Barriers to inclusion embedded in the Disability Discrimination Act 1992 (Cth)

How the DDA Works

What follows in this section is a simple outline of the way DDA works, as informed by thirty years of jurisprudence. This outline is necessary to contextualise the nuanced judicial ‘interpretation’ of the Act which has had a bold impact on the education rights and opportunities of many students with disability. The DDA prohibits discrimination across a range of ‘protected areas’, such as work, the provision of goods and services, accommodation, sport and, relevantly, education (section 22).

Key Definitions: educational authority and disability

‘Educational authorities’ are prohibited from discriminating against students on the ground of their ‘disability’ (DDA, section 22). Section 22 provides a non-exhaustive list of potentially discriminatory conduct, most notably, perhaps, refusal to enrol and exclusion. The terms ‘educational authority’ and ‘disability’ are defined very broadly in the DDA (section 4). An educational authority is ‘a body or person administering an educational institution’ and educational institution is further defined as ‘a school, college, university or other institution at which education or training is provided’. ‘Disability’ is defined to cover physical, intellectual, sensory and psychiatric conditions. The definition explicitly covers learning disabilities. Moreover, the definition covers temporary conditions such as illnesses and injuries. In 2009, the section 4 definition was amended to clarify that a disability also includes ‘behaviour that is a symptom or manifestation of the disability’.

Discrimination

The DDA prohibits both direct and indirect discrimination. While these causes of action are mutually exclusive (Australian Medical Council v Wilson (1996)), the circumstances of the treatment of a person with disability may allow both varieties of discrimination to be pleaded. The case Minns v New South Wales [2002], for example, concerned a student with Attention Deficit Hyperactivity Disorder who was excluded because he could not comply with his mainstream school’s behaviour code. Both direct discrimination, arising from the exclusion, and indirect discrimination, arising from the imposition upon him of a condition that he complies with the behaviour code, were pleaded.

The burden of proof of discrimination falls upon the complainant (Waters v Public Transport Corporation (1991) 173 CLR 349, 393). Various ‘elements’ must be proved to satisfy proof of either variety of discrimination. For each of these elements, associated ‘tests’ have been developed by the courts. It will be seen that the ways in which some of these tests have been articulated has eroded the protective power of the DDA.

Direct discrimination (DDA section 5(1)) is different and ‘less favourable’ treatment of a person with disability ‘on the ground’ of their disability. The treatment of the person with disability is compared with the treatment of a person without disability (a ‘comparator’) in ‘circumstances that are not materially different’ to work out whether it is ‘less favourable’. The reason for the treatment is examined to establish whether it is causally related to the person’s disability. Direct discrimination might arise, for example, if a student with disability is refused enrolment at a school because of their disability, while a student without disability is enrolled. In Finney v Hills Grammar School [1999], a student who used a wheelchair was denied enrolment in the kindergarten year at an independent school. Because other students without her disability were offered enrolment, direct discrimination could be proved and was, in fact, admitted by the respondent school which sought to rely instead, and unsuccessfully, on the ‘unjustifiable hardship’ exemption from unlawful discrimination (See sections 3 and 7.1 of the judgement; exemptions are addressed, below).

Indirect discrimination is less overt. It arises when treating people in the same manner has a discriminatory impact on a person with disability. Indirect discrimination (DDA section 6(1)) arises from the imposition of an unreasonable ‘condition’ on a person with disability. Until 2009, it was necessary to prove that the person with disability could not ‘comply’ with the condition, while a ‘substantially higher proportion’ of people without the relevant disability could comply. The elements of indirect discrimination in the DDA were amended, however, to shift the focus of the enquiry from a sometimes highly technical statistical analysis of who could and could not comply with a potentially discriminatory condition, to a more meaningful evaluation of the impact of the condition on the person with disability (Disability Discrimination and Other Human Rights Legislation Amendment Act, 2009 (Cth)). Since 2009, it must be proved instead that the person with disability cannot comply, and the condition has the effect of disadvantaging them. In discrimination law matters, disadvantage may be accepted as a matter of ‘judicial notice’ – a kind of judicial ‘common knowledge’ – obviating the need for specific evidence of disadvantage (see, for example, Hickie v Hunt and Hunt (1998), a sex discrimination case). If the respondent can prove that the condition is ‘reasonable’, however, the imposition of the condition will be lawful. An example of indirect discrimination also arises in the context of mobility impairment. When a venue is accessible only by steps, a condition is inferred that to access the venue a person must be able to climb steps. A person who uses a wheelchair cannot comply with that condition, and the condition disadvantages them because they cannot enjoy the benefits that might flow from accessing the venue. See, for example, Kinsela v Queensland University of Technology [1997] where it was held that the imposition of a condition that graduates must climb steps to a stage to receive the degree certificates amounted to unlawful discrimination against a student who used a wheelchair.

