DECISION Complaint proved in part; consequential orders made; compensation not awarded.

CITATION Davies v State of Victoria (Victoria Police) (Anti-Discrimination) [2000] VCAT 819

On this complaint coming on for hearing and on hearing Mr Grabau, of Counsel for the Complainant and Mr Hanks QC for the Respondent, THIS TRIBUNAL MAKES THE FOLLOWING ORDERS-

1. It finds this complaint proved in that the Respondent directly discriminated against the Complainant in contravention of section 13 of the Equal Opportunity Act 1995 (the Act) by rejecting his application to join the Respondent's Police Force on the basis of the Complainant's colour vision deficiency.
2. Pursuant to section 136(a)(iii) of the Act, the Tribunal makes the following orders, with a view to redressing the loss, damage or injury suffered by the Complainant in consequence of the contravention of the Act referred to in paragraph 1-
   • (a) The Respondent must in all respects treat the Complainant's application to join the Police Force made in 1997 as if it were still on foot;
   • (b) The Respondent must assess that application without reconsidering whether the Complainant now meets any eligibility requirement which, when the application was made, the Complainant met;
   • (c) The Respondent must assess the Complainant's colour vision deficiency by means of the Farnsworth D15 colour vision test. If Mr Davies passes this test, he must be allowed to proceed to the next step of the selection process on the basis that his colour vision perception is sufficient to enable him to perform adequately the requirements of employment as a Police Constable. If the results of the test show that he is a deuteranope, his application may be rejected on the basis that he is unable to perceive one of the distinct colours that this Tribunal has found are part of the reasonable requirements of this employment. If he fails but is not a deuteranope, then simulation testing is necessary to test his actual ability to perform the requirements of this employment.
   • (d) If the Complainant fails the D15 test and is not a deuteranope, the Respondent must assess the Complainant's actual ability to perform the tasks required of a Police Constable by a test other than a colour vision test. That test referred to in this paragraph must be a simulation test in a form agreed between the parties or, in default of agreement, determined by the Tribunal;
   • (e) The tests mentioned in paragraphs (c) and (d) are to be undertaken at a time and place agreed between the parties or, in default of agreement, determined by the Tribunal;
   • (f) If the Respondent considers that the results of these tests demonstrate that the Complainant is able adequately to perform the tasks required of a Police Constable, the Respondent must deal with the application in the same ways it would deal with the application of a person who had met its colour vision standard.
   • (g) If the Respondent considers that the results of these tests demonstrate that the Complainant cannot perform adequately the tasks required of a Police Constable, the Respondent may reject the Complainant's application.
3. The Complainant's application for other orders under section 136(a) of the Act is refused.
4. To the extent that it claims indirect discrimination, the Tribunal finds the complaint not proved and dismisses it.
5. The Tribunal reserves its decision as to costs. The Complainant has applied for costs. The Respondent may apply for costs by filing and serving an application on or before 17 January 2000.
6. If the parties wish to make submissions on the question of costs they must do so as follows-

(a) the parties must file with the Tribunal and exchange written submissions on or before 27 January 2000.

(b) the parties must file with the Tribunal and exchange any submissions responding to submissions filed under paragraph (a) on or before 7 February 2000.

On 1 September 1997 Mr Davies lodged with the Equal Opportunity Commission (the Commission) a complaint under the Equal Opportunity Act 1995 (the Act) against the State of Victoria (Victoria Police). In its letter of referral to this Tribunal, the Commission states that, pursuant to section 108 of the Act, it considered the complaint non-conciliable. The reference to this provision would appear to be an error. Complaints which the Commission considers non-conciliable are dealt with by s113 of the Act, and the Complainant has a right to refer such a complaint under that section. Mr Davies then requested the referral of the complaint to VCAT. The Commission referred the complaint by letter dated 13 July 1998, received by the Victorian Civil and Administrative Tribunal (VCAT) on 31 July 1998.

The parties have filed Particulars of Complaint and Particulars of Defence. The matter came on for hearing on 1, 2 and 16 September and 8 October 1999. After the hearing of evidence, the Tribunal directed that closing submissions from the parties were to be made in writing. Those closing submissions have subsequently been filed and served. The Tribunal also required submissions to be made to it concerning the effect of a decision of the Human Rights and Equal Opportunity Commission (handed down after the hearing of this proceeding had commenced) and of a decision of the High Court (handed down after closing submissions had been filed). These further submissions have also been filed and served.

Mr Davies claims that the Respondent (the State) directly discriminated against him in a way prohibited by the Act by rejecting, on the basis of his colour vision deficiency, his application to join the Victoria Police Force. The State concedes that it rejected Mr Davies' application on the basis of his colour vision deficiency and that it would not, at the same stage of the application process and in the same or similar circumstances, have rejected the application of a person without the colour vision deficiency. The effect of this concession is that the State concedes it treated Mr Davies less favourably than, in the same or similar circumstances, it would have treated a person without Mr Davies' colour deficiency, and that the basis of this treatment was Mr Davies' colour vision deficiency. The State also concedes that a colour vision deficiency is an impairment within the meaning of the Act. The effect of s8 of the Act is that direct discrimination occurs if a person treats someone with an attribute less favourably than, in the same or similar circumstances, that person would treat someone without that attribute or with a different attribute, where a substantial basis of that treatment was the attribute. Impairment is an attribute within the meaning of the Act (ss4 and 6). The State, then, concedes that it directly discriminated against Mr Davies when it rejected his application to join the Police Force.

But the State argues that this direct discrimination was not prohibited by the Act, because certain provisions of the Act apply to take that conduct outside the scope of what the Act prohibits. The provisions on which the State relies are ss 22, 23 and 80 of the Act.

Mr Davies also claims indirect discrimination. Under s9 of the Act, when applied in the context of this case, indirect discrimination occurs if the State has imposed a requirement, condition or practice with which Mr Davies (as a person with the impairment of colour deficiency) cannot or does not comply, with which a higher proportion of people without that impairment can or do comply, and which is not reasonable.

In these reasons for decision, we do not deal in detail with the claim of indirect discrimination. The Particulars of Complaint did not clearly formulate a requirement, condition or practice that could form the basis of a claim of indirect discrimination. When counsel for Mr Davies opened Mr Davies' case, the Tribunal asked him to clarify the claim of indirect discrimination. His answer left the Tribunal uncertain as to whether the requirement, condition or practice relied on was the use by the State of colour vision testing to determine finally whether to accept or reject an applicant for the Police Force, or the application of that test to finally determine the Complainant's application to join the force, or the failure to take into account any indicators, (apart from the results of colour vision testing) to determine the Complainant's ability to perform the requirements of the employment. The closing submissions for the Complainant have not made this claim clearer. In our view, the Complainant has not formulated the requirement, condition or practice that can found a claim of indirect discrimination with sufficient clarity to enable the Tribunal to consider that claim. If the Tribunal went ahead and tried to consider that claim, it would be forced itself to formulate from the evidence some requirement, condition or practice that might found a claim of indirect discrimination. In our view, the onus of proving a claim of indirect discrimination is on the Complainant. It is for the Complainant to clearly formulate the requirement, condition or practice that founds the claim. If this is not done, it is not for the Tribunal to remedy that deficiency by itself formulating some requirement, condition or practice. This is especially so where, as in this case, the Complainant is represented by Counsel.

On this basis, the claim of indirect discrimination fails and should be dismissed. We therefore consider this case as a case of direct discrimination only.

We first set out the relevant law. We then make findings of fact. Finally, we apply the relevant law to those findings.

This is the first time when this Tribunal and its predecessors the Anti-Discrimination Tribunal and the Equal Opportunity Board have been required to consider in detail whether colour vision and standards for entry into the Police Force might contravene the Equal Opportunity Act 1995 or the Victorian predecessors to that legislation. It is the first time that the Tribunal has been required to consider in detail the meaning of ss22 and 23 of the Equal Opportunity Act 1995. Counsel for the parties have assisted us by submissions and by referring us to a large number of cases.

We first set out the provisions of the Act on which the Respondent relies.

Equal Opportunity Act 1995

22. Exception - special services or facilities

(1) An employer may discriminate against another person on the basis of impairment in any of the areas specified in section 13 or 14 if -

• (a) in order to perform the genuine and reasonable requirements of the employment -

  • (i) the other person requires or would require special services or facilities; and
  • (ii) it is not reasonable in the circumstances for those special services or facilities to be provided; or

• (b) the other person cannot or could not adequately perform the genuine and reasonable requirements of the employment even after the provision of special services or facilities.

(2) In determining whether or not a person can or could adequately perform the requirements of the employment, all relevant factors and circumstances must be considered, including -

• (a) the person's training, qualifications and experience;
• (b) the person's current performance in the employment, if applicable.

1. Exception - reasonable terms of employment

An employer may set reasonable terms or requirements of employment, or make reasonable variations to those terms or requirements, to take into account -

(a) the reasonable and genuine requirements of the employment;

(c) any special limitations that a person's impairment or physical features imposes on his or her capacity to undertake the employment;

(d) any special services or facilities that are required to enable him or her to undertake the employment or facilitate the conduct of the employment.

1. Protection of health, safety and property
   • (1) A person may discriminate against another person on the basis of impairment or physical features if the discrimination is reasonably necessary-
     • (a) to protect the health or safety of any person (including the person discriminated against) or of the public generally;

Section 12 of the Act provides that "this Act does not prohibit discrimination if an exception in Part 3 (whether or not in the same Division as a provision prohibiting the discrimination) or part 4...applies". Sections 22 and 23 appear in Division 1 of part 3 - the same Division that contains s13 which, among other things, prohibits an employer from discriminating against a person in determining who should be offered employment, or by refusing or deliberately omitting to offer that person employment. The word "exception" appears in the heading to both ss 22 and 23. Section 80 appears in Part 4. Part 4 is headed "General exceptions to and exemptions from the prohibition of discrimination".

Sections 22, 23 and 80 are exceptions to the prohibition of discrimination. They are covered by section 12. The effect of that section is that, if conduct falls within one of the exceptions, it is not prohibited by the Act. The burden of proving that an exception applies is on the Respondent (Ellis v Mt Scopus Memorial College, ADT of Vic, 29 March 1996; Colyer v State of Victoria [1999] 3 VR 759 at 774 (per Kenny JA with whom Calloway and Brooking JJA agreed). The Ellis case and the Colyer case suggest that, at least in the context of a proceeding such as a complaint proceeding where a contest between parties is concerned, the burden of proving an exception should lie on the person claiming it. This approach is consistent with what was said in the Explanatory Memorandum for the Bill that eventually became the Equal Opportunity Act 1995, and with the general rules of interpretation relating to exceptions and provisos.

These provisions do not appear in the same form in the predecessors to the Equal Opportunity Act 1995. For example, s39(da) of the Equal Opportunity Act 1984 made it not unlawful to discriminate on the basis of impairment if this was reasonably necessary to protect public health. Section 80(1)(a) of the Equal Opportunity Act 1995 is wider and covers discrimination on the basis of impairment and physical features, and also extends to discrimination reasonably necessary to protect the health or safety of any person, as well as what is reasonably necessary to protect public health or safety.

Similarly, s27B(3)(c) of the Equal Opportunity Act 1977 and s21(4)(g) of the Equal Opportunity Act 1984 were expressed in significantly different terms from ss 22 and 23 of the Equal Opportunity Act 1995. For example, rather than referring to "genuine and reasonable requirements of the employment" they refer to "the work reasonably required" of the person concerned.

