REASONABLY ADAPTED TO A LEGITIMATE END
24 Having answered the first Lange question "yes" it is necessary to address the second question: is the law reasonably appropriate and adapted to serve a legitimate end the fulfilment of which is compatible with the maintenance of the system of government prescribed by the Constitution?
25 The first aspect of the test was discussed by Pincus JA in Sellars v Coleman [2001] 2 Qd R 565 ("Sellars") at 567:
"The Lange test - that the law must be 'reasonably appropriate and adapted to serve a legitimate end' - requires one to deduce the 'end' or purpose of the law. If the purpose is not 'legitimate', an example of illegitimacy being a purpose of suppressing political discussion, then the law is invalid. If the purpose is legitimate, then the law may still be bad, as going further in the direction of suppression of communication than could be thought to be appropriate or adapted to the law's purpose."
Special leave to appeal to the High Court was sought from the decision of the Court of Appeal in Sellars. That application was dismissed by Gaudron and Gummow JJ in June, 2002 on the basis that "no error of principle is to be discerned in the approach of the Court of Appeal to the question whether that right [to freedom of political communication] is infringed by [the law there under consideration]": see Coleman v Sellars (2002) 23(12) Leg Rep SL5.
26 The parties to this appeal referred the Court to an extensive array of reports, including reports of the Commission, of Parliamentary Committees and of Parliamentary debates. The apparent purpose of this material was to identify the relevant object of the law. This is similar to the sort of use that is made of such materials in US jurisprudence. Care must be taken with such material in Australia. Undoubtedly material may be referred to for the purpose of identifying the mischief to which legislation is directed. However, the extensive use of these sorts of materials in order to establish, as a fact, what Parliament "intended" is neither necessary nor appropriate. This was discussed by Stephen J in Dugan v Mirror Newspapers Ltd (1978) 142 CLR 583 at 600-601:
"The important place which legislative history has assumed this century in the United States seems to me likely to mislead if sought undiscriminatingly to be applied to our own legislation, the product of a legislature operating on the Westminster model. Even on its home ground its full acceptance is of relatively recent growth. Holmes J could still speak, in 1922, of speculation about the purposes or construction of a statute based upon the vicissitudes of its passage through the legislature as 'a delicate business'…What was said by Jackson J in Schwegmann Bros v Calvert Distillers Corporation , that the court should 'not go beyond Committee Reports, which presumably are well considered and carefully prepared', whether or not it represents the majority view in the United States, does draw attention to one reason why United States' practice in statutory interpretation is likely to prove inappropriate to our Westminster system. From Congress' legislative process there emerges legislation differing quite markedly in form from that to which we are accustomed in this country; accompanying that legislation are Committee Reports of which there are here no local equivalents. Professor Corry's treatment of all this in his paper, 'The Use of Legislative History in the Interpretation of Statutes' demonstrates why the quite different legislative environment of the United States is inherently unlikely to provide appropriate guidance to the use of legislative history in our parliamentary system." (references omitted)
27 Indeed, there is no need for any evidence to establish the relevant legislative object. So long as a legitimate objective exists it does not matter whether Parliament referred to it or was even aware of it. Again we refer to Pincus JA in Sellarsat 576:
"In the present case there is no reason to think that the purpose of the by-law was other than legitimate; it was not made for the purpose of burdening 'freedom of communication about government or political matters'. To reach such a conclusion, one does not take evidence from those who made the law, but simply considers the likely purpose of the law, on its face. The mall in question is a minute part of the area governed by the council and was established to provide comfortable access on foot to the premises which line it. The purpose of a prohibition of public addresses in the malls is likely to have been to preserve those wishing to use them from being harangued about any matters - political or otherwise - by public addresses."
28 The parties also referred us to international electoral practice and even to the practice adopted in relation to the election of the recent Constitutional Convention. However, given the fundamental and basic differences between Australian election systems, involving compulsory elections with preferential voting, with the systems to which we were referred, Marshall J found that such material was not particularly helpful. We agree.
29 In this case the legislative limitation or burden is the requirement that a party be registered before it receive the various privileges available to registered parties. On its face the requirement of registration seems to be part of the legitimate objective of the regulation of elections. As it was put by the Ontario Court of Appeal in Figueroa v Canada (AG) (2000) 189 DLR (4th) 577 at [117]:
"By limiting identification of party affiliation on the ballot to registered political parties, the Act ensures that party affiliations listed on the ballot will be limited to those organizations that have the indicia normally associated with a political party (e.g., a minimum number of supporters, a leader, officers, an agent and an office), and are prepared to submit to the significant regulatory and reporting conditions established under the scheme."