Reasonable adjustment

In 2009, the DDA was also amended to extend the scope of direct and indirect discrimination by providing that a failure to make ‘reasonable adjustment’ for a person with disability may also amount to direct or indirect discrimination (Disability Discrimination and Other Human Rights Legislation Amendment Act 2009 (Cth)). School-based examples of such a failure might be expected to arise in the context of assessment. If an appropriate extension of time available to complete an assessment item is not provided and a student fails, and is excluded as a result, direct discrimination may be asserted. The imposition of rigid ‘conditions’ on assessment items, unadjusted for disability, might amount to indirect discrimination. The scope of reasonable adjustment was considered in an assessment context in Sklavos v The Australasian College of Sklavos v Australasian College of Dermatologists (2017) (hereafter Sklavos (2017)), which is considered in detail, below.

Disability Standards

Obligations regarding reasonable adjustment in the protected area of education are expanded upon in the Disability Standards for Education, 2005 (Cth) (DSE). The DSE are subordinate legislation authorised by DDA section 31 and were intended ‘to clarify, and make more explicit, the obligations of education and training service providers under the DDA and the rights of people with disabilities in relation to education and training’ (DSE Explanatory Statement, 2005, p 1). Compliance with the DSE excuses education institutions from the scope of Part 2 of the DDA, which, in section 22, covers the prohibition of discrimination in the education area (DDA, section 34). Breach of the DSE is unlawful (DDA, section 32), but no penalty is specified in the legislation. Instead, breach of the DSE is indicative of non-compliance with the DDA and allows the pleading of discrimination under the DDA. The key obligation under the DSE, reasonable adjustment, is required in key areas of education, so that students with disability are educated ‘on the same basis’ as students without disability (clauses 4.2, 5.2, 6.2, and 7.2). Consultation with the student, and, where appropriate, their parents or guardians, is also mandated, providing an opportunity for the student’s views on reasonable adjustment to be presented (clause 3.5). While detailed consideration of the DSE is beyond the scope of this article, they are nevertheless relevant to the success or failure of the DDA as it operates to protect students with disability (see Dickson, 2014 for analysis of the DSE).

Exemptions

The DDA provides for exemptions which, if proved by the respondent, may excuse prima facie discriminatory conduct (see, for example, Division V). For example, schools may discriminate in order to comply with a court (DDA section 47) or to prevent the spread of an infectious disease (DDA section 48, Beattie v Maroochy Shire Council [1996]). The most notorious of these exemptions is ‘unjustifiable hardship’: an educational institution may ‘discriminate against another person on the ground of a disability of the other person if avoiding the discrimination would impose an unjustifiable hardship on the discriminator’ (DDA section 29A). The unjustifiable hardship exemption is also a limit on ‘reasonable adjustment’ as mandated by the DSE: an educational authority ‘must comply with the Standards to the maximum extent not involving unjustifiable hardship’ (section 10.2). The burden of proof of an exemption lies on the respondent (see Cooper v Human Rights and Equal Opportunity Commission (1999) p. 492).

State and territory anti-discrimination legislation

In any explanation of the workings of the DDA, it should also be noted that the DDA is not the only Australian legislation designed to protect students with disability. Each state and territory also has a generic anti-discrimination act which prohibits discrimination by educational institutions on the ground of disability (or ‘impairment’). This duplication is expressly tolerated by the Commonwealth (DDA section 13(3)). As such, the DDA does not carry the whole burden of eliminating education discrimination and promoting education rights. Some litigants may prefer to rely on state legislation, typically because matters are resolved before tribunals not courts (though appeals to courts may be possible). When a matter goes to a tribunal, proceedings may be less formal and less financial risk may be involved for the complainant. Because DDA cases are resolved by Courts, currently the Federal Circuit and Family Court, or the Federal Court, a body of ‘precedent’ has built up, however, which affects not only the application of the DDA, but also the application of the state acts which define discrimination in the same or very similar terms. In this way, the DDA exerts an influence over the resolution of disputes brought under State Acts too. Detailed examination of that effect on state law, however, is beyond the scope of this article.