Given these differences, we consider it unwise to apply to ss 22, 23 and 80 interpretations given to earlier and different provisions of Victorian Equal Opportunity legislation.

The parties urged us to rely on a number of other cases which, they submitted supported their various arguments. Sometimes, they have each used different parts of the same case in support of their arguments.

We consider that the cases cited to us do not have direct application to this case for the following reasons-

(1) They involve statutory provisions which are significantly different from the ones we are considering. For example, some deal with legislation which refers to "inherent requirements of the employment" or "genuine occupational requirements" rather than "genuine and reasonable requirements of the employment", as do ss 22 and 23 (see Qantas Airlines Limited v Christie [1998] 152 ALR 365; R v State of Victoria, HREOC, 8 August 1997; **X v Commonwealth (**HC of A, 2 December 1999); Flannery v O'Sullivan (No.2) ([1993] EOC 92-501).

(2) They concerned an issue quite different from the issue in the present case. On this basis, we have found of little assistance a number of cases relied on for the Complainant. These were cases in which the Federal Administrative Appeals Tribunal reviewed decisions of the Secretary of the Federal Department of Air Navigation which had imposed various conditions on the granting of commercial pilot's licenses and private pilot's licenses to people with colour vision deficiencies (Re Pape and Secretary, Department of Aviation, AAT, 4 August 1987; Re Dennison v Civil Aviation Authority[1989] AATA 84; , [1989] 19 ALD 607). These cases were decided in the context of legislation which permitted the granting of conditional licenses. The only question in those cases was whether some of the conditions imposed when certain kinds of pilot's licenses were granted to particular applicants with colour vision deficiencies should properly have been imposed. The cases did not concern the question of whether the applicants were entitled to the issue of an unconditional license. They are of little assistance when considering the present case in the context of the application of the Victorian Equal Opportunity Act 1995.

(3) The cases involve significantly different fact situations from the present case -

• (a) Some concerned impairments other than colour vision deficiency, such as insulin dependent diabetes (Burrows v Chief Commissioner of Police, [1994] EOC 92-654, affirmed in part on appeal [1996] EOC 92-741), myopia and other kinds of visual impairment (Flannery's case, cited above; Melvin v Morningside Community Service Inc [1996] EOC 92-830; Allender v Department of Human Services, Victoria [1997] EOC 1992-904).
• (b) Some concerned colour deficiencies more severe than that which Mr Davies has. For example, the evidence in R's case was that R suffered from extreme colour vision deficiency. A further example is Loscialpo v New South Wales Police Service (HREOC, 2 September 1999), which concerned a Complainant who was moderately protanomalous - that is, whose red colour receptors did not function efficiently. On the medical evidence in that case, Mr Loscialpo would have difficulty in distinguishing red and orange traffic lights and might fail to see red traffic lights.
• (c) Some concerned occupations other than Police Constable, such as Protective Services Officer (Crombie v Commonwealth of Australia, HREOC, 20 November 1998; Eyden v Commonwealth, HREOC, 24 February 1999), Ambulance Driving (Smith v The St John Ambulance Australia - South Australia Inc [1991] EOC 92-383, affirmed by the Full Court of the South Australian Supreme Court [1992] EOC 92-448), or as a Carriage Trimmer in a railway depot (Bugden v State Rail Authority [1991] EOC 92-360, [1992] EOC 92-434).
• (d) The evidence before the Courts and Tribunals that heard those cases was significantly different from the evidence given in this case. For example, in , the Human Rights and Equal Opportunity Commission accepted the only medical evidence before it. That evidence came from witnesses called by the Respondent, and was to the effect that colour vision diagnostic tests make use of surface colours that are representative of those used in the real world, that those tests provide reasonable simulation of the workplace environment and have excellent predictive ability in identifying those who cannot reliably recognise distant signal lights. The evidence before us in this case is significantly different. It includes extensive medical evidence challenging, and extensive medical evidence supporting, the predicative power of diagnostic colour vision tests in relation to "real world" tasks.

Nevertheless, the Tribunal has found some of the cases to which it was referred useful to illustrate the difficulties of construing provisions analogous to ss22 and 23 of the Equal Opportunity Act 1995, and the range of factors which may be considered in applying those provisions to particular facts. In particular, it has found the High Court's decision in X's case useful in its approach to the interpretation of s15(4) of the Commonwealth Disability Discrimination Act 1992 - a provision similar (although by no means identical) to s22 of the Equal Opportunity Act 1995.

The State relies on s22(1)(b). Section 22 applies to discrimination either against employees or potential employees. It covers all the conduct mentioned in ss13 and 14 - ranging from determinations about the offering of employment to dismissal of an existing employee. Section 22(1)(b) applies to discrimination where the person "cannot or could not adequately perform the genuine and reasonable requirements of the employment even after the provision of special services or facilities". In determining this, all relevant circumstances must be taken into account (s22(2)). The "special services or facilities" referred to are the ones referred to in s22(1)(a)(i) - that is, the ones that the person requires or would require in order to perform the genuine and reasonable requirements of the employment. Just as the majority judgements in X's case said that the two paragraphs in s15(4) of the Commonwealth Disability Discrimination Act could not be construed in isolation from each other, so also s22(1)(a) and (b) of the Equal Opportunity Act 1995 (Vic) cannot be construed in isolation from each other. Section 22(1)(b) covers the case where a person cannot adequately perform the reasonable and genuine requirements of the employment, even after the provision of the special services or facilities referred to in s22(1)(a). Section 22(1)(a) is, in our view, designed to cover the case where the person could perform the reasonable and genuine requirements of the employment after special services and facilities were provided, but where it would not be reasonable for the employer to provide them. Each of the paragraphs covers a different but related situation.

There are at least three questions that need to be answered in determining whether the exception in s22(1)(b) applies.

(1) What are the genuine and reasonable requirements of the employment? In our view, these are wider than the inherent or essential requirements of the employment, and whether a particular duty, ability or qualification is a genuine and reasonable requirement of employment is not to be determined by seeing whether, if that duty, ability or qualification were omitted, the employment would still be the same (Christie's case). The requirements of the employment and the requirements of the employer are not necessarily the same thing, although they will often be the same. The requirements of the employment refers to what the job or position requires to be done, as well as what is necessary to do those duties. The term covers the whole range of these requirements, and not just the "essential" ones. The requirements must be requirements "of the employment". In other words, they must relate to and derive from the employment. They must be genuine requirements. An employer cannot invent requirements which are not truly requirements of the employment. They must be reasonable requirements. As was pointed out in the context of indirect discrimination, "reasonable" is a more demanding test than one of convenience but a less demanding test than one of necessity (Secretary, Department of Foreign Affairs and Trade v Styles [1989] FCA 342; [1989] 23 FCR 251 at 263 per Bowen CJ). So, a requirement for abilities or qualifications that are grossly disproportionate to the degree of difficulty of the duties, might not constitute a reasonable requirement of the employment.

In Flannery's case (cited above), the Queensland Anti-Discrimination Tribunal, in considering a provision similar but not identical to s22(1)(b), drew a distinction between the requirements of the employment and the ability to perform them. We do not consider that, in s22(1)(b), the genuine and reasonable requirements of employment could never include requirements concerning certain abilities and skills relevant to particular duties of the employment. A requirement for a certain degree of physical or mental fitness might be one of these.

There is a suggestion in the judgement of Gummow and Haine JJ in X's case (a judgement with which Gleeson CJ and Callinan J agreed) that the inherent requirements of employment must also have regard to the circumstances in which the employment is carried on and the dangers to which an employee may be exposed or may expose others. In our view, these matters are also relevant in determining the genuine and reasonable requirements of employment for the purposes of s22(1) of the Equal Opportunity Act 1995 (Vic).

To determine what are the genuine and reasonable requirements of the employment, we should look at all the circumstances surrounding the employment, including any contract of employment, any statutory functions, and any organisational or operational requirements of the employment. (see generally Crombie's case, cited above; Jamal v Secretary, Department of Health and Ors [1988] EOC 92-234).

(2) What special services or facilities does or would the person require in order to perform the genuine and reasonable requirements of the employment?

The terms "special services or facilities" and "special services, benefits or facilities" are used in other exceptions in the Act (see for example ss32, 39 and 82). In other exceptions the word "services" is used, without the word "special" (for example s19).

In our view, "services" and "facilities" cover a wide range of measures, and would include additional assistance by way of staff support, equipment, devices or other benefits. The facilities must be "special" - that is, they must be designed or provided to help the person with the impairment perform the requirements of the employment or they must be of a kind not provided to or needed by applicants or employees without that impairment.

(3) Even with the special services or facilities, can the person adequately perform the genuine and reasonable requirements of the employment?

The employer does not have to show that the person cannot perform the genuine and reasonable requirements of the employment at all. On the other hand, performance may still be adequate even though it is not perfect. Adequate performance falls somewhere between these two extremes. In our view, the question of whether a person is able adequately to perform the genuine and reasonable requirements of the employment must be judged on those requirements as a whole. A person may still be able to perform adequately the requirements of the employment as a whole, even though he or she is unable to perform, or unable adequately to perform, a particular requirement. Whether an inability to perform a particular requirement adequately will amount to inability to perform adequately the requirements of the employment as a whole will depend on the extent of the inability, the comparative importance of the requirement to the employment as a whole, and the consequences of failure to perform the requirement. A complete inability to perform a major requirement of the employment in circumstances where non-performance with that requirement will result in a risk of death or serious injury, should be viewed differently from a slight deficiency at performing a minor requirement of employment where there is no likelihood that a risk to life or property would result.

To fall within the provision, an employer must show that a person "cannot" perform the requirements of the employment adequately. It is not enough that the person has difficulty in performing them if, despite this, he or she can still perform them adequately. Inability to perform them must be assessed in a practical way. One of the relevant factors is whether the person would be able to perform the genuine and reasonable requirements of the employment with reasonable safety to himself or herself and to others with whom he or she will come into contact in the course of the employment. This factor must be assessed having regard to the type and nature of the risk which exists (X's case, judgement of Gummow and Haine JJ, with which Gleeson CJ and Callinan J agreed; see also the judgement of McHugh J).

The exception does not require the employer to consider whether some group of people, defined by reference to a particular attribute, is or is not or would not be capable adequately of performing genuine and reasonable requirements of the employment. The employer must consider the adequacy of performance of the individual against whom he or she proposes to discriminate. All relevant circumstances must be taken into account, including training, qualifications and experience of the person and (where the person is already in the employment) the persons performance in that employment.

The exception applies both to ss13 and 14 of the Act. It applies both to the offering of an application for employment, and to existing employees. This accounts for the use of the words "cannot or could not adequately perform" the requirements of the employment, in s22(1)(b). These would contemplate two situations. The first is where an existing employee's actual performance demonstrates to the employer that he or she cannot adequately perform the requirements of the employment. The second is where a person seeks or applies for employment, and the prospective employer must judge whether that person could adequately perform the requirements of the employment, if he or she were employed by that employer. This judgement will be a prediction. But what kind of prediction must it be?

In our view, it must be a prediction that will enable the employer (and ultimately the Tribunal) to conclude that it is more probable than not that the person will not be able to perform adequately the requirements of the employment. The judgement will need to be genuinely made and will need to be based on information capable of supporting that judgement. An assumption, based on no information or on scant information, that a person's impairment will make him or her unable to perform adequately the requirements of the employment will not be enough. (See generally Smith's case, cited above).

