IMPERFECTION AND INCONVENIENCE: BOILERMAKERS’ AND THE SEPARATION OF JUDICIAL POWER IN AUSTRALIA

∗ GABRIELLE J APPLEBY

‘Persuade thyself that imperfection and inconvenience are the natural lot of mortals, and there will be no room for discontent, neither for despair.’

Ieyasu Tokagawa Shogun (1543-1616)

I IMPERFECTION AND INCONVENIENCE: AN INTRODUCTION

This article explores the imperfection and inconvenience of the 1956 decision of R v Kirby; Ex parte Boilermakers’ Society of Australia.1 It does so by drawing on several illustrations borne out in cases from the last 20 years and thereby fitting somewhat into the more contemporary theme of this special edition of the University of Queensland Law Journal. Boilermakers’ consolidated a number of previous High Court decisions forbidding non-judicial bodies to exercise federal judicial power,2 and some earlier judicial comments to the effect that federal courts could not exercise non-judicial power,3 to establish two negatively drawn principles, subject only to narrow exceptions. The twin principles are that a non-chapter III court cannot exercise the judicial power of the Commonwealth, and that a federal chapter III court cannot exercise anything other than the judicial power of the Commonwealth. Part I of this article provides an analysis of the imperfection of the reasoning employed by the majority in Boilermakers’ in drawing these principles. Part II of the article establishes the inconveniences of the decision by reference to four areas of jurisprudence and governance, each of which will be illustrated using a number of cases from the last two decades. In each area, the Boilermakers’ doctrine’s rigidity has reduced Parliament’s flexibility in responding to contemporary problems and expectations. The first area is the inhibition on the development of a system of efficient and accessible administrative decision-making tribunals. The second is the prohibition on cross-vesting of judicial power across the State and federal systems leading to difficult and technical choice of laws questions for litigants. The third is the

∗ Senior Lecturer, University of Adelaide Law School. I would like to thank Professor James Allan for the invitation to contribute to this themed edition of the University of Queensland Law Journal, and my colleagues Anna Olijnyk and Dr Matthew Stubbs for their comments on earlier versions of this article. Mistakes remain, of course, my own. 1 (1956) 94 CLR 254 (‘Boilermakers’’). 2 New South Wales v Commonwealth (1915) 20 CLR 54 (‘Wheat Case’); Waterside Workers’ Federation of Australia v J W Alexander Ltd (1918) 25 CLR 434 (‘Alexander’s Case’). 3 Alexander’s Case (1918) 25 CLR 434, 464 (Isaacs and Rich JJ). Contra the majority (Latham CJ, Rich, McTiernan JJ) in R v Federal Court of Bankruptcy; Ex parte Lowenstein (1938) 59 CLR 556 (Dixon and Evatt JJ dissenting); Victorian Stevedoring & General Contracting Co Pty Ltd v Dignan (1931) 46 CLR 73, 98 (Dixon J). Re Judiciary and Navigation Acts (1921) 29 CLR 257 was decided as a matter of jurisdiction and the definition of a ‘matter’ rather than on the basis that advisory opinions did not constitute judicial power of the Commonwealth. 266 University of Queensland Law Journal 2012

inability to modernise the military justice system through the introduction of greater independence and impartiality while maintaining a separate military system. The last is the impossibility of introducing a statutory bill of rights that maintains the supremacy of Parliament through the system of statements of incompatibility. Part II of the article demonstrates the High Court’s preoccupation with the doctrine as enunciated in Boilermakers’ rather than the objectives of the doctrine.4 Tony Blackshield warned of ‘the error of mechanically applying the words of a judgment as if they were the words of a statute’.5 Similarly, Callinan J said that ‘Reasons for judgment can only state principles, and not express rules as instruments and enactments do.’6 This ought to be remembered when considering the Boilermakers’ doctrine, ameliorating the rigidity of the judicially constructed test with the objectives and constitutional context in which it was stated. While I agree with Tokagawa that this mortal construct must be consigned to imperfection and inconvenience, I disagree that we should persuade ourselves to become content with this position, although many judges have forced themselves to be.7 We should strive for betterment: can the objectives underpinning Boilermakers’ be achieved in a less imperfect, less inconvenient way? My objective in this article is not simply to deconstruct the current jurisprudence relating to the separation of judicial power. In Part III therefore I consider an alternative limitation on the Parliament’s power that focuses on maintaining the independence, impartiality, and status of those bodies that exercise judicial power. This is a purposive, functionalist driven approach in lieu of the formalism of the Boilermakers’ doctrine.8

II PART I –THE IMPERFECTIONS OF BOILERMAKERS’

A The preceding decisions

Boilermakers’ was decided against the background of the Wheat Case,9 Alexander’s Case,10 and Dignan.11 These early decisions must be understood in their political context. The High Court in the Wheat Case considered the constitutionality of the Inter-State Commission Act 1912 (Cth). The framers had made provision for the