Barriers to Inclusion Constructed from the Terms of the DDA

The fact that cases set legal precedents is one reason, perhaps, schools who have been sued fight so hard to defend claims of disability discrimination. But ‘precedents’ in a more generic sense are relevant, too, in explaining why schools put up such a fight. Even in cases where the law is not in dispute, if the application of the law to facts results in a finding of discrimination, fairness may demand that other students whose circumstances are similar to those of the successful complainant must be similarly accommodated. For each student who wins, then, many others will stand behind them in a queue for assistance.

Barriers to inclusion have been constructed from the definitions of what we may refer to as ‘classic’ direct discrimination and indirect discrimination, the varieties of discrimination prohibited by the DDA since its inception. The more recently added variety, discrimination flowing from a failure to make reasonable adjustment, has also been interpreted in an obstructive manner.

Direct Discrimination

As already noted, proof of direct discrimination in DDA section 5(1), less favourable treatment, requires a comparison between the treatment of the student with disability and a ‘comparator’ student in ‘circumstances that are not materially different’. Moreover, discrimination must be ‘on the ground of disability’, caused by the disability. The comparator and causation issues are particularly difficult in cases where an underlying disability manifests as ‘problem behaviour’, violence to others or disruption to learning.

Three controversial ‘problem behaviour’ school cases were brought under Queensland legislation during the 1990s (L v Minister for Education for the State of Queensland [1995] (‘L [1995]’), P v Director-General, Department of Education [1995] (‘P [1995]’) and K v N School [1996]), while the infamous Purvis case was the first such direct discrimination in education case to be litigated under the DDA (Purvis v State of New South Wales (2003) (‘Purvis (2003)’) In each of the Queensland cases, where a similar definition to the definition of direct discrimination in the DDA was applied (Anti-Discrimination Act 1991 (Qld), section 10), the intimate relationship between the impairment and the behaviour it caused was acknowledged. The comparator chosen in the Queensland cases was a student without the impairment and without the behaviour: an ‘ordinary’ student with ‘ordinary’ behaviour. Such a comparator student, of course, would not have been excluded, and in each of the three cases there was a prima facie finding of direct discrimination. Nevertheless, the unjustifiable hardship exemption (see, below) was relied on to find that the discrimination was lawful. While ‘unjustifiable hardship’ was proved and the Queensland cases were ultimately lost, the approach taken in them at least compelled a finding of discrimination – an upfront acknowledgement that the student had been treated less favourably. A similar approach could not be taken by the High Court in Purvis (2003) because at the time it was litigated the unjustifiable hardship exemption was not available post enrolment. It could be used to refuse enrolment, but not to exclude.

In Purvis (2003), a majority of judges of the High Court of Australia construed section 5(1) of the DDA very narrowly to decide that, on the facts, there had been no unlawful discrimination. The case concerned Daniel Hoggan, who had recently started Year 7 at a mainstream high school. As a baby, Daniel had developed encephalitis which resulted in a brain injury. The brain injury manifested, relevantly, as disinhibited behaviour. Because of his disinhibited behaviour, Daniel lashed out when upset and was suspended several times for minor acts of violence against staff and other students at the school, and against property at the school. When the violence did not stop, Daniel’s guardian, Mr Purvis, was notified that Daniel would be excluded from his mainstream school and enrolled instead at a ‘special’ school. Purvis took action on Daniel’s behalf, alleging that the exclusion would amount to direct discrimination on the ground of disability. In many education discrimination cases, parents or guardians will initiate legal action on behalf of the affected student because the student does not have the legal capacity. Usually, however, the case name will include the student’s name as complainant. Initially, the Purvis case was heard by the HREOC which found in Daniel’s favour. The matter was then appealed to the Federal, Full Federal and High Courts.

It was held by the majority judges of the High Court in Purvis (2003) that, even though a disability included its behavioural manifestations, ‘violence’, in Daniel’s case, (Chief Justice Gleeson, p. 100, Justices Gummow, Heydon and Hayne, p. 157), the comparator for proof of less favourable treatment was to be given the same ‘circumstances’, the same ‘problem behaviour’, as the complainant. As such a comparator, in Daniel’s case, would also be ‘violent’, they would also be excluded: there was no less favourable treatment. Moreover, the majority judgesheld that behavioural manifestations could be treated separately from the underlying impairment which triggered it, to determine the ‘true basis’, the cause, of the decision to exclude. The true basis, in Daniel’s case, was not the ‘disability’, but, to quote Chief Justice Gleeson, ‘the violent conduct…and…[the school principal’s] concern for the safety of other pupils and staff members’ (p. 102). As such, Daniel’s treatment was not ‘on the ground of’, caused by, his disability.