Can additional evidence given at hearing be taken into account?

In this case, there was available to the Tribunal information which was not available to the employer at the time it made the decision to reject Mr Davies' application. After the rejection of that application, Professor Vingrys (of the Melbourne University's College of Optometry) carried out a number of additional colour vision tests on Mr Davies. He gave evidence about these tests and his conclusions about Mr Davies' ability to perform the requirements of employment of a Police Constable. Dr McKenzie, an Ophthalmologist, also performed additional colour vision tests on Mr Davies after the rejection of his application to join Victoria Police. Dr McKenzie gave evidence about these tests and his conclusions as to Mr Davies' abilities to perform adequately the requirements of employment as a Constable with Victoria Police. Some of the tests administered by Professor Vingrys and Dr McKenzie had not been administered by the employer when it made its decision to reject Mr Davies.

Professor Crassini (Professor of Health and Behavioural Sciences at Deakin University) gave evidence about the ability of colour vision tests to predict performance of real world tasks. While literature dealing with that subject was available to the State when it made its decision to reject Mr Davies, Professor Crassini's report was not.

How is the Tribunal to deal with this evidence? In our view, it is the Tribunal's task to decide, on the evidence before it, whether the Respondent has satisfied it that it was more probable than not that Mr Davies would not have been able to perform adequately the requirements of the employment. It is not limited to the information which was before the employer when the decision to reject Mr Davies was made. If this were not so, the Tribunal would be unable to assess whether the employer had sufficient information on which to base the decision. At the hearing, the employer is allowed to embellish the information available at the time of the decision by further evidence, and the Complainant is allowed to call other evidence which challenges the employer's conclusion.

In our view, the purpose of s22(1)(b) is to permit an employer to refuse to employ, or to dismiss or take other action in relation to a person who, because of impairment, is unable to perform adequately the requirements of the employment. It balances the consideration of permitting an employer to choose not to employ a person who cannot perform the required work against the prohibition on impairment based discrimination. But, for the exception to apply, the employer must base the decision on the circumstances of the individual, and the inability to perform the employment requirements must be real. In our view, the provision does not protect an employer who bases a decision on unfounded assumptions about the ability of an impaired person or of people with a particular impairment. It also does not protect an employer if the employer is unable to demonstrate that the person was unable adequately to perform those requirements. The employer must demonstrate that his or her decision was based on reasonable information and enquiry. The employer may add to this information by further evidence at the hearing. At the hearing, the person who is the subject of the decision may lead evidence questioning the accuracy of the information of which the employer relied, the reasonableness of any enquiries undertaken by the employer, and the reasonableness of the conclusions drawn by the employer from the information. In deciding whether the exception applies, the Tribunal will take into account all the evidence put before it. In our view, it is not the intention of the provision to enable an employer to exclude or dismiss from employment, on the basis of impairment, a person whom the employer has not demonstrated is unable adequately to perform the requirements of the employment.

This is not to say that an employer who, on the medical evidence available at the time and taking into account the individual circumstances of a person, excludes that person from employment on the basis that the person, because of impairment, is unable adequately to preform the employment requirements, would be unable to rely on s22(1)(b) if, later, new medical evidence came to light showing that the person could in fact have performed those requirements. An employer cannot be expected to take into account medical knowledge which did not exist at the time the relevant decision was made.

Section 35 of the Interpretation of Legislation Act 1984 permits us to have regard to the objectives of the Act and underlying scheme of the Act in construing s22(1)(b). There is, in our view, sufficient doubt about the proper interpretation of that provision to warrant us having regard to those objectives and that scheme. We consider that the way in which we have interpreted the provision accords with the ordinary and natural meaning of the words in it and is consistent with the scheme and objectives of the Act. That scheme is (among other things) to prohibit discrimination in relation to prospective and actual employment, but subject to an extensive number of exceptions. As the second reading speech for the Bill that eventually became the Equal Opportunity Act 1995 indicates, the exceptions attempt to strike a balance between the public interest in eliminating discrimination and other public interests. One of these is to ensure that those who are employed in particular positions are those who are capable of undertaking them. The approach we have adopted to the construction of s22(1)(b) is consistent with this underlying scheme.

One of the objectives of the Act is to eliminate discrimination as far as possible (s3). A crucial part of the definition of discrimination is that it must be substantially based on an attribute, or a characteristic that people with that attribute generally have or are presumed to have (s7). It has been pointed out in a large number of cases that one of the reasons why the Act prohibits discrimination based on presumed characteristics of an attribute is to prevent stereotyping - that is, to prevent people being treated in a particular way, not on the basis of individual merit, or even on the basis that they have an attribute, but on the basis of an assumption about what flows from the attribute which they have (Waters v Public Transport Corporation [1991] HCA 49; [1991] EOC 92-390; Smith's case, Bugden's case, cited above). Our approach to the construction of s22(1)(b) is consistent with this objective of the Act, because it requires the person who wishes to rely on this exception to demonstrate actual inability to perform adequately the requirements of the employment, rather than perceived or assumed inability to do that.

In construing s22(1)(b), we have followed an approach broadly similar to that followed by the South Australian Equal Opportunity Tribunal and affirmed by the Full Court of the South Australian Supreme Court in Smith's case (cited above). That approach recognises that the provision permits an employer to make a prediction about the ability of a job applicant with an impairment to perform the requirements of the employment. It emphasises that the judgement must be made on the basis of the individual. It stresses the consistency of this construction with the underlying scheme and objectives of the Act. However, we do not (as did the South Australian Equal Opportunity Tribunal) formulate the test for the application of the provision in this case as involving consideration of the questions whether the word genuinely and reasonably required renders it reasonable for the employer to have a colour perception test, whether the employer adopted reasonable means of testing colour perception, and whether the test was administered to the job applicant in a reasonable manner. Instead of this approach, we have followed very closely the wording of the provision and what we consider to be the ordinary natural meaning of those words.

This permits an employer to set "reasonable terms or requirements of employment" or to make reasonable variations to them, to take into account a number of listed factors. The factors are listed in paragraphs (a) to (c). There is nothing in the paragraphs to indicate whether they are intended to be cumulative or disjunctive. In our view, the provision is not only intended to operate when all of those factors are present, but when any of them are present. If all of the factors had to be present before the provision could operate, it would mean that the provision would apply only where there is a person whose impairment or physical features impose limitations on his or her capacity to undertake the employment and special service or facilities are required to enable that person to undertake the employment or to facilitate the conduct of the employment. The matters listed in paragraphs (a) to (c) are so different from each other that we do not think that Parliament intended that they must be present together before the section could apply. But we consider that at least one of the factors must be present before the section can apply.

Although the provision does not expressly permit the employer to discriminate, the provision is one of the exceptions to the Act. The operation of s12 means that the conduct which is covered by s23 is not prohibited discrimination under the Act. Although s23 does not expressly permit an employer to take whatever action is necessary to comply with a reasonable term or requirement of employment that falls within the section, it is out view that the provision must cover this by necessary implication. Parliament could not have intended that the provision would permit an employer to set a particular term of employment, but not to carry it out.

The provision permits an employer to set "reasonable terms or requirements of employment". This is a far wider expression than "genuine and reasonable requirements of the employment". Indeed, that expression appears in paragraph (a) of the section. However, the terms and requirements must still be related to the employment. The terms or requirements must be reasonable, having regard to all the circumstances. This would enable the Tribunal to consider whether, for example, the term was disproportionate to the requirements of the employment itself. We prefer to express the test in this way, rather than trying (as did Counsel for the Complainant) to relate it more closely to the concept of reasonableness which is part of the definition of indirect discrimination. Section 23 is not part of that definition. It is an exception to the prohibitions of the Act.

The purpose of the terms must be to take into account one of the factors listed in paragraphs (a) to (c).

So, reasonable terms or requirements of employment may be set to take into account the reasonable and genuine requirements of the employment (s23(a)). This would, for example, permit reasonable fitness standards or standards of qualification to be set sufficient to enable employees to meet the reasonable and genuine requirements of the employment, where those standards and qualifications were not themselves part of those reasonable and genuine requirements.

Secondly, the provision permits an employer to set reasonable terms or requirements in employment to take into account "special limitations" that a person's impairment or physical features impose on his or her capacity to undertake the employment (s23(e)). In our view, this would, for example, permit an employer to set terms of employment for a particular individual which take into account that individual's impairment based limitations. An illustration of such a term might be that the individual would not be required to undertake certain duties of the employment which he or she, because of the impairment, was unable to perform. The provision is directed to enabling an employer to make positive adjustments to the duties of employment to accommodate a particular individual's impairment or physical features. It is not, in our view, directed to permitting an employer to exclude such an individual because of incapacity to perform the duties of employment which is based on the person's impairment or physical features (for a similar conclusion in relation to s36 of the Act, see McDonald v the Marine Board of Victoria, VCAT, (AD), 17 November 1998).

Thirdly, the provisions permit an employer to set reasonable terms or requirements of employment to take into account "special services or facilities" required to enable a person to undertake the employment or "to facilitate the conduct of the employment". Apart from the words "to facilitate the conduct of the employment", this paragraph is similar in its terms to s22(1)(a)(i). In our view, the special services or facilities referred to in s23(c) are services and facilities directed to an individual with an impairment or particular physical features, rather than services or facilities directed at the employer's workforce as a whole. This is what makes the services or facilities "special". The words "him or her" in paragraph (c) refer to the person mentioned in paragraph (b) - that is the person with a particular impairment or physical features. In our view, although the second part of the paragraph does not expressly say so, the paragraph (when read as a whole) is directed to special services provided to an individual with an impairment or particular physical features to enable that individual to undertake the employment or to make it easier for that individual to undertake the employment. This is what is meant by "to facilitate the conduct of the employment". The purpose of s23(c) is similar to that of s23(b). It enables an employer to make reasonable adjustments to the terms of employment to accommodate a particular individual with an impairment or particular physical features. It does not cover terms that would exclude such a person from the employer's workforce.

The Tribunal has considered the meaning of this provision in Hall v Victorian Amateur Football Association ([1999] VCAT AD 30 EOC 92-997). In that case, the Tribunal pointed out that the words "reasonably necessary", when used in relation to conduct which was reasonably necessary to protect the safety or health of a person or the public, did not have the same meaning as "reasonable" or "necessary". It said that the conduct in question "must, in all circumstances, be reasonably necessary - that is, on a reasonable judgement" to protect the class which the conduct is designed to protect. It identified as relevant considerations in applying the provision the objective consideration of whether the conduct was in fact reasonably necessary for the specified purpose, and also the belief of the person who seeks to rely on the provision, the enquiries reasonably made by that person and the information reasonably available to that person as to whether or not the conduct was reasonably necessary for that purpose. It said -

"It is not enough if the Respondent believes the ban is necessary for the specified purpose if the ban is not on a reasonable judgement necessary for that purpose". (At p.6).

In Hall's case, the Tribunal identified the following factors as relevant and needing to be balanced against each other in considering whether or not s80(1)(a) would apply (at pp 7-8) -

"In this case, the relevant factors can be set out as a series of questions as follows

(1) What is the class whose health and safety are to be protected? What is the size of that class?

(2) What is the risk from which that class is being protected? What is the magnitude of that risk? What are the consequences to the class to be protected if the risk becomes reality?