4 Geoffrey Sawer, Australian Federalism in the Courts (Melbourne University Press, 1967) 165. 5 Tony Blackshield, ‘Case Note: New South Wales v Commonwealth: Corporations and Connections’ (2007) 31 Melbourne University Law Review 1135, 1142, referring to the adoption of Gaudron J’s words in Re Pacific Coal Pty Ltd; Ex parte Construction, Forestry, Mining and Energy Union (2000) 203 CLR 346, 375, in New South Wales v Commonwealth (2006) 229 CLR 1 (‘WorkChoices Case’), 114-5. See also Caltex Oil (Aust) Pty Ltd v Feenan [1981] 1 NSWLR 169, 173 (Lord Diplock). 6 Coleman v Power (2004) 220 CLR 1, 111 (Callinan J). 7 On his retirement Sir Anthony Mason accepted that the separation of powers had ‘generated some less than inspiring jurisprudence’, for which he must accept some of the blame: Anthony Mason, ‘A New Perspective on the Separation of Powers’ (1996) 82 Canberra Bulletin of Public Administration 1, 1. 8 See formalist/functionalist divide explained in Peter Gerangelos, ‘Interpretative Methodology in Separation of Powers Jurisprudence: The Formalist/Functionalist Debate’ (2005) 8 Constitutional Law and Policy Review 1, 2-4; Cheryl Saunders, The Constitution of Australia: A Contextual Analysis (Hart Publishing, 2011) 199. 9 (1915) 20 CLR 54. 10 (1918) 25 CLR 434. 11 Victorian Stevedoring & General Contracting Co Pty Ltd & Meakes v Dignan (1931) 46 CLR 73 (‘Dignan’).

Vol 31(2) Boilermakers’ and the Separation of Judicial Power 267

Inter-State Commission in s 101 of the Constitution to alleviate some of the difficulties caused by the creation of the new free-trade area and the language of s 92 of the Constitution. Under s 101, the Commission was to have ‘such powers of adjudication and administration as the Parliament deems necessary for the execution and maintenance, within the Commonwealth, of the provisions of this Constitution relating to trade and commerce, and all laws made thereunder.’ An appeal was available from the Inter-State Commission to the High Court ‘as to questions of law only’.12 Andrew Inglis Clark wrote in 1901 that the Inter-State Commission, so far as it was vested with powers of adjudication, ‘will be a court of judicature’.13 J M Finnis said: ‘The Inter- State Commission was undoubtedly intended by the founders to have, or be able to have, judicial powers’.14 But in the Wheat Case, a 4:2 majority (Griffith CJ, Isaacs, Powers and Rich JJ; Barton and Gavan Duffy JJ dissenting) found the conferral of judicial powers on the Inter-State Commission by the Inter-State Commission Act invalid,15 under what would become the first limb of the Boilermakers’ doctrine. As a justice of the High Court in the Wheat Case, Isaacs J achieved his intention expressed as a delegate to the drafting Convention in Melbourne, to ‘eliminate the constitutional creation of the Inter-State Commission.’16 While not eliminating the Commission, the judgment emasculated the body so that its utility was undermined; it was eventually disbanded in 1920.17 Isaacs J’s opposition to the Inter-State Commission was matched by his opposition to the appointment of High Court justices to the Court of Conciliation and Arbitration. The Court was established in 1904 with a President who had to be appointed from the justices of the High Court for a seven-year term, eligible for reappointment.18 In Alexander’s Case, a 4:3 majority of the High Court (Barton, Isaacs, Rich, Powers JJ; Griffith CJ, Higgins and Gavan Duffy JJ dissenting) struck down legislation conferring judicial functions on the Commonwealth Court of Conciliation and Arbitration on the basis that a body other than a court could not exercise judicial power. Isaacs and Rich JJ in Alexander’s Case also suggested what would become the second limb of the Boilermakers’ doctrine. They warned against the commingling of functions in judicial bodies, commenting that ‘A court of law has no power to give effect to any but rights recognized by law’.19 In Dignan, Dixon J, generally thought to be the architect of the majority decision in Boilermakers’, had also suggested a restriction that mirrored the second limb.20 In a course of reasoning that largely follows the first strand employed by the majority in Boilermakers’ and heavily influenced by, inter alia, Isaacs J’s judgment in the Wheat Case, Dixon J came to the conclusion that:

12 Constitution s 73(iii). 13 Andrew Inglis Clark, Studies in Australian Constitutional Law (Legal Books, 1901, 1997 reprint) 186. See also John Quick and Robert Garran, The Annotated Constitution of the Australian Commonwealth (1901 ed, Legal Books, Sydney, 1976 reprint) 743; and comments of Barton J in the Wheat Case (1915) 20 CLR 54, 74. 14 John Finnis, ‘Separation of Powers in the Australian Constitution’ (1968) 3 Adelaide Law Review 159, 172. 15 Part V. 16 Official Record of the Debates of the Australasian Federal Convention, Third Session, Melbourne, 11 March 1898, 2279 (Isaac Isaacs). 17 The Commission was reconstructed briefly in 1984 under the Hawke Labor Government (implementing a Whitlam Labor Government initiative), but it would only operate for five years before being subsumed into the Industries Assistance Commission. 18 Commonwealth Conciliation and Arbitration Act 1904 (Cth) s 12(1). 19 (1918) 25 CLR 434, 464. 20 (1931) 46 CLR 73.