The Purvis (2003) case was shocking at the time it was decided in that it allowed the separation of the manifestations of Daniel’s impairment from the underlying impairment itself for the purpose of working out whether discrimination had occurred. Such a separation disregarded the causal relationship between the underlying brain impairment and its behavioural effects. Daniel could not control his behaviour. A comparator arguably could. The decision to exclude was based on behaviour inextricably linked to Daniel’s disability.

Much academic commentary was excited by the decision (see, for example, Edwards, 2004; Rattigan, 2004; Campbell, 2007; Dickson, 2005). But though the commentary has slowed, and the shock has faded, the case is still in the background – in the bedrock – of disability discrimination law in Australia making it impossible to succeed in a direct discrimination case except in the baldest, most ignorant cases. Smart lawyers, perhaps, will almost always be able to massage the facts to provide a plausible, lawful, ‘true basis’ for exclusionary treatment.

There is an inference to be drawn that the majority judges of the High Court manipulated the definition of direct discrimination because they wanted to find in favour of the school. Justices Gummow, Heydon and Hayne, in their joint judgement, dismissed international treaty obligations which would have supported a finding for Daniel, calling them as ‘aspirational’ (pp. 153, 156). Chief Justice Gleeson found that the Purvis (2003) case concerned a clash between competing rights:

The present case illustrates that rights, recognised by international norms or by domestic law, may conflict. In construing the Act, there is no warrant for an assumption that, in seeking to protect the rights of disabled pupils, Parliament intended to disregard Australia’s obligations to protect the rights of other pupils (p. 98).

The majority judges expressed a concern that an interpretation of the DDA which compelled schools to tolerate ‘criminal’ behaviour, to include a person who attempted ‘to burn a school down’ or ‘kill someone’ (Chief Justice Gleeson, p. 100) could not, in turn, be tolerated (Justices Gummow, Heydon and Hayne, p. 162, Justice Callinan, pp. 173–4).

The risks posed by violent behaviour should not be dismissed – schools do owe obligations to all their students and their staff. In Purvis (2003), two majority (Chief Justice Gleeson, p. 99) and Justice Callinan p. 164) and the two minority judges (Justices McHugh and Kirby, p. 94) provided some commentary on the right to inclusion in mainstream schools. All agreed that a limit on any such right arose where the inclusion impinged on the safety of other students and staff. Three of the four implied that a further limit may arise when educational opportunities of other students are adversely affected. It should not be forgotten, however, that at first instance Commissioner Innes of HREOC found that more could have been done by the school to mitigate the effect of Daniel’s disinhibited behaviour. Justices McHugh and Kirby in the High Court agreed and would also have found for Daniel, on a similar basis (p. 176). On this analysis, if more support had been rendered to Daniel and to his teachers, then exclusion – direct discrimination – could have been avoided because the behaviour triggering the exclusion would have been mitigated to an acceptable level.

It appears that in 2009, the Commonwealth Parliament acknowledged that the law as stated in the Purvis (2003) decision was flawed, and three things were done to attempt to shift the balance in favour of students with disability (Disability Discrimination and Other Human Rights Legislation Amendment Act 2009 (Cth)). First, the definition of disability was amended to put it ‘beyond doubt’ that a disability included its manifestations. Secondly, the unjustifiable hardship exemption was made available post enrolment. This meant that schools could give a student with ‘problem behaviour’ a ‘chance’ and know that if the behaviour proved ‘risky’ that the student could be excluded. Thirdly, although the DSE had been in force since 2005, and obliged the making of reasonable adjustment, such an obligation was inserted in the DDA itself.