(3) To what degree will the ban protect the health and safety of the class? Will it eliminate or reduce the risk to the health and safety of that class?

(4) Does the ban contain within itself any risk to the health and safety of the class?

(5) Are there measures currently in place to protect the health and safety of the class from that risk? Are they effective to protect the health and safety of that class from that risk?

Will the ban give that class a protection from that risk of a kind or degree that those current measures do not give?

(6) Are there nondiscriminatory alternatives that will give the class protection from the risk that is equal to or better than the ban? If there are, is there any reason why it may be impracticable for the respondent to adopt these alternatives?

(7) Did the respondent, at the time of the ban, believe that the ban was reasonably necessary to protect the health and safety of the class? On what information or enquiries was this belief based? What information on the matter was reasonably available to the respondent?"

The Tribunal had before it extensive medical evidence from recognised experts in the colour vision field. This includes evidence from Professor Crassini (a Professor in the Health and Behavioural Sciences faculty at Deakin University), Dr McKenzie (an Ophthalmologist), both of whom were called as witnesses by the Complainant. It also includes evidence from Professor Vingrys (a Professor in the Department of Optometry and Vision Sciences of the University of Melbourne), and Dr Vissenga (the Medical Officer of the Victoria Police). The Tribunal has before it extensive medical literature on the subject of colour vision testing. The Tribunal has taken all this evidence into account in making its findings on the medical aspects of this case.

There are two factors which have made the Tribunal's task in this case more difficult.

The first is that Mr Davies has never been a Constable in the Victorian Police Force. This means that the evidence as to whether he could or could not perform those duties was based on judgement and prediction. As to this, the predictions of some of the witnesses were based on Mr Davies' performance in colour vision testing. The opinions of other witnesses (including Mr Davies himself) were based on his performance of tasks in his work or recreational life. The degree of similarity between those tasks and the task of a Police Constable, and whether it can be reasonably inferred that good performance in one will mean good performance in the other, are in dispute.

The second matter concerns the predictive power of colour vision testing. The question is whether the performance of a person in colour vision testing can tell us anything about how that person will perform tasks in the real world. The medical experts agreed that no detailed research has been undertaken as to this, but nevertheless gave opinions as to the predictive power of these tests. Their opinions differed significantly as to the extent of the predictive power of those tests.

The findings on this topic are based primarily on the medical evidence and medical literature before us.

The human eye has three receptors which are responsible for colour vision. They are receptors for the wavelengths of light that correspond to the colours that are called red, green and blue. A person has defective colour vision when one or more of these receptors are lacking, do not function or do not function effectively. The medical term to describe a person in whom the colour green receptor is lacking or does not function is deuteranope. The medical term used to describe a person in whom the colour green receptor functions less than effectively is a deutan or a deuteranomalous trichromat. A deutan will be able to see the colour green (red, green and blue), but, depending on the severity of the condition, they have difficulty in detecting various shadings along the red/green spectrum of colours.

A number of different kinds of colour vision tests were in issue in this case. We say something about each of them.

The Ishihara test requires a person to identify a series of different numbers. The numbers are formed by a series of dots. They appear on a background also of dots. The hue or shade of the number and the hue or shade of the background are different. The standard commonly required for successful completion of this test is not a perfect score. It is generally accepted that a person will "pass" this test if he or she makes four or less errors in identifying the numbers.

People without a colour vision deficiency make errors on the Ishihara test. They may make errors for a variety of reasons unrelated to colour. Errors may be caused by such things as fatigue or confusion. They may make more than four errors and "fail" the test. However, on the whole of the evidence before us, it appears that those who "fail" this test are more likely to be people with a colour vision deficiency. This means that the group of people who "fail" the test will comprise mainly but not exclusively people with a colour deficiency.

The Farnsworth D15 test (the D15 test) requires a person to state correctly the sequence of 15 caps, each of a different hue. The shades or hues range across the whole of the colour spectrum. This test, and the test we subsequently describe, form the second level of testing if a person "fails" the Ishihara test. There was a difference between the evidence of Professor Vingrys and Dr McKenzie as to what score on this test would be expected from a person without defective colour vision - that is, with "normal" colour vision. On the whole of the evidence, we are satisfied that a person with "normal" colour vision would have a score of 1.4 or less on that test. It would not be expected that a person with "normal" colour vision would score higher than this.

The Farnsworth 100 Hue test (the 100 Hue test) requires a person to distinguish between 85 pots of colour each of a different shade or hue. The gradations in hue between one pot of colour and the next are small. The hues range across the whole of the colour spectrum. This test requires a refined ability to distinguish one hue from the other. Professor Vingrys illustrated the difference between the D15 and a 100 Hue test by an analogy with the runners in a race. The difference in refinement was like the difference in trying to determine which of the runners had two legs, compared with trying to determine which of the runners might eventually compete in the Olympic Games.

The scoring in the 100 Hue test is on a continuous scale, based on the number of errors made.

The Farnsworth Lantern test (the Lantern test) requires a person to identify the colours of a sequence of lights. The lights are red, green or white. They are shown two lights at a time, one light above the other. A person is required to identify the colour of each light. The most commonly accepted standard for "passing" this test is, if there are nil errors on the first run or, less than two errors in total on the second or third runs. The test was initially developed to be applied to pilots, particularly with reference to the ability to recognise lights on a runway when landing an aircraft. The wavelength of the lights is different from the wavelength of traffic lights. Intensity of the lights is similar to that which would be experienced in driving conditions of the poorest visibility.

On the evidence, it appears that the Ishihara test will be "failed" by both those who are colour deficient to any degree, and by some who have "normal" colour vision. It tells nothing about the severity of the colour vision defect.

There were differences in the medical evidence as to whether the results of a person on the D15 test, a 100 Hue test or Lantern test tell us anything about the severity of that person's colour vision defect. Dr McKenzie maintained that the only way in which the relative severity of a person's colour vision deficiency could be determined was to compare that person's results on those tests with the results of other people with colour vision deficiency who had undergone those tests. It was clear from the evidence before us that this research has never been done. It was clear from the evidence that, while some people with normal colour vision would "fail" the Ishihara test, those people would pass the D15 test, a 100 Hue test or Lantern test. It was also clear that a proportion of people with a colour vision deficiency could "pass" the D15 test, a 100 Hue test or Lantern test.

On the whole of the evidence, we are satisfied that the colour vision tests we have mentioned (if the Ishihara test is combined with one of the other tests) are capable of determining that, most probably, a person has a colour vision deficiency. In comparison with the rest of the population with a colour deficiency however, the D15 test is capable of distinguishing between a person in whom a colour receptor functions less effectively and one in whom that receptor is lacking or does not function, because the latter will not only make errors in distinguishing fine differences in hue but will also make errors in distinguishing major or gross differences in hue.

Our findings here are made taking particular account of the evidence of Dr Vissenga, Professor Crassini, Dr McKenzie and Dr Manalopolous (an Ophthalmologist).

In early 1997, Mr Davies applied to enter the Victoria Police Force as a Constable. He undertook and passed a written test to determine his skill in subjects such as English and arithmetic. In March 1997, he underwent the Ishihara test and the Lantern test. He failed both. At that time he was not asked to undertake any further colour vision tests. His application to join the Police Force was rejected.

From 1997 to 1999, Dr Manalopolous, Dr McKenzie and Professor Vingrys have administered various colour vision tests to Mr Davies. He has consistently failed the Ishihara test, the D15 test and the Lantern test, and his performance in the 100 Hue test had consistently been well beyond the range of errors that would be expected from a person with normal colour vision. Mr Davies can see the distinct colours such as red, green, orange, yellow and blue. The errors he makes are in the red-green spectrum. He is a deutan or deuteranomalous trichromat- that is, the green colour receptor in the eye functions less effectively than it should.

There is disagreement in the evidence about the severity of Mr Davies' colour vision deficiency. The assessment of the evidence is made complicated by the fact that different witnesses used the word "mild", in relation to colour vision, in different ways. Sometimes, it was used to indicate severity of colour deficiency. At others, it was used to indicate the extent of ability to perform real world tasks.

It was Professor Vingrys' opinion that, although Mr Davies' error scores would normally be taken as "borderline", the fact that the errors were repeated over numerous tests suggested a far more serious colour vision deficiency. Dr McKenzie, although indicating in his written report on Mr Davies (Exhibit E) that Mr Davies approached being a deuteranope, said in evidence that in fact he considered Mr Davies to have a mild colour vision deficiency. On the evidence, we find that Mr Davies has an unambiguous colour vision defect of the kind we have described. The state of the evidence does not enable us to say what the extent of his defect is. Indeed, the state of the evidence indicates that it may be difficult to determine this, given the nature of the tests administered.

Can performance in colour vision testing be used to predict performance in real world tasks?

The evidence before us was not directed to whether a "pass" on a colour vision test will mean that a person will be able adequately to perform certain real world tasks. The evidence was directed to whether a failure to pass the Ishihara test and one of the other tests we have mentioned can be used to predict that the person will not be able adequately to perform certain real world tasks. The predictive power of such a test will, of course, depend on whether the real world task in question is a colour-related task. The views of the witnesses who gave evidence as to this matter were, in our opinion, influenced by whether they regarded the tasks required of a Constable in the Victoria Police Force as colour-related tasks or not. So, Professor Vingrys regarded colour perception as an important element of those tasks, and concluded that performance in colour perception testing could be used to predict performance of those tasks. On the other hand, Professor Crassini regarded colour perception as only one of many elements involved in those tasks, and on that basis gave evidence that performance in colour perception tests was not a good predictor of performance of these tasks. However, Professor Vingrys did not go so far as to state that performance in colour perception tests was a certain predictor of performance in the real world tasks of a Police Constable. He said only that there was a "reasonable link" between performance in the tests and performance of the tasks.

On the evidence there is a clear difference in opinion between Professor Vingrys and Professor Crassini as to the predictive power of performance in the tests in relation to real world performance. Professor Vingrys expressed the opinion that, from Mr Davies' poor performance in the Lantern test, it was possible to predict that he would have difficulty in distinguishing red or green lights in the foreground where there was a background of white lights, in particular at night time. It was Professor Crassini's opinion that this prediction could not be made, because of the difference in the relative complexity of the task in the colour perception test and the task in the real world. In his view, the tasks in the real world involve many cues, including colour-related cues.

On the evidence, we are satisfied that, the more complicated the real world task, the less will be the predictive value of colour perception tests. This will be particularly so if the real world task involves many indicators or cues, of which colour is only one.

The Tribunal has before it a considerable amount of medical literature relating to colour perception testing (Exhibits L, N, O and 6 to 9). That literature indicates that there has been change and development in views about colour perception testing from the 1980's until now. For example, literature from the 1980's which is before the Tribunal does not examine or recommend simulation testing as an alternative or supplement to colour perception testing (see for example J E Sheedy, "Police vision standards", exhibit N). By the 1990's, the literature discusses the possibility of simulation testing and, in 1996, simulation testing was used to determine whether or not six colour vision defective candidates for the Californian Correctional Services should be employed (see Johnson et al., "Final report on entry level for Correctional Officer vision standards for the California Department of Corrections, 1992, exhibit 8; Johnson et al., "Report on colour vision evaluation for the California Department of Corrections", 1996, exhibit 9).