268 University of Queensland Law Journal 2012

Parliament is restricted both from reposing any power essentially judicial in any other organ or body, and from reposing any other than that judicial power in such tribunals.21

B The decision in Boilermakers’

Reacting to the decision in Alexander’s Case, the Commonwealth Parliament reconstituted the Commonwealth Court of Conciliation and Arbitration with judges appointed in accordance with s 72 of the Constitution. Nonetheless, a 4:3 majority in Boilermakers’ (Dixon CJ, McTiernan, Fullagar, and Kitto JJ; Williams, Webb and Taylor JJ dissenting) still found the conferral of judicial powers on it invalid. First, the High Court found that the body still failed the constitutional definition of a ‘court’ on the basis it primarily exercised non-judicial powers (its arbitral function). As such, its exercise of judicial power was in breach of the principle used in the Wheat Case and Alexander’s Case: the first limb of the Boilermakers’ doctrine.22 The High Court also provided us with the second limb of the Boilermakers’ doctrine: federal courts can only exercise judicial power,23 although the addition of this limb was in obiter.

C The two foundations of Boilermakers’

In Boilermakers’, the majority’s reasoning underpinning the drawing of the two- limbed doctrine rested on two foundations, both of which are imperfect on further examination.24 First, the majority argument looked to the structure of the Constitution to support the doctrine’s negative principles. They looked at the distribution of the three ‘powers’ of government in different branches, closely following the division in the United States Constitution. Our Constitution owes a large debt to the draft Constitution of Andrew Inglis Clark which itself was heavily modelled on the American document.25 American theory and jurisprudence, however, has always acknowledged the flexibility that must be built into the separation of powers. Thus, the reliance on the United States comparator does not seem to support the narrowly drawn and formalistic principles enunciated by the High Court. Charles de Secondat, Baron de Montesquieu, one of the first proponents of the principle in his Defense de L’espirit des Lois,26 argued that separation of powers requires both the separation of functions and the separation of persons exercising those functions. Thus, it diffuses the concentration of power and protects individual liberty against tyrannical abuse of power. (Strangely enough Montesquieu based his analysis on the English system, which is not generally thought to centrally embody the

21 Ibid 98. 22 Boilermakers’ (1956) 94 CLR 254, 270. 23 Ibid 271, 277-8. 24 For other critiques of the reasoning employed by the majority in Boilermakers’ see Leslie Zines, The High Court and the Constitution (5th ed, Federation Press, 2008) 216-18, 295-9; Sawer, Australian Federalism in the Courts, above n 4, 165; Geoffrey Sawer, ‘The Separation of Powers in Australian Federalism’ (1961) 35 Australian Law Journal 177; Finnis, above n 14; James Stellios, ‘Reconceiving the Separation of Judicial Power’ (2011) 22 Public Law Review 113 25 See William G Buss, ‘Andrew Inglis Clark’s Draft Constitution, Chapter III of the Australian Constitution, and the Assist from Article III of the Constitution of the United States’ (2009) 33 Melbourne University Law Review 718. 26 First published in 1748.

Vol 31(2) Boilermakers’ and the Separation of Judicial Power 269

separation of powers;27 commentators have argued that he was describing an idealised form of the British Constitution.)28 Philip Pettit explains that Montesquieu’s rationale for dispersing power does not require exact division, ‘provided that power still remains effectively dispersed.’29 The American founding fathers, while embracing the concept of diffusion of power espoused by Montesquieu, also accepted some flexibility in the doctrine.30 This was for three reasons. The first was the difficulties of defining the different powers with any rigour, the second was the desirability of some overlap to facilitate a system of checking and balancing between the three branches, and the last was the necessary overlapping and commingling of functions in the interests of ‘good government’.31 The emphasis on ‘good government’ reflects ideas of efficiency in government administration. Even without the influence of the United States separation of powers, the majority in Boilermakers’ argued that the ‘face of the Constitution’ made the position clear, that the express grant of the three powers of government to different branches in the first three chapters necessarily implies the separation of government power (at least in relation to judicial power). The majority said: ‘It would be difficult to treat it as a mere draftsman’s arrangement … This cannot all be treated as meaningless and of no legal consequence.’32 The majority bolstered their argument regarding the separation of judicial power by reference to the prescriptive nature of Chapter III itself,33 particularly that s 71 is an ‘exhaustive statement of the manner in which the judicial power of the Commonwealth is or may be vested’.34 This is also fallacious. Robert Garran recollected that the division was, in fact, simply a ‘draftman’s neat arrangement, without any hint of further significance.’35 The maxim expressio unius est exclusio alterius has been expressed to be ‘a valuable servant, but a dangerous master’,36 that is, it is better used as additional evidence of an otherwise determined outcome, rather than the primary evidence of the outcome.37 Its application to s 71 and Chapter III ignores the complete context in which these provisions are found in the Constitution. As has recently been emphasised by Gummow and Bell JJ in Williams v Commonwealth, ‘constitutional coherence’ must be the touchstone for constitutional interpretation.38 The approach of the majority in Boilermakers’, while recognising the existence of the overlap between the executive

