committee on Electoral Reform, 1983/84 Amendments to Commonwealth Electoral Legislation, above n 167, 31 [3.39]. The Australian Electoral Commission’s recommendation was for a form of certification that provided that a ‘medical condition precludes [the person] from involvement in the electoral process’ at 30 [3.38].
 Joint Standing Committee on Electoral Matters, above n 60, 29 (table 2.1).
 Electoral Act 1993 (NZ) s 80(1)(c).
 Orr, above n 161, 60.
 Schriner, Ochs and Shields, above n 166, 95.
 McGinty v Western Australia  HCA 48; (1996) 186 CLR 140, 221–2 (Gaudron J).
 ALRC Report, above n 2, 268.
 See, eg, Guardianship and Management of Property Act 1991 (ACT) s 7B(a); Guardianship and Administration Act 1990 (WA) s 45(3)(a).
 Hence the opening of pre-voting at or around the close of nominations such as in the Electoral Act s 200D.
 Carney, ‘Supported Decision-Making’, above n 33, 53.
 Adam Winkler, ‘Expressive Voting’ (1993) 68 New York University Law Review 330, 368.