The literature before us generally supports colour perception testing for entry to the Police Force and the Correctional Services. It recommends that candidates be initially tested with a Pseudoisochromatic Plate test such as the Ishihara test, and, if they fail that test, with the Nagel Anomoaloscope or D15 test. The literature does not deal extensively with the Lantern test, and contains no recommendation which proposes that it should be used in preference to the D15 test or the Anomoaloscope. The literature suggests that the Anomoaloscope and the D15 test can be used in a way that will enable the extent of a colour vision defect to be measured. This is supported by the evidence of Professor Vingrys, who said that, although he had not used the D15 test to evaluate the severity of Mr Davies' colour vision deficiency, that test could be used in that way. Much of the literature before us expresses the view that colour perception is a significant element of the work of a police officer, and that a person who fails the D15 test will not have sufficiently good colour perception to be able to do that work. The conclusion that colour perception is a significant element in police work is based on information obtained from descriptions of that work taken from manuals and other documentation prepared by the Police Forces which the particular authors were considering, sometimes from questionnaires completed by members of those Police Forces, and sometimes through discussions with focus groups of members of those Police Forces. Although the literature concludes that a person who fails a D15 test will not have the necessary colour perception to be able to perform that part of the work of a police officer in which colour perception is significant, the literature (apart from the two reports by Dr Johnson that we have mentioned) does not refer to any research or practical testing that might form the basis for this conclusion.

Our findings on this aspect of the case take particular account of the evidence of Professor Vingrys, Professor Crassini, Mr Glover (Assistant Director of Policy and Strategic Services for Victoria Police), Inspector Loison (a Police Officer with 19 ½ years of operational experience) and Mr Smith (formally an Inspector with Victoria Police, with 18 years of operational experience). Our findings also take particular account of the two reports by Johnson et al. which we have mentioned earlier.

The Victoria Police currently has a simulation test to determine the agility of applicants to join the force. The test requires the applicant to get over an obstacle five feet six inches in height. This height has been set on a recent review of the test for entry into the Police Force, and is lower than the height previously set. It is designed to replicate the height of the fence of a house. In this test, no attempt was made to replicate the stressful environment that might be encountered in the real world tasks of a police officer who is required to jump the fence in the course of his duty. Nevertheless, it was clear that a person who successfully completed this test was regarded as having the necessary agility to be able to perform the duties of a Police Constable, irrespective of the fact that those duties will be carried out in a more stressful environment.

The Victoria Police do not administer any simulation test for candidates for entry into the Police Force who fail the Ishihara test and the Lantern test.

The literature before us indicated that simulation testing for those who wish to enter occupations such as the Police Force or Correctional Services and who failed diagnostic colour visions tests, is a very recent development. The only literature before us that relates to the administration of these tests is the Johnson report of 1996. Those tests were administered to candidates for entry to the Californian Correctional Service. They were administered to candidates who, on diagnostic testing, were shown to have a colour vision deficiency. The tests included tests such as requiring the candidates to tell, by the colour of the uniform, which of a line of people were prison officers and which were prisoners. The line of people was placed at some distance from the candidates. Candidates were required to complete five other tests, each based on a different scenario. They failed three out of the six of these tests. It was recommended that all be refused entry to the Correctional Service.

In evidence, Professor Vingrys and Professor Crassini criticised these tests on different grounds. Professor Vingrys criticised them because, in his view, they could not determine the ability to perform real world tasks as accurately as could a clinical colour perception test. Professor Crassini criticised them because, in his view, the test did not correctly simulate the real world tasks of the employment. They forced candidates to ignore the many other factors, apart from colour, which a person would use to distinguish a prisoner from a prison officer. They relied solely on the ability to differentiate colours, which the real world task did not. They were less complex than the real world task would have been. For example, the test we have described comprised a static line of people, whereas the real world situation would more probably have involved people who were moving and the environment would be more stressful than the test environment. Professor Crassini said that simulation tests which more closely replicated real world tasks would be both time consuming and expensive. But he did not go so far as to say that they would be impracticable to formulate or perform.

The only evidence before us of an attempted simulation test for a person who had, failed the Ishihara test, was the test administered by the then medical officer for the Victoria Police Force to Mr Smith, when Mr Smith applied to enter the Police Force in 1973. The test was to take him to St Kilda Road and ask him to identify and describe passing cars, people and objects. After this test, the medical officer recommended that Mr Smith be accepted into the Police Force. He was accepted. There is no evidence that any other tests of this kind have ever been conducted by the Victoria Police Force. This test was not put to the expert medical witnesses for comment. Taking into account their evidence concerning the difficulties of simulation testing, it is our view that this test presents more difficulties than the simulation tests conducted by Dr Johnson. Although the test presents the candidate with more complex tasks, involving some cues or factors other than colour, it still does not accurately simulate a number of conditions that would be encountered in real world tasks in the Police Force, and its nature makes it difficult to control the circumstances under which it is undertaken so that it can be repeated under identical conditions.

On the evidence, we are not satisfied that it is impossible or impracticable to conduct simulation testing for those who wish to join the Police Force and who fail diagnostic colour perception tests. Nevertheless, those tests need to be designed in a way that will replicate, as closely as possible, the real world tasks of the Police Force. They will be more time consuming and expensive than the diagnostic colour perception test. We note that it would be difficult if not impossible in this simulation test to replicate the stress often present in the situations in which a police officer will carry out his or her duties, but we also note that there is no attempt to replicate situations of stress in tests, such as the agility test, which candidates for the Police Force currently undergo. On this basis, in considering whether it is impracticable or impossible to formulate and administer simulation tests, we do not consider it necessary that the simulation test must attempt to replicate a situation of stress.

The proportion of the total population that is colour deficient is approximately 8%. Assuming that the proportion of colour deficient applicants for the Police Force will be similar to that of the general population, additional simulation tests will be required for a maximum of 8% of applicants. We take this into account in determining the feasibility of performing simulation tests. It was not suggested that simulation tests would involve the actual task - for example, pursuing another vehicle at high speed and in heavy traffic.

Our findings on this aspect of the case we take particular account of the evidence of Professor Vingrys and Dr McKenzie, and the report by Shaw and Gledhill, "Constable Selection Project - Report on Job Review Phase - Medical Physical Skills and Ability - Recommendations Tests and Standards (1995, exhibit 6).

Some people with colour vision deficiencies use filtering lenses to assist them. An example of one of these lenses is the Chromogen lens. On the evidence, it appears that Mr Davies wears Chromogen lenses, but it is not clear whether he wears them constantly or occasionally.

The Chromogen lenses and lenses like it filter out certain wavelengths. They help a person with a colour vision deficiency by varying the brightness cues which come to the eye. They make it easier for a colour vision deficient person to pass an initial test such as the Ishihara test. They do not correct a person's colour vision deficiency. They do not improve a person's performance in the second level of tests such as the D15 test or the 100 Hue test. They may impair visual depth perception. They may sometimes impair visual perception, particularly at night, by filtering out some visual information.

None of the literature recommends that these lenses be worn by a person when undergoing a colour perception test.

What are the general and reasonable requirements of employment as a Police Constable?

Our findings on this matter are made taking particular account of the evidence of Mr Glover and Inspector Loison, and of the documentation for positions as Constables in the Police Force (exhibits D and J).

We accept that it is the paramount duty of the Police Force to keep the peace and prevent offences against the peace (Police Regulation Act 1958 s13). It is unnecessary for us to identify all the genuine and reasonable requirements of employment as a Police Constable. We only deal with those which were identified by the Respondent as requirements with which Mr Davies was alleged to be unable to comply. We only identify the requirements for employment as a Constable. This is based on the evidence that all those who enter the Police Force begin with these duties and, throughout their working life in the Police Force are expected to perform these duties if the occasion arises, even though they may progress to some specialist area of the force. We are satisfied on the evidence that a Police Constable is required to do all the duties which we now describe.

We find that the following requirements, identified by the Respondent are genuine and reasonable requirements of the employment of Police Constable-

Factors which make this activity different from normal driving activity are that it may occur at high speeds, that it may occur either by day or night and in any kind of weather conditions, and in any kind of conditions of volume or flow of other traffic, and that it involves stress. On most occasions, there will be two people in the pursuit vehicle - a driver and another police officer. It is the duty of the passenger to communicate with police radio base and to observe and record relevant details as to the identity of people, objects and location. However, Police Force guidelines permit a pursuit to be undertaken by a police officer in a vehicle alone, in cases of emergency. Inspector Loison said that, in his 19 ½ years of operational police duties, he had been engaged in between 15 and 20 pursuits in a vehicle. Mr Smith estimated the number of pursuits in which he engaged while on operational police duty as about 10 in his 17 year career. We find that a Constable on operational duty would be engaged in approximately 1 vehicle pursuit per year. Inspector Loison said that, of the 15 to 20 vehicle pursuits which he had undertaken, he had been alone in the vehicle for 6 of them. In these 6, he had driven the vehicle and also recorded observations and communicated with police base.

During the pursuit, the driver will need to observe the colour of traffic lights and traffic signals, to describe objects and people for the purpose of identification, and to identify location, often in conditions of poor visibility. Many cues or indicators would be used to perform these tasks, including shape, size, position, appearance, and style of clothing. Colour is one of the cues. Although it is a significant cue, it is only one of many. It will only become critical to the performance of the task if all other cues fail or it is the only means by which the task can be performed. For example, Inspector Loison gave evidence of an incident where Police Officers had observed a vehicle at a crime scene, and identified it by its make and colour. As it turned out, they were mistaken in the make of vehicle but correct in its colour. The vehicle was subsequently located, and the area in which it was first seen and its colour were the critical factors leading to its identification.

On all the evidence, we are satisfied that perception of distinct colours (such as red, yellow, green blue and orange) is critical for the performance of vehicle pursuits. We are not satisfied that any finer form of colour differentiation is critical to that task, given the complexity of the task, the number of other cues available and the infrequency of occasions on which shade (as opposed to distinct colours) will be the only key to successful performance of the task.

Part of this requirement involves consideration of the risk which the driving of a police vehicle in pursuit, in circumstances of emergency, might pose to the occupants of that vehicle, and of other vehicles and pedestrians in the vicinity.

The consequence of an error of judgement during a vehicle pursuit may be a collision or accident resulting in injury or death, either of those in the pursuit vehicle, of those in any other vehicle, or of pedestrians. These consequences are very serious, but they must be balanced against the likelihood that they would follow only because of an error of identification of shade. We are not satisfied that this likelihood is so great that it should affect our conclusion.

This task will frequently be involved in the investigation of alleged offences. It will need to be undertaken by day or night and in all weather conditions and conditions of visibility. In his evidence, Inspector Loison gave one example of where shade, as distinct from colour, was the key element in identifying a person who had broken another persons jaw during a fight. Inspector Loison had made the identification. The only thing he had noticed about the person was that he was wearing a biscuit colour jumper with a darker stripe. The offender was identified based on this information. This was the only example that Inspector Loison gave in his evidence of a situation where shade, rather than colour, was the key to identification.

On the whole of the evidence, we reach a similar conclusion to that set out under requirement (1). We conclude that perception of distinct colours is critical to this task, but that finer colour perception than this is not. We base this conclusion on the number of other clues available for identification purposes, Inspector Loison's evidence that colour was important in approximately 5% of cases where identification of people or things is required, and the fact that there is only one instance given in evidence before us where shade (as distinct from colour) was the only key to identification.