27 See Eric Barendt, ‘Fundamental Principles’ in David Feldman, English Public Law (Oxford University Press, 2004) 3, 38; Geoffrey Marshall, Constitutional Theory (Clarendon Press, 1971) 124; contra Duport Steels Ltd v Sirs [1908] 1 All ER 529, 541 (Lord Diplock); R v Secretary of State for the Home Department; ex parte Fire Brigades Union [1995] 2 AC 513, 567 (Lord Mustill). 28 See, eg, M J C Vile, Constitutionalism and the Separation of Powers (Clarendon Press, 1967) 84-5; Stanley de Smith and Rodney Brazier, Constitutional and Administrative Law (Penguin Books, 7th ed, 1994) 18. 29 Philip Pettit, Republicanism: A Theory of Freedom and Government (Oxford University Press, 1997) 179-80. 30 Madison, Federalist No 47, (in Clinton Rossiter (ed), The Federalism Papers (Penguin, 2003) 299. 31 Madison, Federalist No 37, (in Rossiter, above n 30) 223. See also discussion in William B Gwyn, ‘The Indeterminacy of the Separation of Powers in the Age of the Framers’ (1989) 30 William and Mary Law Review 263, 263. 32 Boilermakers’ (1956) 94 CLR 254, 275. 33 Ibid 270, 272. 34 Ibid 270. 35 Robert R Garran, Prosper the Commonwealth (Angus and Robertson, 1958) 194. 36 Colquhoun v Brooks (1888) 21 QBD 52, 65 (Lopes LJ). 37 See explanation in D C Pearce and R S Geddes, Statutory Interpretation in Australia (LexisNexis Butterworths, 6th ed, 2006) 138. 38 (2012) 86 ALJR 713, [157].

270 University of Queensland Law Journal 2012

and legislative arms to provide for responsible government, 39 fails to interpret Chapter III in the context that this creates. Responsible government compromises any sort of strict separation between the legislative and executive branches.40 The necessary overlap in these two branches also undermines any implication that by setting out the three branches in their own separate chapters there was an intention to divide the branches and their powers. While the majority in Boilermakers’ acknowledged responsible government led to the commingling of the legislative and executive institutions and powers, they did not accept this should influence the separation of judicial power. Separation of judicial power ignores the constitutional provision in Chapter III that allows appeals to the Queen in Council.41 This provision reminds us that keeping the judicial branch separate runs contrary to the historical commingling of judicial power with other government powers in Britain. In Britain, the Queen in Council acts as the final appeal court for many Commonwealth countries, British overseas territories and dependencies. Under the Judicial Committee Act 1833 (UK), the Queen must refer judicial matters to the Judicial Committee of the Privy Council (constituted by judicially qualified members) for a recommendation,42 and by convention the Queen follows the Judicial Committee’s advice. There are other British examples that highlight the foreignness of separating judicial power from other government functions. The Lord Chancellor was traditionally a senior Minister, the head of the judiciary and the presiding officer of the House of Lords.43 The House of Lords historically included the Law Lords, who performed the dual roles of sitting as the final court of appeal in the United Kingdom and also sitting and voting in the House of Lords, although by convention they generally abstained from politically controversial votes.44 The Boilermakers’ doctrine also ignores the conferral on the Parliament of the power to punish for contempt,45 and that section which establishes an Inter-State Commission with ‘powers of adjudication’ conferred on it by Parliament.46 Finally, it also ignores the ready acceptance by the framers in Chapter III to give the Parliament power to confer federal judicial power on State courts47 – courts that, since colonial times, had never been protected by a strict, formalistic doctrine of separation of judicial power. In many colonies, for example, the Governor in Council also sat as the Court of Appeal.48 There is no evidence in the convention debates that the framers intended to adopt a strict separation of judicial power that would run contrary to the traditions they had known in Britain and the colonies.49

39 Particularly Australian Constitution s 64. 40 Although even in the United States there is some commingling of functions to provide for ‘checks and balances’, see further explanation below. 41 Australian Constitution Section 74. 42 Australian Constitution Section 3. 43 This was changed by a number of constitutional reforms in the early twenty-first century. 44 A W Bradley, ‘The Constitutional Position of the Judiciary’ in David Feldman, English Public Law (Oxford University Press, 2004) 333, 346-7. The House of Lords was removed and replaced by a Supreme Court in constitutional reforms taking effect in 2009. 45 R v Richards; Ex parte Browne and Fitzpatrick (1955) 92 CLR 157. 46 Australian Constitution s 101. 47 Australian Constitution s 77(iii). 48 Indeed, this continued in South Australia until after federation. See, eg, Bradley Selway, The Constitution of South Australia (Federation Press, 1997) 108. 49 Fiona Wheeler, ‘Original Intent and the Separation of Powers’ (1996) 7 Public Law Review 96, 99-102; Finnis, above n 14, 170-7. Note also that both Andrew Inglis Clark and Harrison Moore (at least in the 1902 edition of his work) make no strong statements about the entrenchment of a strict separation of powers: Inglis Clark, above n 13, 31; Harrison Moore, The Constitution of the Commonwealth Australia (John Murray, 1902) 89, Ch V, 280-1.

Vol 31(2) Boilermakers’ and the Separation of Judicial Power 271

The second foundation of the reasoning in Boilermakers’ was that the separation of judicial power was basic to the operation of the federal system.50 The majority said:

The position and constitution of the judicature could not be considered accidental to the institution of federalism: for upon the judicature rested the ultimate responsibility for the maintenance and enforcement of the boundaries within which governmental power might be exercised and upon that the whole system was constructed.51

This second line of reasoning emphasises the importance of the independence of the judiciary but does not provide the next logical step that is required to substantiate the Boilermakers’ doctrine. Why does independence require strict separation of judicial power? Contrast the position taken by the majority in Boilermakers’ to that in the United Kingdom, where separation of powers is not a central constitutional concept, yet the independence of the judiciary has been acknowledged as a fundamental principle for centuries.52

D Two ongoing flaws of Boilermakers’

The imperfections of the Boilermakers’ doctrine can be illustrated by two issues that it has left for subsequent courts to grapple with. The first is the need to define ‘judicial power’, a task which has proven difficult and overly technical. The second is the practical need to allow the conferral of some non-judicial functions on judges, leading to the High Court’s creation of the persona designata doctrine.