Reasonable adjustment and unjustifiable hardship will be considered below. The first two amendments, though, were too late to save the utility of direct discrimination. As noted above, the majority judges in Purvis (2003) had accepted that a disability included its manifestations but had still allowed a ‘carving off’ of those manifestations for the purpose of constructing the comparator and determining causation. Courts still apply the Purvis (2003) comparator and causation tests, despite the definition amendment. The test has been applied to defeat allegations of direct discrimination in a string of ‘problem behaviour’ DDA cases in the last decade: Walker [2011], Abela v State of Victoria [2013] (Abela [2013]), Kiefel v State of Victoria [2013] (Kiefel [2013]), Burns v Director General of the Department of Education [2015] (hereafter Burns [2015]), Connor v State of Queensland [2020] (hereafter Connor [2020])). Even state courts apply Purvis (2003) to school cases. Queensland cases illustrate how far the influence has reached. Despite the approach to direct discrimination adopted in the Queensland ‘behaviour cases’, L, K and P, and the existence in the Queensland legislation of a provision that discrimination on the basis of a protected attribute encompasses discrimination on the basis of a ‘characteristic’ of that attribute which in other cases have allowed courts to avoid following Purvis (2003) (see Anti-Discrimination Act 1991 (Qld) section 8; Woodforth v State of Queensland [2017]), in disability discrimination in education cases involving ‘problem behaviour’, Purvis (2003) is still followed as ‘binding’ authority on the approach to be taken to comparator and causation (see, recently, BB v State of Queensland [2020], para. [16]).

Indirect Discrimination

In many cases asserting indirect discrimination, complainants have struggled with articulating the ‘condition’ they allege causes disadvantage, indicating the complexity of the legislation and making success less likely (see, for example, Walker [2011] and Sievwright [2012]). In Walker [2011], where indirect discrimination was pleaded unsuccessfully, Justice Tracey said.

A respondent is entitled to know what requirements or conditions it is said to have imposed on an applicant. In many cases this will not be difficult because the applicant will be relying on some written rule or prerequisite which has been formulated by the respondent. In a case such as the present, however, where an applicant seeks to rely on inaction by an educational authority, clarity of the claim is essential (para. [184]).

It is fair to say, however, that the most difficult barrier to inclusion constructed from the indirect discrimination provisions of the DDA is located in the reasonableness exception. It is convenient to consider the reasonableness barrier alongside the unjustifiable hardship exemption barrier as they both operate defensively, may be pleaded together, and similar considerations are relevant to both: see ‘reasonableness and unjustifiable hardship’, below.

Reasonable adjustment

Notions of reasonableness also intrude in the context of reasonable adjustment. The 2009 amendments to the DDA inserted new subsections into section 5, direct discrimination (section 5(2)), and section 6 indirect discrimination (section 6(2)) which were intended to oblige reasonable adjustment. In his second reading speech in support of the amendments, then Attorney General Robert McLelland said that ‘[T]he key amendments to the Disability Discrimination Act will introduce an explicit and positive duty to make reasonable adjustments for people with disability’ (Parliamentary Debates, 2008). The drafting of the new subsections is awkward, however, and both have been interpreted in the Full Federal Court education discrimination case, Sklavos (2017), in a way which defeats the purpose of the amendments.

Dr. Sklavos was hoping to be admitted to the Australasian College of Dermatologists, but because of a psychiatric condition, he could not complete the required examinations. He sought ‘reasonable adjustment’ to the assessment regime so that he could demonstrate competency, but the College insisted that the examination requirement was reasonable and refused to provide an alternative. Sklavos argued that a failure to make reasonable adjustment had resulted in direct discrimination against him, in breach of section 5(2) of the DDA, as he was excluded from the College. He argued that the failure to make reasonable adjustment had resulted in indirect discrimination in breach of section 6(2) in that a disadvantageous condition was imposed upon him that he completes the examination. He argued that the imposition of this disadvantageous condition also amounted to ‘ordinary’ indirect discrimination in breach of section 6(1). Finally, he argued that the College had breached its consultation and reasonable adjustment obligations under the DSE. He failed on all counts.

In respect of the direct discrimination claim, the Court focused on the wording of para. (b) of the new subsection:

(b) the failure to make the reasonable adjustments has, or would have, the effect that the aggrieved person is, because of the disability, treated less favourably than a person without the disability would be treated in circumstances that are not materially different.

Influenced by the approach taken to causation in Purvis (2003), the Court found that the placement of the words ‘because of’ meant that it was necessary to prove a causal link between the failure to make reasonable adjustment and the complainant’s disability (Justice Bromberg, pp. 255–256,; Justices Griffiths (p. 290) and Bromwich (p. 303) agree). To illustrate from the facts of the case, Sklavos would have had to prove that the failure to adjust the exam was because of his psychiatric condition. Of course, the failure to make the reasonable adjustment had nothing to do with his psychiatric condition – it was motivated by the College’s insistence that the examination was reasonable and by concern that developing a new assessment regime for Dr. Sklavos would be not only an onerous but also a speculative task because there was doubt that he would be able to complete any kind of assessment. Arguably, failure to make reasonable adjustment will almost always be caused by considerations of cost and inconvenience and almost never by a complainant’s disability. Indeed, it is difficult to imagine a set of facts where a failure to make reasonable adjustment would be because of the complainant’s disability. The approach to direct discrimination in Sklavos (2017) has since been applied in the unsuccessful ‘problem behaviour’ education case, Connor [2020] (see, for example, para. [266]).