The consequences of an error of judgement in this identification task may be that a person who did not commit an offence is initially identified as an offender, or the person who did commit the offence is not identified, or the offender is not apprehended and successfully prosecuted because of errors in the identification of objects or substances connected with the offence. These consequences are serious, but they must be balanced against the likelihood that they will occur only because of an error in identification of hue or shade, and also against the number of times they are likely to occur because of such an error. We are not satisfied that this likelihood is so great as to affect our conclusion.

There was a suggestion in the evidence that a person with defective perception of shades (rather than distinct colours) would have difficulty identifying whether substances such as blood spilt on the ground was fresh blood or not, based only on its shade of redness. We are not satisfied that this would ever be used as the sole means of identifying whether or not blood was fresh, without further forensic examination. There was also a suggestion in the evidence that there may be some objects (such as very small objects left on the ground) which might not be noticed by a person with defective perception of shades, because the person would not distinguish them from the background. There was no evidence before us about how frequently this might occur, how small the object would need to be, how often differences of shade (as distinct from differences of colour) would be critical to identification. We are not satisfied that this identification task is critical to the broader identification tasks we have identified, or that shade perception (as opposed to distinct colour perception) is critical to this part of the task.

This task involves giving evidence to a Court or Judicial Inquiry concerning the identification of people or things. An example is the evidence which Police Officers are required to give in prosecutions for offences. We are satisfied that this is something which Police Officers are required to do frequently in the course of their work.

A number of the witnesses expressed their views about how defects in colour vision might affect the credibility of the evidence which a police officer with a colour vision defect might give. We gave particular weight to the evidence of Inspector Loison and Mr Smith, both of whom have had extensive experience in operational duties in the Police Force. In his evidence, Inspector Loison gave one instance of a case where evidence of the colour of a vehicle had been critical to its identification, and one instance of a case where evidence of the shade of clothing (as distinct from its colour) had been critical to the identification of the person wearing it. He said that in his opinion there was reliance on colour perception in identity evidence given in about 5% of the cases within his experience. Neither Inspector Loison nor Mr Smith could recollect a case where the credibility of the evidence of a police officer who normally wore glasses had been successfully challenged because he or she was not wearing those glasses at the time of the incident which was the subject of the evidence.

On the evidence, we conclude that identification evidence will cover many matters, including height, build, location and so on. One of these may be colour or shade. In the majority of cases, colour or shade will not be the only key to successful identification. We conclude that evidence of colour and shade are important but are not in the majority of cases critical elements in identification evidence.

The Victoria Police Force however, does accept people who wear glasses. The evidence was that there has been no successful challenge to the credibility of evidence given by those officers who usually wear glasses to assist them but were not wearing them at the relevant time. On this basis, and by analogy, it is reasonable for us to infer that it would be unlikely (although not impossible) that the credibility of identification evidence given by a police officer with a colour vision defect would be successfully challenged. If that evidence is the sole identification evidence available, and if colour or shade is the sole means of identification available, however, we must balance this against the likelihood of that being the only identification available, and of the evidence of the officer with a colour vision defect being challenged. Taking this into account, we find that colour and shade perception is important but not critical to the giving of credible identification evidence.

The task described in the evidence before us related to the ability to be able to recognise colour-coded warning signs on containers of hazardous chemicals. A number of these signs were shown to Mr Davies and to Inspector Loison. It appeared from their evidence that they identified these signs, not so much by their colour, but by the words and symbols written on them. It was clear that, on some occasions, and because of the potential hazard involved, a police officer might not be able to approach the sign close enough to read the words and symbols on it, but may still be able to see its colour. There was no suggestion in the evidence that fine perception of shades was necessary in order to identify the colour of these signs.

We find that other cues as well as colour will generally be present to assist identification of warning signs relating to hazardous chemicals. In most cases, the words on the sign, rather than its colour will be relied on for identification. In the rare cases when colour will be the only key to identifying the sign, we are satisfied that distinct colour perception (as opposed to shade perception) is critical to this identification.

The consequence of a failure to identify a hazardous chemical correctly from a warning sign may, at the worst, be that inappropriate procedures for dealing with the chemical are adopted, and death or injury to members of the public or to Police Officers follows. These consequences are serious. However, they must be balanced against the likelihood of incorrect identification of the chemical based on the colour of the sign alone, and against the likelihood that these consequences will follow an incorrect identification. We are not satisfied that this likelihood is so great as to affect out conclusion.

(5) Identifying the target correctly when firearms are being used.

This task relates to the use of firearms in the course of police duties. It is clear from the evidence that this use will only occur in emergency situations. The factors to be taken into account in deciding whether or not the use of a firearm is appropriate will include the nature of the emergency, whether it involves an immediate threat to the life of any person, and other available means of dealing with the emergency. Inspector Loison said in evidence that, in a situation requiring the use of a firearm, a police officer might be ordered to shoot at a person identified only by the colour of the clothes he or she wore. It was his view that colour perception would be necessary to carry out this task correctly. The Inspector did not give any evidence of situations in which this had actually happened.

On the evidence, we are satisfied that a situation of an emergency in which a police officer is required to shoot at a person described only by reference to the colour of his or her clothing would occur rarely if ever. The discharge of a firearm may result in death or serious injury. We would expect that, even in an emergency, care would be taken to describe, as fully as possible in the circumstances, the person at whom the bullet was to be directed. There was no suggestion in the evidence that such a person would be identified by shade rather than by distinct colours of clothing. Although the occasion for discharging a firearm in these circumstances would be rare, the consequences are serious enough for us to be satisfied that distinct colour vision is critical for this task. However, we are not satisfied that perception of shades (rather than distinct colours) is critical for that task. We say this, taking into account that it would be unlikely that the person would be described by reference to shade alone, as opposed to reference to some distinct colour; and also that it is likely that other identification details would be given.

We next consider whether physical fitness to perform the five genuine and reasonable requirements of employment as a Constable that we have just identified is also a genuine and reasonable requirement of that employment. We then consider whether a separate colour vision standard is a genuine and reasonable requirement of that employment or is a reasonable term of that employment.

Is physical fitness to perform duties (1) to (5) a genuine and reasonable requirement or a reasonable term of employment as a Constable?

Section 8(1) of the Police Regulation Act 1958 permits the Chief Commissioner, in accordance with the regulations, to appoint (among other things) so many Constables "as the Governor in Council thinks necessary". Constables are initially appointed on a 2 year probation period. Section 130 gives wide regulation making powers. The Police Regulations 1992 have been made under these powers. Regulation 201 sets out the qualifications for appointment to the Victoria Police Force. These include requirements to be of a specified age, to be of good character, a requirement relating to Australian citizenship or permanent residence, and to have passed a general intelligence test, a written examination and an agility test. Regulation 201(1)(d) sets out the requirement for fitness. It states-

(d) Has the mental capacity and physical fitness to be able to perform the duties of a member of the force and who has completed a medical examination to the satisfaction of the police medical officer;

It was not contended for the Complainant that this requirement was discriminatory. Indeed, this requirement would appear to fall within s69 of the Equal Opportunity Act 1995, which permits discrimination where this is authorised by an Act or Enactment, including a regulation such as this. It would permit a determination that a person was physically unfit to be able to perform the duties of a member of the Police Force. In our view, the medical examination which the regulation refers to is for the purpose of determining whether the person is physically fit and has a mental capacity to be able to perform those duties. This requirement is related to the employment of Police Constable. We find that it is a genuine and reasonable requirement of that employment. Alternatively, if we are wrong in considering this as a reasonable and genuine requirement of the employment, we find that it is a reasonable term of the employment, set to take into account the reasonable and genuine requirements which we have set out under items (1) to (5) above.

Is the Respondent's colour vision standard for entry to the force a reasonable and genuine requirement, or a reasonable term, of employment as a Police Constable?

We now consider whether, apart from the requirement that a person must be physically fit to be able to perform the duties of a Police Constable, the Respondent's colour vision standards for entry to the Police Force constitute in themselves a separate genuine and reasonable requirement of that employment or a separate reasonable term of that employment.

These standards are set out in a section of the Victoria Police Force's Personnel Management Manual relating to recruitment medical policies (exhibit 3), and explained in the evidence of Dr Vissenga (the police medical officer) and Mr Glover (Victoria Police Force's Assistant Director of Policy and Strategic Services). Our findings take particular account of this evidence.

Until about 1990, medical standards for recruitment to Victoria Police were set by the Government Medical Officer. Those standards included colour vision standards. From 1990, those standards have been set and reviewed by force command (that is, by the Chief Commissioner in consultation with other senior Police Officers), on the recommendation of the police medical officer, and after consulting external experts in the various fields. In the field of colour vision, one of the external experts consulted has been Professor Vingrys. The standards are regularly reviewed. The reviews examine whether the standards continue to meet the needs of the force, whether they are consistent with continuing developments in medicine, and whether they remain consistent with evolving community attitudes. We accept Dr Vissenga's evidence that, in the course of this review, Dr Johnson's two reports in relation to the Californian Correctional Service were considered.

Reviews have led to changes in some of the standards. We already mentioned the decreased in height of the fence-like obstacle used in the agility test. The visual acuity standard has also been changed, so as to make provisions for those who wear spectacles. The colour vision standards have not been changed. They remain the same form as they were when set by the Government medical officer in the 1980's. In the recruit medical policy, the introduction states that a recruit medical standard for colour vision "is essential to the carrying out of operational duty. Applicants with significant colour vision defects will not be accepted". The introduction states that the purpose of the medical standards is to ensure that recruits "are capable both physically and mentally of undertaking the duties of an operational member of the force. The standards are designed to ensure, in so far as this is possible, that the applicants will remain fit to undertake their duties during all of their career as members of Victoria Police". The introduction goes on to state that applicants are assessed on their medical history and on medical examination by or on behalf of the police medical officer, including specialist tests.

The part of this standard relating to a colour vision test states -

"Colour vision will be tested by use of the standard Ishihara test. An error score not exceeding 4 is acceptable. Applicants having an error score of 5 or more will be referred for the Farnsworth Lantern test. Nil errors on the first run or less than 2 errors in total on runs 2 and 3 is acceptable".

It was argued for Mr Davies that the statement we have quoted from the Recruitment Medical Policy was somehow inconsistent with the colour vision standard itself. We do not agree. It was clear from the evidence that the policy was simply regarded as an explanation of the colour vision standard.

It was clear from Dr Vissenga's evidence that he regarded these colour vision tests as an indicator of ability to perform the duties of police officer. If a person did not achieve results that were "acceptable" within the terms of the standard we have quoted, Dr Vissenga regarded them as unable to perform the duties of a police officer. He did not undertake any further assessment process to determine this.

We find that the colour visions standards were not a separate requirement or term of employment, apart from the requirement or term that a recruit must be fit to be able to perform the duties of the employment. It is a means used by the employer to determine whether or not recruits are fit to perform those duties.

Is Mr Davies able adequately to perform the reasonable and genuine requirements of the employment?

The State asks us to infer, from the results of Mr Davies' colour vision tests that show he has an unambiguous colour vision defect, that he is unable adequately to perform the reasonable and genuine requirements of the employment of a Police Constable that we have just described. Mr Davies asks us to infer, from work that he has done in various kinds of employment, that he is able adequately to perform those requirements. If you look only at the requirements described in items (1) to (5) above, the question is whether Mr Davies would be able adequately to perform those requirements. In considering whether Mr Davies can comply with requirements (1), (4) and (5), it is relevant to consider whether, if he undertook those tasks, he would expose himself or others with whom he came in contact to an unreasonable risk to safety. If we look at the sixth requirement - the requirement that he must be physically fit to be able to perform the requirements described in items (1) to (5) - the question is whether Mr Davies is physically fit enough to be able adequately to perform those requirements.