E Defining ‘judicial power’

Dixon CJ’s successor, Barwick CJ, was highly disparaging of the Boilermakers’ decision. He said in R v Joske; ex parte Australian Building Construction Employees and Builders’ Labourers’ Federation that it ‘leads to excessive subtlety and technicality’.53 He was probably referring to the need to characterise all government powers neatly into the three categories set out in ss 1, 61, and 71 of the Constitution. Both limbs of Boilermakers’ require the classification of powers into judicial and non- judicial. James Stellios refers to this as ‘an elusive task’;54 Peter Gerangelos says it is ‘vexed, multifaceted, intricate’.55 It is certainly not one that the colonial or British constitutional systems ever required.

Contra Quick and Garran, above n 13, 720, 767 and Harrison Moore, The Constitution of the Commonwealth Australia (John Murray, 2nd ed, 1910) 96-7, 303. Although note Robert Garran’s comments in his memoirs that the first three chapters was a mere draftsman’s arrangement at footnote 35. 50 Boilermakers’ (1956) 94 CLR 254, 276; See also Alexander’s Case (1918) 25 CLR 434, 469 (Isaacs and Rich JJ). 51 Boilermakers’ (1956) 94 CLR 254, 276. It was this aspect of the decision that was particularly emphasised by the Privy Council on appeal: Attorney-General (Cth) v The Queen (1957) 95 CLR 529, 540-1.This largely accords with Dicey’s formulation of the essential characters of federalism to include ‘the authority of the Courts to act as interpreters of the constitution’ A V Dicey, Introduction to the Study of the Law of the Constitution (Liberty Fund, 8th ed, 1915, 1982 reprint) 77, 88-92, and nicely distinguishes the Australian position from that in Britain where there exists no similar doctrine. 52 See, eg, the protection of judicial tenure from summary dismissal by the Crown in the Act of Settlement 1701. 53 (1974) 130 CLR 87 (‘R v Joske’), 90. 54 James Stellios, The Federal Judicature: Chapter III of the Constitution (LexisNexis Butterworths, Chatswood, 2010) 107. 55 Gerangelos, above n 8, 1.

272 University of Queensland Law Journal 2012

The judges have been able to agree that there must be certain ‘core’ judicial functions: for example, the adjudication of criminal guilt,56 and the conclusive determination of disputes about pre-existing rights.57 Otherwise, the process of classifying judicial power rests on weighing up the combination of factors which the court has identified as being relevant to what is judicial power, but in making that assessment ‘no single combination of necessary or sufficient factors identifies what is judicial power’.58 For example, in some instances, judicial power can create rights, a characteristic more readily associated with legislative power.59 In others, the judicial function has extended to making determinations on what would ordinarily be questions of policy.60 Some functions are found to be judicial simply because they have historically been exercised by the courts. Many characters of judicial power are shared with administrative power: administrators often interpret the law, apply it to the facts and resolve disputes between parties.61 The difficulty of classifying many functions as judicial or administrative has led to the development by the Court of the somewhat circumspect ‘chameleon’ doctrine.62 Under this doctrine, the Court accepts that to some extent the nature of the body determines the characterisation of the power on the basis that the power will be exercised consistently with its functions.63 So the adjudication of a dispute will be administrative when conferred on an administrative tribunal (which may not have to comply strictly with judicial process, and the findings of which will not be final) but judicial if conferred on a court, with all the judicial trappings attaching to it.64 In 2007 in Thomas v Mowbray, the Commonwealth Solicitor-General, referring to the cases that have developed the chameleon doctrine, claimed: ‘Those cases have not overruled the Boilermakers’ Case but they have removed much of its rigidity so that it does not matter much any more.’65 Similarly, Fiona Wheeler has argued that the Court’s flexible approach to defining ‘judicial power’ as seen in the use of history and the chameleon doctrine, has in fact been advantageous for the Boilermakers’ test. She said: ‘In practice, the Court has limited the effect of Boilermakers’ by adopting a pragmatic and flexible approach to identifying judicial and non-judicial functions.’66