The approach taken to indirect discrimination in Sklavos (2017) is also damaging. Here the Court focussed on subsection 3 of section 6:

(3) Subsection (1) or (2) does not apply if the requirement or condition is reasonable, having regard to the circumstances of the case.

The Court found that, instead of stepping through the elements of indirect discrimination sequentially, it was legitimate, once a condition had been identified, to approach section 6 by focussing next on subsection (3) – a consideration of the reasonableness of that condition. Again, this approach can be illustrated on the facts of the Sklavos (2017) case. Because the Court accepted that the examination condition was ‘reasonable’, there was no need to consider whether there had been a failure to make reasonable adjustment (Justice Bromberg, p. 274, Justices Griffiths (p. 290) and Bromwich (p. 303) agree). On one analysis, this approach renders consideration of reasonable adjustment redundant and elides two separate issues of reasonableness – adjustment and condition – into one. On another analysis, this approach recognises that reasonable adjustment and indirect discrimination are interrelated, conceptually and practically. Both are concerned with substantive equality, with the truth that unequal treatment may sometimes be required to deliver ‘equality’. The avoidance of a discriminatory ‘condition’ will, in practice, require the making of an adjustment for a student: provision of an Auslan interpreter for a Deaf student, aide support for a student with a learning disorder, ramp access for a student with a mobility impairment, and so on. On this second analysis, it does not matter so much that the obligation to make reasonable adjustment has been rendered redundant by the Sklavos (2017) decision. While the words send a powerful signal, the actual work of delivering equality is capable of being performed by indirect discrimination.

An argument in Sklavos (2017) that obligations under the DSE had been breached were convincing for only one of the judges, Justice Bromberg. He would have referred the case back to the Federal Court for rehearing on that issue (pp. 283–284). Importantly, in what could be good news for students with disability, he suggested that general damages for breach of the DSE may be available (p. 289). This is a novel suggestion, though, as the DDA does not impose a penalty for breach of the DSE (see section 32).

The highly technical interpretation of the law in the Sklavos (2017) case has attracted academic criticism (Taylor, 2019) and calls from disability advocates (for example, People with Disability Australia, 2021) for amendment of the DDA so that it is capable of delivering on the promise of the reasonable adjustment. The amendments were meant to tear down a barrier to inclusion, not to construct ‘a new and impracticable legal hurdle’ (People with Disability Australia, 2021).

Reasonableness and unjustifiable hardship

Protections for discriminators embedded in notions of reasonableness and hardship have the clear potential to limit the social response required to impairment, and, therefore, to erode the right of a person with disability to an inclusive education. They appear, moreover, to be informed by ‘ableist’ prejudice in favour of the preservation of existing education systems established to cater for ‘normal’ students. During the second reading debate of the DDA, a member opposing the legislation said, ‘[e]very time the word “reasonable” is written, another lawyer gets a job. One can debate “reasonable” in the courts of Australia until one is blue in the face’ (Commonwealth, Parliamentary Debates, House of Representatives, 19 August 1992 (Wilson Tuckey)). Less colourfully, ‘reasonableness’ was characterised by Justice Madgwick of the Federal Court, in Clarke v Catholic Education Office, as ‘the most difficult question’ (p. 353). Similarly, the unjustifiable hardship enquiry was described by Tamberlin J, of the Federal Court, in the appeal from Finney v Hills Grammar School [1999], discussed above, as involving ‘the weighing of indeterminate and largely imponderable factors and the making of value judgments’ (Hills Grammar School v Human Rights and Equal Opportunity Commission (2000), p. 314).