30 Consequently, in our view, the registration of political parties under the Act is a necessary aspect of a valid and legitimate legislative objective. Indeed, as we have pointed out above, the appellant does not dispute this. The appellant wants to share in the privileges arising from registration, not abrogate them.
31 Nevertheless, the appellant says that Marshall J fell into error in answering the second Lange question "yes". The appellant says that Marshall J gave undue emphasis to the "margin of appreciation" that the Commonwealth had suggested should be given to Parliamentary legislation. As the quotation from Lange at 561-562 (supra) makes clear, however, the Court in Lange was not suggesting a test whereby the court could strike down a law merely because the court did not view the law as reasonable. In his submissions for the appellant before us, Mr Beach QC drew attention to the apparent difference between the "reasonably adapted" formulation used in Lange and the "reasonably capable of being regarded as appropriate and adapted" formulation used in other cases: see, for example, Brennan J in Cunliffe v The Commonwealth (1994) 182 CLR 272 ("Cunliffe") at 325. However, the apparent difference may not be so great as Mr Beach's submission may have suggested. Reference might be made, for example, to the apparent use of the "reasonably capable of being regarded as appropriate and adapted" test by a number of Justices of the High Court in Levy at 614, and 627-628. Even those of their Honours who seem to have made a deliberate choice to use the "reasonably adapted" test, have, in other cases, used the "reasonably capable of being regarded as appropriate and adapted" test. Compare, for example, Gaudron J in Cunliffe at 388 (where her Honour drew a distinction between the test applicable to a purposive head of power and the test applicable to an implied limitation) with the approach of her Honour in the earlier case of Re Director of Public Prosecutions; Ex parte Lawler (1994) 179 CLR 270 at 286 and the later cases of Langer v The Commonwealth (1996) 186 CLR 302 at 334 and Muldowney v South Australia (1996) 186 CLR 352 at 376 in all of which her Honour used the "reasonably capable of being regarded as appropriate and adapted" test in respect of the implied limitation.
32 The correct approach, at least for an intermediate court of appeal, is probably that identified by Kirby J in Levy at 644-647. After identifying the relevant differences between the approaches taken by individual Judges in particular contexts his Honour concluded at 646-647:
"In Australia, without the express conferral of rights which individuals may enforce, it is necessary to come back to the rather more restricted question. This is: does the law which is impugned have the effect of preventing or controlling communication upon political and governmental matters in a manner which is inconsistent with the system of representative government for which the Constitution provides? Such cases do exist. But in the nature of their source in Australian constitutional law they will be fewer than the multitude of First Amendment cases which have engaged the attention of the courts of the United States.
Whilst bearing in mind the foregoing discussion, the test to be applied is that recently stated in the unanimous opinion of the Court in Lange v Australian Broadcasting Corporation [at 567-568]." (references omitted)
After setting out the relevant quotation from Lange his Honour, added in a footnote:
"Note that [in Lange] at 561-562 the Court acknowledged the different formulae which had been adopted within the Court in relation to the second step but, without resolving the differences, used the 'appropriate and adapted' formula."
33 The appropriate task for this Court is to apply the test as stated in Lange, acknowledging that the form of words used in Lange was intended to reflect different formulae that had been used in previous cases; which different formulae may, or may not, have reflected minor differences in principle between the judges that used them.
34 But the Lange test should not be confused with a test of "reasonableness". The test itself contains within it a "margin of appreciation" of the proper role of the Parliament. As was said by Doyle CJ in the South Australian Full Court in Rann v Olsen (2000) 76 SASR 450 at 483:
"In answering the question now being addressed by me, it is not for the Court to substitute its judgment for that of Parliament as to the best or most appropriate means of achieving the legitimate end. The question for the Court is whether the law is 'reasonably appropriate and adapted to serve a legitimate end.' In doing so the Court must give due weight to the judgment of Parliament, while acknowledging its own ultimate responsibility to determine whether the law infringes the Constitution: ACTV at 144 Mason CJ; Cunliffe at 324 per Brennan J; Leask at 595 per Brennan CJ, at 636 per Kirby J. The extent to which the Court must defer to the judgment of Parliament in such a case has been the subject of conflicting opinions in the High Court, but at a certain point the Court must do so.