56 Chu Kheng Lim v Minister for Immigration (1992) 176 CLR 1, 27 (Brennan, Deane and Dawson JJ). 57 R v Trade Practices Tribunal; Ex parte Tasmanian Breweries Pty Ltd (1970) 123 CLR 361, 374 (Kitto J); Brandy v Human Rights and Equal Opportunity Commission (1995) 183 CLR 245, 268 (Deane, Dawson, Gaudron and McHugh JJ). 58 Attorney-General (Cth) v Alinta Ltd (2008) 233 CLR 542, 577 (Hayne J). 59 See, eg, Thomas v Mowbray (2007) 233 CLR 307 note particularly comments at 328 (Gleeson CJ). 60 See explanation of what is permissible in Thomas v Mowbray (2007) 233 CLR 307, 348-51 (Gummow and Crennan JJ). 61 These characteristics are shared by judicial power, see particularly the classic statements by Griffith CJ in Huddart, Parker & Co Pty Ltd v Moorehead (2909) 8 CLR 353, 357; and Kitto J in R v Trade Practices Tribunal; ex parte Tasmanian Breweries Pty Ltd (1970) 123 CLR 361, 374. 62 This type of power was first referred to in British Imperial Oil Co Ltd v Federal Commissioner of Taxation (1926) 38 CLR 153, 176 (Isaacs J); and the label adopted in R v Quinn (1977) 138 CLR 1, 18 (Aickin J). 63 Sawer, ‘The Separation of Powers in Australian Federalism’, above n 24, 180. 64 R v Trade Practices Tribunal; ex parte Tasmanian Breweries Pty Ltd (1970) 123 CLR 361, 373 (Kitto J). 65 Thomas v Mowbray (2007) 233 CLR 307, 316 (David Bennett QC). 66 Fiona Wheeler, ‘The Boilermakers Case’ in H P Lee and George Winterton, Australian Constitutional Landmarks (Cambridge University Press, 2003) 160, 171; see also Fiona Wheeler, ‘The Separation of Judicial Power and Progressive Interpretation’ in H P Lee and Peter Gerangelos, Constitutional Advancement in a Frozen Continent: Essays in Honour of

Vol 31(2) Boilermakers’ and the Separation of Judicial Power 273

Whether or not the development of the chameleon doctrine and the High Court’s more recent flexible approach to determining what is judicial power has been beneficial in its amelioration of the effect of the Boilermakers’ doctrine, the point still remains that it is only because of the doctrine’s rigid separation of judicial power from other government functions that this has had to occur at all. A less imperfect alternative would change the focus of this inquiry from defining ‘judicial power’ so as to comply with the limbs of Boilermakers’, to determining whether the conferral of a power on a body (judicial or non-judicial) is incompatible with the independence, impartiality and status of the judiciary, or the exercise of judicial power. This alternative test of incompatibility will be returned to in Part III.

F Persona Designata

Dixon himself was deployed during his tenure as a Justice of the High Court as Minister Plenipotentiary to America during World War II. Judges are often appointed as Royal Commissioners, tribunal members, authorities to issue surveillance warrants, and one has even been appointed Director-General of ASIO.67 These appointments are usually justified because the judge’s independence, impartiality and integrity are perceived as desirable in these important government roles. But how are they allowed against the background of the strictness of the Boilermakers’ second limb? Such appointments are made in the judge’s personal capacity. In Grollo v Palmer the High Court accepted that provided the judge consents to such an appointment, and the appointment is not incompatible with their status as a judge of a federal court and their exercise of judicial power, these types of persona designata conferrals are constitutional and do not breach Boilermakers’.68 The doctrine has been heavily criticised. Geoffrey Sawer argued that one could not ‘draw a very sharp distinction between a judge and a court’.69 He went on: ‘In practical language, it makes no sense to say that a judge can discharge non-judicial functions unless a distinction between his judicial-official and his other-official personality can be drawn.’70 Leslie Zines explained the problem with the doctrine:

If contact with non-judicial functions is, as the Boilermakers’ case suggests, likely to undermine the impartiality and independence of the judiciary, that influence will be no

George Winterton (Federation Press, Annandale, 2009) 222; Stellios, The Federal Judicature, above n 54, 207; and Cheryl Saunders, ‘The Separation of Powers’ in Brian Opeskin and Fiona Wheeler (eds) The Australian Federal Judicial System (Melbourne University Press, 2000) 3, 13; Saunders, above n 8, 200. 67 Enid Campbell and H P Lee, The Australian Judiciary (Cambridge University Press, 2001) chapter 7. 68 (1995) 184 CLR 348, 364-5 (Brennan CJ, Deane, Dawson and Toohey JJ). The incompatibility test was further explained in Wilson v Minister for Aboriginal and Torres Strait Islander Affairs (1996) 189 CLR 1. These cases follow the earlier decisions of Drake v Minister for Immigration & Ethnic Affairs (1979) 46 FLR 409 and Hilton v Wells (1985) 157 CLR 57. 69 Sawer, ‘The Separation of Powers in Australian Federalism’, above n 24, 182. 70 Ibid. See also Mason, above n 7, 5. Mason’s criticisms have been referred to in the more recent judgment of French CJ and Kiefel J in Wainohu v New South Wales (2011) 243 CLR 181, when they begrudgingly extended the doctrine to the conferral of powers on State judges: [49].

274 University of Queensland Law Journal 2012

less if the judge voluntarily undertakes the function in a ‘personal capacity’ than if he or she does so as a member of a court.71

If the consent of a judge overcomes any potential threat to judicial independence, and a test of ‘incompatibility’ suffices to ensure the integrity of the judicial institution, why would this not also suffice in terms of federal courts?72 The absurd legal fiction of persona designata demonstrates that the strictness of Boilermakers’ is not necessary. The alternative test of incompatibility will be returned to in Part III.