We are not prepared to draw the inference for which Mr Davies contends. He gave evidence that he has driven a motor vehicle for more than 18 years and has only had one accident, and that he has for some years driven a truck without accident. He said that he had done this in all weather conditions and both by day and night. While we are prepared to infer that he will be able to undertake, as a police officer, duties that require the driving of a motor vehicle otherwise than in a pursuit situation, we are not prepared to infer that he is able to drive without risk in a pursuit situation. He has not driven in this situation.

While we accept Mr Davies' evidence that he has at various times held a pistol license, we are not prepared to infer from this that he will be able to identify a target correctly in an emergency which requires almost instantaneous judgement and response, of the type that a police officer would encounter in the course of duty. We are also asked to infer that Mr Davies would be able to perform the duties of a police officer, based on what he has already done during various kinds of employment. There was uncontradicted evidence that he is able to identify electronically colour coded components, to identify colour coded tickets commonly used in the public transport system, to detect a small piece of litmus paper dropped on a railway platform, and (as manager of a car racing team) to identify and describe racing cars during races carried out under all weather conditions. Mr Davies had not carried out the duties which we have described under items (1) to (5) above, and has not carried out duties so similar that we can infer from them anything about his ability to carry out the duties of a police officer. Although he gave evidence that he had worked for one month in public transport transit patrol, that employment was of such short duration that we are not prepared to use it as a basis for Mr Davies' capacity to perform duties common to transit patrol officers and police officers.

We are equally unable to infer that he could not perform adequately those duties.

The Respondent asks us to infer from the fact that Mr Davies has not gained an acceptable result (in the terms of its medical standards) on colour vision testing, that he is unable to perform adequately the duties of a Police Constable that we have identified above. We are not prepared to draw that inference from those results. Our reasons are these-

(a) Although it is possible to say on those results that Mr Davies' deficiency is unambiguous, the exact extent of his colour vision deficiency has not been measured. From the test results, there is nothing to indicate whether Mr Davies has a colour vision deficiency so severe that it might impact on his ability to perform the duties of a Police Constable.

(b) The results of colour vision testing are not certain predictors of performance of real world tasks. The tests concentrate only on colour, whereas real world tasks have a complexity and give a diversity of cues which colour tests do not. The real world tasks involved in the duties we have described under items (1) to (5) are of this kind.

(c) It may be possible to use the results of colour vision testing as a predictor of performance in real world tasks, where a particular kind of colour perception was a critical key to the performance of the task. We have found that, while perception of distinct colours is critical to the performance of some of the tasks we have identified under items (1) to (5), perception of shades (as opposed to perception of distinct colours) is not. Mr Davies had perception of distinct colours, but his perception of shades along the red/green axis is defective. In the context of our findings, his performance in colour vision tests which test perception of shades cannot be used as a predictor for performance of real world tasks in which perception of shades is not a critical component. There is one part of requirement (2) above in which shade perception is important. However, that part of the requirement is not so important that a failure to be able to perform it adequately could justify a finding that Mr Davies was unable adequately to perform that requirement as a whole; and still less, that he was unable to perform adequately the totality of the requirements in (1) to (5).

(d) We have identified as a consideration relevant to whether Mr Davies cannot perform requirement (1), (4) or (5) the degree of risk to which he and others with whom he came into contact would be subjected if he were to perform either of those requirements. The harm which is involved here is death or serious injury. That the risk of it occurring is not so high as to warrant a conclusion that Mr Davies could not, because of the degree of that risk, perform those requirements. In drawing this conclusion, we rely on the relatively small number of occasions of which Mr Davies would be required to drive a pursuit vehicle at high speed or to discharge a firearm at a person, the fact that he has perception of distinct colours to help in his tasks, and the variety of non-colour cues available to assist him and the fact that we are unable to say conclusively whether the risks to safety involved in Mr Davies performing these tasks with his particular degree of colour vision deficiency would be greater than the risk to safety when these tasks are performed by a person who has normal colour vision.

If we are not prepared to infer that, because of his defective colour vision, Mr Davies would be unable adequately to perform the duties we have identified under items (1) to (5), or would not be fit enough to enable him adequately to perform them, it follows that the Respondent has not satisfied us that he would be unable adequately to perform these reasonable and genuine requirements of the employment of a Police Constable. If, in the alternative, the fitness to perform the reasonable and genuine requirements listed in items (1) to (5) is a reasonable term of that employment, then the Respondent has not satisfied us that Mr Davies is unable to comply with that term. As to whether the colour vision standard is itself a separate term of employment, apart from the term relating to physical fitness, we have found that it is not.

Section 22(1)(b) requires us to consider whether Mr Davies would be unable adequately to perform the genuine and reasonable requirements of employment as a Police Constable listed in items (1) to (5) even after special services and facilities were provided by the employer. It was not argued for the Complainant that the employer should have imposed some restrictions on the duties which he is to perform. Indeed, all the evidence before us is that every Constable must perform all the duties of a Constable, and every police officer (of whatever rank) must, when occasion requires, perform all the duties of a Constable. The only special services or facility to which the Complainant points are the filtering lenses. We have found that, although these lenses may help a person to pass the Ishihara test, they do not correct a colour vision deficiency and, in the real world, and in tests such as the D15 test, do not improve a person's colour vision. They may even impair it. We find that Mr Davies' ability to perform the genuine and reasonable requirements of this employment would not be improved if these lenses were provided by the employer.

The Respondent concedes that, but for the application of s22(1)(b), 23 and 80(1), it directly discriminated against Mr Davies because of his colour vision deficiency in contravention of s13 of the Act. It concedes that, because of that deficiency, it treated Mr Davies less favourably than it would have treated an applicant for the Police Force without that deficiency, and rejected Mr Davies' application.

It was argued for Mr Davies that the State also discriminated against Mr Davies by treating him less favourably than it did Mr Smith. In our view, the comparison for direct discrimination purposes cannot be between Mr Davies and Mr Smith, because their circumstances are not the same or similar. There is no evidence that the same colour vision standards were applied to Mr Smith and Mr Davies. Their applications to enter the Police Force were twenty-six years apart. Moreover, the evidence was that, although Mr Smith had a mild colour vision deficiency, he was able to pass the D15 test. Mr Davies was not able to do this.

We have found that the Respondent has not satisfied us that Mr Davies could not perform the reasonable and genuine requirements of employment as a Police Constable. We have found that the provision of filtering lenses would not have affected this conclusion. It follows that the Respondent has not satisfied us that s22(1)(b) applies.

We have found that the colour vision standard was not a separate term of employment in itself, but was related only to the fitness of a person to perform the reasonable and genuine requirements of employment as a Police Constable. We have found that the Respondent has not satisfied us that Mr Davies was not fit enough, because of his colour vision deficiency, to be able to perform adequately the reasonable and genuine requirements of that employment.

If we are wrong and the colour vision standard is a separate term of employment, then we find that this term (used as an arbiter for determining whether or not employment should be offered) is not a reasonable term to take into account the reasonable and genuine requirements of the employment. It is not reasonable because, as we have found, it cannot of itself accurately predict whether or not a person in Mr Davies' situation will be unable to adequately perform the duties of the employment. Our task is not to decide whether the colour vision standard would be a reasonable term of employment, in relation to recruiting for the Police Force generally, or as applied to someone other than Mr Davies. It is our task to decide whether it is a reasonable term of employment in its application to Mr Davies. We find that the term is not a reasonable term of employment in its application to Mr Davies. It follows that the Respondent has not satisfied us that s23(1)(a) applies in this case.

Finally, the Respondent argues that s80(1)(a) applies. It argues that its treatment of Mr Davies was reasonably necessary to protect the safety of people and the public generally. We now set out our findings on this aspect of the case, in terms of the questions formulated in Hall's case.

(a) The relevant conduct of the State is its refusal to permit Mr Davies to enter the Victoria Police Force because of his defective colour vision.

(b) The class whose health and safety are to be protected varies according to the nature of the requirements of employment identified in (1) to (5) above. In relation to vehicle pursuit, shooting in an emergency, and colour coded warnings on containers of hazardous chemicals, the class to be protected is Mr Davies himself and police officers and members of the public who might be affected by an accident or collision caused during the pursuit, by a shot fired during the emergency or by a spill or explosion caused by inappropriate handling of the chemical. In the case of the duties relating to the giving of evidence and the identification of people and things in relation to suspected offences, the class affected is (directly) any person who may be incorrectly identified or any offender who may escape identification and (indirectly) any members of the public who may be affected because the offender remains at large and the public generally who may lose confidence in the Police Force because of the inability to successfully prosecute offenders and the decline in public protection and safety that may result.

(c) The kind of magnitude of the risks involved, and the consequences of those risks becoming realities, differ depending on the nature of the duties as Police Constable which Mr Davies would perform. A driving error during a pursuit by a vehicle may result in an accident which may cause serious injury or death to Mr Davies himself, another police officer or a member of the public. An error of identification when shooting in an emergency situation may also result in injury or death to another police officer or a member of the public. The consequences of these risks becoming realities are very serious.

An error in identification of a person or thing in relation to a suspected offence may lead to an incorrect identification of a suspect or to the real offender remaining at large. Indirectly, this may lead to a weakening of confidence in the Police Force. If the real offender remains at large, it may also lead to further offences. These risks, although serious, are more indirect and less likely to occur than the risks we have first identified. Similar comments apply to the risk that, if credible identification evidence is not given, a prosecution will be unsuccessful.

On the evidence, we are not satisfied that there is a high or even moderate risk of these consequences occurring. The Respondent had not satisfied us that Mr Davies would be unable adequately to perform the tasks we have mentioned. We are also satisfied that, in the performance of these tasks, cues are available other than cues related to colour, which Mr Davies could use to help him perform these tasks. Some of the risks we have identified (such as the risk of loss of confidence in the Police Force) are so indirect and speculative that they are insignificant.

(d) On the evidence, the Respondent's decision to refuse Mr Davies entry to the Police Force does not in itself contain additional risk to the health or safety of the classes we have identified.

(e) The only measure currently in place to reduce or eliminate any risks that a colour vision defective person might pose in carrying out policing duties is the colour vision standard for entry to the force. This must be balanced against our finding that failure to meet that standard is, in Mr Davies' case, not a certain predictor that he will be unable to perform police duties. We also take into account the fact that there may still be police officers engaged in operational duties who were not tested for colour vision deficiency when they entered the force and in fact have defective colour vision. The colour vision standard does not apply to these people.

(f) In our view, there are non-discriminatory alternatives which will protect the class we have identified from the risks we have identified. These alternatives are discussed under heading 5 in relation to the orders we propose to make. They involve additional testing of a person who does not meet the colour vision standard, to determine whether or not that person is able to perform the genuine and reasonable requirements of employment as a Police Constable. In our view, the protection which this alternative offers is better than simply banning people with defective colour vision from joining the Police Force because the alternative is directed to testing ability to perform the real tasks of the employment. We are not satisfied that it would be impracticable for the Respondent to adopt this alternative.