It is tempting to search for other words in which one might express the question posed for this Court, but in the end one must return to the words chosen by the High Court in Lange. The question for the Court is not a question of whether there are other and better ways of achieving the legitimate end. It involves considering whether the impediment to freedom of speech generally is a result of a measure that is reasonably appropriate and adapted to achieving a legitimate end." (references omitted)
We again refer to what was said by Pincus JA in Sellarsat [10]:
"The real question in the case, then, is whether the law is 'reasonably appropriate and adapted to serve' the legitimate end. Does it, trying to achieve that end, too greatly burden freedom of communication about government or political matters? Lange does not define the degree of suspicion or harshness which must be adopted, in scrutinising such a law; but the High Court could hardly have intended that Australian courts should readily conclude that laws passing the 'legitimate end' test are invalid because, to put it shortly, they are unreasonable."
35 It is clear from his reasons that Marshall J did not superimpose a "margin of appreciation" test on top of a "reasonably appropriate and adapted test". What he did was refer to a "margin of appreciation" as an integral aspect of determining what was reasonably appropriate and adapted. His Honour was right to do so.
36 There are two aspects where the appellant says that the registration requirements in the Act go beyond what is reasonably appropriate and adapted to achieving the legislative object. The first is the requirement for 500 members. The appellant says that any requirement of more than two members (presumably being the minimum to have a "party") is too many. But there is no reason why the minimum requirement should be the only available requirement. At the very least the Parliament must be able to take into account issues such as the extent of public support enjoyed by the party. Maybe it can also take account of the degree of recognition of the party by the voters. The Parliament could hardly be required to arrange the publication on the ballot of party affiliations if the only effect of doing so is to create confusion. It is also likely that Parliament may take into account the potential farce of the ballot paper being so large that the public lose confidence in the electoral system. Presumably it is for this reason that there are statutory requirements that a candidate must have at least fifty signatories to his or her nomination form: see s 166 of the Act. The number "500" may well, in one sense, be an arbitrary number, but nothing was put before us to suggest that it is inappropriate. It was not suggested that a political party having what might be seen as "public support" would be unable to comply with the 500 member requirement. Indeed, the DLP, although it has not had a federal member for many years, still apparently has 500 or more members. As we understand it, it is the requirement to provide a list of 500 members none of whom are overlapping (or if they are, who will choose the DLP as their party of choice) which is causing it difficulty.
37 This is not to deny that if the required number of members were sufficiently large it might be in breach of the implied limitation. To take an extreme case, if it were apparent that only one political party could comply, it is hard to see how such a requirement could be reasonably adapted to a legitimate object.
38 The second objection of the appellant was that the no overlap rule in s 126(2A) of the Act, combined with the increased investigatory powers in s 138A of the Act are not "reasonably appropriate and adapted" to a legitimate purpose. At one level it is probably sufficient merely to say that this requirement has the effect of changing the requirement from 500 members to "500 members who are prepared to acknowledge their membership and whose membership is not relied on by another political party for the purpose of being registered". In practical terms, this may operate to increase the number of required members. Given that the number of 500 is itself arbitrary there is probably no reason to think that an increased number (whatever it is) changes the scheme into one that is not reasonably appropriate and adapted to the legitimate end we have previously identified. Certainly there was nothing before us to suggest what the number might be. The Court was, however, referred to material which suggested that these changes were directed to a particular problem, namely a party registering a number of other parties with the same 500 members, but with new party names that might be attractive to the electorate and then using these "dummy" parties to direct preference votes. Apparently this problem has occurred in New South Wales. The appellant argued, in effect, that there were better or other ways to address such issues and that, in any event, it was not likely to be as significant a problem in relation to Commonwealth Parliamentary elections that it had been in New South Wales. Mr Beach QC argued that the relevant mischief would be better addressed, for example, by legislation directed to those responsible for the management of the party. Insofar as the mischief consisted of inappropriate party names, he argued that it would be better addressed by legislation directed to that topic. But these arguments do not answer the second limb of the Lange test. The question is whether the legislation is reasonably adapted and appropriate to a legitimate objective, not whether some different or other legislative approach might have been more effective.
39 In addition, the attempt to identify a relevant "legitimate" objective by the detailed perusal of Parliamentary materials can mislead. For example, Mr Beach QC also argued that the legislation was invalid for not being reasonably appropriate and adapted to a legitimate object in so far as it empowered the Commission to seek the names and addresses of members. He said that this interfered with the privacy of members. Even accepting that it does so, the disclosure of names of members might well be reasonably appropriate and adapted to the requirements of transparency and accountability within the electoral system in the same manner, for example, as is the disclosure of funding to political parties: see s 305B of the Act. In the circumstances it is unnecessary to pursue this issue any further.
40 In our view Marshall J was correct in his analysis and application of the reasonably appropriate and adapted test. The answer to the second Lange question is "yes". Consequently, the relevant provisions of the Act are not invalid by reason of any implied constitutional limitation respecting the freedom of political communication.