III PART II – INCONVENIENCES OF THE LAST 20 YEARS

Boilermakers’ intended to promote objectives that emphasise the importance of an independent and impartial judicial system. An independent and impartial judiciary promotes the dispersal of power and in that way offers some protection of individual liberties, and also provides an authoritative arbitrator on federal matters. But while history may have emphasised the importance of these objectives, it has not demonstrated the necessity of a strict separation of judicial power to achieve them. This at the heart of Barwick CJ’s criticism in R v Joske when he said the principal conclusion in Boilermakers’ was unnecessary ‘for the effective working of the Australian Constitution.’73 What the imperfect formalism of Boilermakers’ has given us then, is inconvenience. Williams J, in dissent in Boilermakers’, repeatedly emphasised the importance of a constitutional system which would allow for practicability and flexibility in constitutional arrangements.74 This part will consider four areas in which cases decided in the last two decades demonstrate the inflexibility and inconvenience of the Boilermakers’ decision. Other inconveniences that the Boilermakers’ doctrine has given us that have not yet come before the courts will not be considered.75

A Administrative Tribunals

1 Federally: Brandy v Human Rights and Equal Opportunities Commission (1995)

Today’s modern regulatory state requires flexibility to deal with a plethora of social, political and legal questions, many of which were unforeseen at the time of framing the Constitution. The state must deal quickly and efficiently with a vast number of disputes that may arise in specific areas, including trade practices, copyright infringement and human rights breaches. All constitutions must grapple with the question of limiting power but also ensuring flexible empowerment to meet these future needs. The Boilermakers’ doctrine adds unnecessary inflexibility to our arrangements. The case of Brandy v Human Rights and Equal Opportunities Commission demonstrates the limits Boilermakers’ places on the powers of administrative tribunals.

71 Zines, above n 24, 298. See also Sawer, Australian Federalism in the Courts, above n 4, 165; Kristen Walker, ‘Persona Designata, Incompatibility and the Separation of Powers’ (1997) 8 Public Law Review 153, 167. 72 Zines, above n 24, 299. 73 (1974) 130 CLR 87, 90. 74 Boilermakers’ (1956) 94 CLR 254, 301, 307, 309. 75 See discussion of some of these in James Duffy, ‘Problem-Solving Courts, Therapeutic Jurisprudence and the Constitution: If Two is Company, is Three a Crowd?’ (2011) 35 Melbourne University Law Review 394, 406-7; Stellios, ‘Reconceiving the Separation of Judicial Power’, above n 24, 124.

Vol 31(2) Boilermakers’ and the Separation of Judicial Power 275

Even at the time the Constitution was drafted, the framers recognised that there may be some disputes that require a quick determination by a specialist body. Most notably, the framers made provision for the Inter-State Commission in s 101 of the Constitution. However, as already outlined above, the Inter-State Commission was quickly rendered useless by the High Court in the Wheat Case under what would become the first limb of Boilermakers’. Referring to the demise of the Inter-State Commission, Justice Rae Else-Mitchell commented that ‘the strict separation of powers doctrine that has been formulated in this country ... has resulted in some sort of stultification of our governmental processes or their potentialities, and as I have indicated it seems to lead to an inhibition of the development of a sound system of administrative law in this society.’76 In 1995, Else-Mitchell’s warnings played out in the case of Brandy v Human Rights and Equal Opportunities Commission. The case concerned a challenge to the powers of the Commission (HREOC) to conciliate and arbitrate disputes under Commonwealth anti-discrimination legislation. The Racial Discrimination Act 1975 (Cth) was amended in 1992 to allow a decision of the HREOC to be registered in the Federal Court at which point it would be enforceable ‘as if it were an order made by the Federal Court’.77 Prior to the amendments, the HREOC’s determinations were non- binding, and to achieve conclusiveness, a litigant would have to apply separately to the Federal Court. The Federal Court then had to be satisfied that a breach of the legislation had occurred. In introducing the changes in 1992, the Prime Minister referred to remarks of Wilcox J in Maynard v Neilson78 that the system created considerable costs if a person wanted to pursue a claim in the Federal Court and undermined the integrity, status and therefore influence of the HREOC hearing and determination.79 The High Court in Brandy applied the first limb of the Boilermakers’ doctrine to the new scheme, finding that the HREOC’s decisions could not be registered in the Federal Court, as this gave the HREOC’s powers the element of conclusiveness that made a previously non-judicial power an exercise of judicial power. The decision undermined the objectives of the 1992 amendments. If litigants had to commence Federal Court proceedings separately many benefits were undermined, including efficiencies in time and cost and the availability of expert commissioners.80 The federal government responded by introducing a system of Federal Magistrates Courts in 2000.81 The Federal Magistrates Court has been plagued with controversy, predominantly on the basis of its cost and lack of use by litigants. There is evidence that since the enforcement of federal anti-discrimination laws moved to the federal courts, most complaints have been lodged under the State system.82

76 Commentary provided by Rae Else-Mitchell in Sawer, ‘The Separation of Powers in Australian Federalism’, above n 24, 194. 77 Racial Discrimination Act 1975 (Cth) s 25ZAB. 78 (1988) EOC 92-226. 79 Commonwealth, Parliamentary Debates, House of Representatives, 3 November 1992, 2397. 80 Sarah Joseph and Melissa Castan, Federal Constitutional Law: A Contemporary View (Thomson Lawbook Co, 3rd ed, 2010) 191. 81 Federal Magistrates Act 1999 (Cth). 82 Beth Gaze, ‘Has the Racial Discrimination Act contributed to eliminating racial discrimination? Analysing the litigation track record 2000-04’ (2005) 11 Australian Journal of Human Rights 171, 181.