The concept of ‘reasonableness’ pervades the DDA. Indeed, it has become more pervasive as the DDA been amended over time. As seen above, it is relevant to both proof of section 6, indirect discrimination, and the purported positive obligation to make reasonable adjustment. As seen, above, reasonableness is relevant, too, to reasonable adjustment as obliged by the DSE. Furthermore, an almost impossibly complex relationship has developed between reasonableness and unjustifiable hardship across the DDA and the DSE. Unjustifiable hardship is an exemption which may be relied on by a respondent to defeat a prima facie case of either direct or indirect discrimination in education (DDA, section 29A). For direct discrimination, it is the key exemption relied on to defeat liability. For indirect discrimination, reasonableness is considered before unjustifiable hardship and the implication is that even an unreasonable condition may be excused if to avoid it would cause unjustifiable hardship to the discriminator. Unjustifiable hardship also sets the limit for reasonable adjustment in the DDA: ‘an adjustment to be made by a person is a reasonable adjustment unless making the adjustment would impose an unjustifiable hardship on the person’ (section 4). For the DSE, unjustifiable hardship is expressed as an exemption from the obligation to make reasonable adjustment: a school ‘must comply with the Standards to the maximum extent not involving unjustifiable hardship’ (section 10.2(3)). As with indirect discrimination, then, reasonableness is considered before unjustifiable hardship, and even a reasonable adjustment could potentially be avoided if to make it would impose unjustifiable hardship.

What can be discerned from the confusion is a conclusion that concepts of reasonableness and hardship, operating independently or together, have worked to defeat the majority of those discrimination claims in which less favourable treatment or the imposition of a discriminatory condition has been proved. However, as reasonableness has become more influential in the way cases are decided under the DDA, unjustifiable hardship has, arguably, become less influential. This is because, even though hardship may have been pleaded to exempt discrimination, reasoning in the cases seldom gets to the point where hardship needs to be considered by a court. The Scarlett Finney case, discussed above (Hills Grammar School (2000), is a rarity in this respect and that case sent a clear message that schools should not catastrophise the impact of enrolling a student with physical impairment. As such, very few DDA physical impairment cases have found their way to court since that case. In the DDA cases that do go to court – usually ‘problem behaviour’ cases – that court will apply Purvis (2003) and find there is no direct discrimination, so consideration of hardship is redundant. A court will find that any alleged indirect discrimination is ‘reasonable’, so consideration of hardship is redundant. Further, it may be anticipated that courts will apply Sklavos (2017) and find, similarly, that consideration of reasonable adjustment is redundant, and by implication so too is consideration of the hardship limit on reasonableness.

Case law (Secretary, Department of Foreign Affairs and Trade v Styles (1989), p. 263) and relevant sections of the DDA (section 11) and DSE (sections 3.4(2) and 10 (2) (3)), suggest that the relevant circumstances for determination of reasonableness and hardship overlap. Indeed, because the law requires that ‘all relevant circumstances’ must be considered for proof of either, either enquiry may cover the same territory. DDA section 11 and DSE section 3.4(2) both contemplate, for example, a balancing consideration of the effect of the disability on the complainant, and the effect on them and others in the school community should potentially discriminatory treatment be allowed to continue, and the cost of avoiding the discriminatory treatment. While the legislation might contemplate a nuanced balancing of competing relevant circumstances, the ‘lost’ education discrimination cases suggest two circumstances have invariably proved to be determinative: impact of the disability on the safety or learning of others; and excessive cost to the discriminator of avoiding the discrimination.

The case Minns v State of New South Wales [2002] (hereafter Minns [2002]), already mentioned above, illustrates the barrier that is raised when it can be demonstrated that inclusion of a student with disability challenges existing systems by affecting the safety or disrupting the learning of others. It was suggested by Gleeson CJ in Purvis (2003) that Daniel’s case was not framed as one of indirect discrimination in order to avoid the requirement of proving that a discriminatory term was ‘not reasonable’ (p. 98). Minns [2002] was the first ‘problem behaviour’ case pleaded as indirect discrimination under the DDA. Direct discrimination was pleaded too, but rejected by the Court via application of the comparator test ultimately adopted by the High Court in Purvis (2003). That test had already been formulated and applied by the Full Federal Court in an interim appeal in the Purvis litigation before the matter reached the High Court (Purvis v New South Wales (2002)), and before Minns [2002] was decided. The complainant, Ryan Minns, who had been diagnosed with Attention Deficit Hyperactivity Disorder, was facing exclusion from his mainstream school after a series of incidents of challenging behaviour. Federal Magistrate Raphael explicitly drew attention to the similarities with the Purvis (2003) case, commenting that it concerned ‘violent and anti-social behaviour very similar to that exhibited by Ryan Minns’ (para. [191]) and, like the Majority in Purvis (2003), his Honour was concerned about the impact of that behaviour. The complainant’s case failed in Minns [2002] because the condition imposed, compliance with the conduct required by the school discipline policy (para. [247]), was held to be ‘reasonable’ (para.s [256]-[257].