(g) We are satisfied that the Respondent, at the time it rejected Mr Davies' application to join the Police Force, believed that it was reasonably necessary to do this because his colour vision defect meant that, as a Police Constable, he would pose a risk to himself and the safety of other police officers and members of the public. The Respondent's belief was based on the opinion of the police medical officer, Dr Vissenga, who had in turn consulted with Professor Vingrys and examined relevant medical literature. That literature included the two Johnson reports (exhibits 8 and 9). The belief was based on all of the information which was reasonably available to the Respondent.

Balancing all of these factors, the Respondent has not satisfied us that it was reasonably necessary to refuse employment in the Police Force to Mr Davies, after he had failed to meet the colour vision standard applied by the Police Force and without further testing his ability to meet the requirements of a Police Constable. It follows the Respondent has not satisfied us that s80 applies.

We have found that none of the exceptions relied on by the Respondent apply. Accordingly, we find that the Respondent has contravened s13 of the Act by directly discriminating against Mr Davies on the basis of an attribute - his colour vision deficiency.

Under s136 of the Act, the Tribunal, if it finds any part of the complaint proved, may make any of the following orders-

• An order that the Respondent refrain from committing any further contraventions of the Act in relation to the Complainant;
• An order that the Respondent pay to the Complainant an amount to compensate the Complainant for loss, damage, or injury suffered in consequence of the contravention;
• An order that the Respondent "do anything specified in the order with a view to redressing any loss, damage or injury suffered by the Complainant as a result of the contravention".

Under s136(b) the Tribunal may find any part of the complaint proved but nevertheless decline to make orders.

The Complainant seeks a declaration from the Tribunal that the Respondent has contravened the Act. On a reading of the submissions for the Complainant, it seems that what is being sought is a finding rather than a declaration. The Tribunal has already made this finding.

• A declaration that his colour vision deficiency is not to be an impediment to his admission to the Victoria Police Force;
• A declaration that the manner of application of the colour vision standards of the Victoria Police Force are discriminatory and contravene the Act;
• An order that the Complainant be admitted to the Victoria Police Force;
• Compensation;
• Costs;
• Such further or other orders as the Tribunal thinks fit.

We first deal with the request for a declaration that the colour vision standard of Victoria Police is discriminatory and contravenes the Act. On a reading of the Complainant's submissions it appears that the Complainant is inn fact seeking an order under s136(a)(i) or (iii) rather than a declaration. The Respondent has conceded direct discrimination based on Mr Davies' colour vision deficiency, and we have found that this direct discrimination was in contravention of s13 of the Act. As we said at the beginning of our reasons for decision, we have not dealt with Mr Davies' claim of indirect discrimination, because this claim was inadequately formulated. Had we dealt with that claim and found that the colour vision standard indirectly discriminated against Mr Davies, an order of the kind which Mr Davies now seeks might have been appropriate. But we have not made that finding or dealt with the indirect discrimination claim. On this basis, we do not consider it appropriate to make this order.

We next deal with the request for a declaration that Mr Davies' colour vision deficiency is not an impediment to his admission to the Victoria Police Force. On a reading of the submissions for the Complainant it appears that what is sought is in fact an order under s136(a) rather than a declaration. The effect of such an order would be to require the Respondent to disregard Mr Davies' colour vision deficiency in determining whether he should be permitted to join the Victoria Police Force. We are not satisfied that it is appropriate to make this order. On the evidence, it appears that the core of the direct discrimination which lead to Mr Davies' rejection as an applicant for the Police Force was not so much the requirement that he "pass" specified colour vision tests. Rather, it was the assumption that, having failed those colour vision tests, Mr Davies would be unable to perform adequately the requirements of the employment. In particular, it was the making of this assumption based on tests the predictive ability of which is uncertain, without further testings Mr Davies' ability to perform the requirements of the employment. An order in the form sought by the Complainant does not address what we consider to be the conduct at the heart of the contravention of the Act which we have found the Respondent committed.

The Complainant then seeks an order that he be admitted to the Victoria Police Force. Our examination of the Police Regulation Act 1958 and the Police Regulations 1992 has shown that eligibility for appointment to the Police Force depends on qualifying in accordance with r2.01 of the Police Regulations, and that appointment to the force is in the discretion of the Chief Commissioner. Mr Glover gave evidence that, as well as the requirements in the regulations concerning an intelligence test, physical and mental fitness to perform the duties of the employment, and an agility test, there is also a final interview by a selection panel. On the evidence, we are satisfied that Mr Davies passed the intelligence test. But there is no evidence that (apart from the colour vision test) he undertook any further medical test, undertook the agility test, or was interviewed by the selection panel. The Tribunal has a very wide discretion to make orders to redress the loss, damage or injury resulting from contravention of the Act (Sinnapan and Anor v State of Victoria [1995] VicRp 27; [1995] 1 VR 421; [1995] 2 VR 242). But, before it could require the Chief Commissioner to exercise the discretion to employ Mr Davies, the Tribunal would (in our view) have to be satisfied that, but for his colour vision deficiency, the only proper exercise of the Chief Commissioner's discretion would have been to appoint Mr Davies to the force (as to this, see generally Burrows v Chief Commissioner of Police [1996] EOC 92-741).

We are not satisfied that this is the case with Mr Davies. He has not undergone a number of the tests which other applicants are required to undergo. He has not been interviewed by a selection panel. Moreover, the exact extent of his colour vision deficiency has not been determined, and he has not undergone any further tests to determine his actual ability to perform the requirements of the employment. These factors make it inappropriate to order that Mr Davies be admitted to the Police Force.

We now consider whether we should make other orders to redress the loss, damage or injury resulting from the Respondent's contravention of the Act. What is that loss, damage or injury? It is the loss of the opportunity to be tested in a way which will determine the exact extent of Mr Davies' colour vision deficiency, and Mr Davies' actual ability to perform the duties of the employment. To redress this loss, we consider it appropriate to make three orders.

The first is that the Respondent treat Mr Davies' application to join the Police Force as if it were still on foot, and that it treats him as having complied with those eligibility requirements for entry to the force with which he complied at the date he made the application.

The second is that the Respondent must test Mr Davies' colour vision using the D15 test. The evidence was that it was possible to distinguish deuteranopes from other forms of colour deficiency using that test.

If Mr Davies passes this test, he must be allowed to proceed to the next step of the selection process on the basis that his colour vision perception is sufficient to enable him to perform adequately the requirements of employment as a Police Constable. If the results of the test show that he is a deuteranope, his application may be rejected on the basis that he is unable to perceive one of the distinct colours that we have said are part of the reasonable requirements of this employment. If he fails but is not a deuteranope, then simulation testing is necessary to test his actual ability to perform the requirements of this employment.

The third is that the Respondent test Mr Davies' ability to perform the requirements of the employment by a test other than a colour vision test. We would envisage that this test would be a form of simulation test, either a computerise simulation test or a test involving real world tasks. We would envisage that the form of test chosen should take into account the complexities of the real world task to be performed. The form of test should be agreed between the parties, and in default of agreement, should be determined by the Tribunal. We note the comments in Sinnapan's case on the extensive powers of the Tribunal to make orders which it is required to supervise.

If those who administer these simulation tests to Mr Davies are satisfied that he would be able to perform the task of employment as a Police Constable, Mr Davies will proceed to the next stage of the selection process.

The Complainant then seeks compensation. The elements of this compensation are not clear. If it is compensation for the humiliation caused to Mr Davies by the Respondent's decision to reject his application for the Police Force, we have little evidence before us which would enable us to quantify that loss. We have only the evidence of Mr Davies that, since the age of 16, he had wanted to be a policeman and, initially rejected as a cadet because of his defective colour vision, he still continued to try to join the Police Force. On the evidence, he has been in continuous employment since the date of which he made the application to join the force which is the subject of this complaint. It appears from the written submissions made on Mr Davies' behalf that he does not claim loss of income.

Apart from the scant evidence, there is a more fundamental reason why we consider that an award of compensation in this case would be inappropriate. We have identified the loss resulting from the Respondent's contravention of the Act as the loss of opportunity to have Mr Davies' fitness to perform the duties of the employment determined by assessing the severity of his colour vision deficiency and then by another test designed to assess his actual performance. We are satisfied that the earlier orders we have suggested redress that loss. To award compensation in these circumstances would go beyond what is necessary to redress the loss which Mr Davies has suffered (as to this, see Holdaway v Qantas Airlines Ltd [1992] EOC 92-430). We also consider it inappropriate to award compensation to reflect any difference of income between what Mr Davies would have received if he had joined the Police Force in 1997 and what he received from his various employers between then and now. While it is fair to say that Mr Davies' application to the Respondent was rejected because of his colour vision defect, we are not satisfied that, if it had not been for that defect, he would have been employed in the Police Force. As we have said, his actual ability to perform the tasks of that employment have not been tested. Nor has he undergone all the stages of the selection process. In these circumstances, it would be inappropriate to award compensation for loss of income.

The Complainant applied for costs. We have taken into account that we have found the complaint only proven in part. We have reserved our decision on the question of costs. In our order, we have set a timetable to permit the Respondent to apply for costs if it wishes, and to permit the parties to make written submissions on the question of costs.

We find that the Respondent directly discriminated against the Complainant in contravention of s13 of the Act when, on the basis of the Complainant's colour vision deficiency, it rejected the Complainant's application to join the Police Force. The Respondent has not satisfied that its conduct is covered by any of the exceptions in the Act on which it sought to rely. At the heart of the Respondent's conduct is the assumption that Mr Davies, because of his colour vision deficiency, cannot perform adequately certain tasks required to be performed by a Police Constable. This is an assumption made without determining the exact extent of Mr Davies' colour vision deficiency, and without assessing (by a test other than a colour vision test) his actual ability to perform the tasks required of a Police Constable. We have made orders designed to test his actual ability to perform the tasks required of a Police Constable. If, taking into account the results of those tests, the Respondent is satisfied that he is able to perform those tasks, he will proceed to the next stage of the selection process.

We recognise that our orders will require a change in the method of assessment used for Mr Davies and, presumably, for people with colour vision deficiencies similar to his. There is not evidence before us on which we could be satisfied that the change which our orders require in Mr Davies' case is impracticable either because the tests are too costly or too difficult to administer.

We recognise that our decision may have application to people whose circumstances are not materially different from Mr Davies'. But the Tribunal's task is a limited one.

It is not our task to examine, in general terms, the policy underlying colour vision testing for the Police Force. Our examination is confined to the provisions of the Act and to Mr Davies.

Our decision should not be taken to mean that all colour vision testing contravenes the Act. We have not found that it is not a legitimate part of the selection process. However, because the tests are an uncertain predictor of actual performance, further tests of actual ability to perform tasks of the employment need to be undertaken. We are not satisfied that it would be impracticable to conduct such tests.

Our decision is not authority for the proposition that testing of actual performance must be undertaken for every person with a colour vision deficiency who fails the colour vision tests for entry into the Police Force. For example, a person who fails the tests because he or she lacks one or more colour receptors should be distinguished from a person such as Mr Davies, a person with a colour vision deficiency similar to that of Mr Davies but who also has severe loss of visual acuity may well be in a position quite different from that of Mr Davies.

It may be that, in the light of our findings, the Respondent might wish to consider whether the Farnsworth Lantern test is still an appropriate colour vision test for applicants to the Police Force. The D15 test might be a more appropriate test.