Federal Magistrate Raphael emphasised that it was reasonable that schools have and enforce codes of conduct: such codes were necessary to enable ‘all students to benefit from the educational opportunities offered and the requirement to allow this to happen in a safe environment’ (para. [247]). The critical issue, however, was the reasonableness of the condition that Ryan comply with the code, and not the reasonableness of the code itself. Ryan alleged that his disability made it ‘impossible for him to behave in a manner compliant with the discipline policy’ (para. [250]. A ‘relevant circumstance’ which weighed in favour of the school was that it had administered the discipline policy flexibly to accommodate Ryan’s disability. The persuasive impact of Ryan’s behaviour upon his reasoning as to reasonableness is evident in the words of Federal Magistrate Raphael:

I am of the view that the requirement that was placed upon Ryan to comply with each of the school’s disciplinary policies as modified was reasonable in all the circumstances. The classes in which Ryan was placed would be unable to function if he could not be removed for disruptive behaviour. The students could not achieve their potential if most of the teachers’ time was taken up with handling Ryan. The playgrounds would not be safe if Ryan was allowed free rein for his aggressive actions. Therefore the claim for indirect discrimination must fail in the manner in which it is put (para. [263]).

Thus, in determining the reasonableness issue against Ryan Minns, the benefit to Ryan of inclusion was trumped by the potential detriment to others in the school community. The language employed by Federal Magistrate Raphael is clearly reminiscent of the language of the Majority in the High Court in Purvis (2003) who were so concerned about the risks to others in Daniel’s school community should his enrolment be maintained. Similar ‘problem behaviour’ cases decided since Minns [2002] framed as indirect discrimination have all failed: Walker [2011], Abela [2013], Kiefel [2013], Burns [2015], Connor [2020].

Courts have also demonstrated a willingness to find that excessive cost to a school or school system may render an adjustment unreasonable or amount to unjustifiable hardship. In most of the failed indirect discrimination cases referred to above, increased aide support had been sought by the complainants on the basis that such support would not only facilitate their learning but also mitigate the effect on others of their behaviour. A condition had allegedly been imposed, and it was argued in those cases, along these lines: ‘the complainant must receive their education without the assistance of a full-time aide’. Aide support is expensive, and a barrier to inclusion is constructed from the reluctance of courts to order its provision. The case Sievwright [2012] is not a ‘problem behaviour’ case but one, nevertheless, where expensive support was sought for the complainant, Jade. Jade had multiple disabilities and sought one-on-one aide assistance in the classroom, among other supports. In the Queensland cases, L [1995] and P [1995], discussed above, the hearing tribunals found that excessive cost to the state could not be raised to defeat an obligation to provide support. In L [1995], for example, it was found that should extra support be allocated, ‘the financial burden on the Respondent [the State] is insignificant, having regard to its resources’ (n.p.). In P [1995], it was decided that providing the extra support may simply have ‘required some re-allocation of resources and even the raising of some extra revenue’ (n.p.). Of course, as noted above, those cases still failed, rendering the finding with regard to the cost of support redundant. A different approach to ‘financial burden’ was taken, almost twenty years later, by the Federal Court in Sievwright [2012]. Justice Marshall emphasised, citing Purvis (2003), that ‘[t]he obligations of the State in respect of individual children must be considered alongside the wider legal responsibilities which teachers and administrators owe to all students’ and that it was appropriate for the State to implement ‘objective eligibility criteria’ for access to funding (para. [207]). The ‘objective criterion’ in respect of Jade was that she must have an IQ lower than 70, indicative of intellectual impairment, to attract extra support [para. [43]). Jade’s IQ fluctuated between 68 and 80, as measured at different times by different experts, but the Court accepted that, at the operative time, she did not qualify for extra support: ‘provision of a full-time aide for Jade was beyond the financial capability of her school’ (para. [208]). Justice Marshall was impressed by evidence that ‘providing a full-time aide to all students who had a IQ in the vicinity of Jade’s… would double the current PSD [disability support] budget requirements and result in a need for the State to engage 20,000 extra staff’ (para. [209]). It appears that, unlike the decision makers in Queensland twenty years ago, the Federal Court does not view governments as having ‘deep pockets’ and proof that what Justice Marshall called ‘enormous cost’ (para. [209]) might flow from the setting of a precedent in one single case is a barrier to the ordering of support in that single case.