276 University of Queensland Law Journal 2012

2 States: Commonwealth v Anti-Discrimination Tribunal (Tas) (2008)

Boilermakers’ may also cause difficulties for the efficient administration of justice in State tribunals. In Commonwealth v Anti-Discrimination Tribunal (Tas),83 the Full Federal Court was asked to determine whether the Commonwealth and one of its agencies, Centrelink, were subject to the jurisdiction of the Anti-Discrimination Tribunal (ADT). The Commonwealth successfully argued Centrelink and the Commonwealth were not ‘persons’ to whom the Anti-Discrimination Act 1998 (Tas) applied. The Commonwealth also argued that even if they were persons to whom the Act purported to apply, this would be unconstitutional because it would mean the ADT was exercising the federal judicial power of the Commonwealth in contravention of the first limb of Boilermakers’. Because the Commonwealth had been successful on its statutory interpretation argument, the judges did not have to consider the Boilermakers’ question. Nonetheless, Kenny J did. The argument followed a number of propositions.84 The first was that the Tribunal was not a ‘court’ of the State capable of being vested with federal jurisdiction under s 77(iii) of the Constitution. The second was that the Tribunal was exercising judicial power in determining the dispute under the Act. The third was that because the Commonwealth was a party to the matter, this must be federal judicial power. The final proposition rested on an implication that s 75(iii) required that any exercise of jurisdiction over the Commonwealth be federal jurisdiction. Kenny J accepted all of the propositions.85 In relation to the final point, her Honour concluded:

Federal jurisdiction is attracted whenever there is an exercise of judicial power in respect of a matter of the kind described in ss 75 and 76 of the Constitution. Furthermore, Ch III precludes the existence of residuary State judicial power in respect of any such matter. It follows from this that a State Parliament cannot confer State judicial power on either a State court or a tribunal in respect of a matter in ss 75 and 76 of the Constitution, because these matters attract federal jurisdiction in which only Commonwealth judicial power is exercisable.86

If Kenny J’s position is correct, it could lead to a number of outcomes, each with its own difficulties. Leaving the system as it is (which is what has occurred since the decision to date) means that the Commonwealth is not subject to the jurisdiction of State tribunals when they exercise judicial power. This hardly accords with general notions of justice and equality before the law. Kenny J’s position would also mean that matters between residents of different States, or between a State and a resident of another State could not be within the jurisdiction of State tribunals exercising judicial power.87 The States could amend their legislation to remove judicial powers from tribunals, as was done at the federal level after Brandy, but this of course would result in the same inconvenience at the State level, where a litigant could not achieve finality through the tribunal system. Alternatively, the States could change the nature of their

83 (2008) 169 FCR 85. 84 See the further explanation of the issues in the case in Gabrielle Appleby, ‘Chapter III and the State Tribunal System: Guarantee of Liberty or Unprofitable Inconvenience?’ (2009) 17 Australian Law Librarian 142; and an excellent analysis of the reasoning in Geoffrey Kennett ‘Fault lines in the autochthonous expedient: The problem of State tribunals’ (2009) 20 Public Law Review 129. 85 In taking this position she was largely following an earlier New South Wales decision, Attorney-General (NSW) v 2UE Sydney Pty Ltd (2006) 236 ALR 385. 86 Commonwealth v Anti-Discrimination Tribunal (2008) 169 FCR 85, [222] (Kenny J). 87 Australian Constitution s 75(iv).

Vol 31(2) Boilermakers’ and the Separation of Judicial Power 277

tribunals so that they meet the standards of a ‘court of a State’ under s 77(iii) of the Constitution. However, this also carries with it inconveniences – it would undermine many of the informalities and efficiencies that make the tribunal system more flexible and thereby accommodating to litigants.

B Cross-Vesting: Re Wakim; Ex parte McNally (1999)

The Family Court and the Federal Court were introduced in the 1970s,88 creating a system of federal courts in parallel with State courts exercising federal jurisdiction. In the 1980s, the Commonwealth, State and Territory governments worked together to establish a system of cross-vested jurisdiction to help streamline these parallel systems for litigants. By cross-vesting federal jurisdiction in State courts and State jurisdiction in federal courts, litigants would be spared often difficult choice of laws questions at the outset of litigation. While not establishing a single system of courts, it overcame many of the difficult jurisdictional disputes that arose in the course of navigating the new federal system. The cross-vesting scheme was introduced in 1988 and worked well for a number of years,89 before being struck down by the High Court in Re Wakim; ex parte McNally.90 Former Commonwealth Solicitor-General Gavan Griffith described the High Court’s decision in Re Wakim as his ‘worst loss’.91 The majority held that the second limb of Boilermakers’ precludes the vesting of State judicial power in federal Chapter III courts. McHugh J, in the majority, explained the decision in two ways. First, the absence of an express constitutional provision allowing the vesting of State jurisdiction in federal courts in the presence of an express provision that does allow the vesting of federal jurisdiction in State courts92 necessarily precluded the former. Secondly, Chapter III exhaustively defines the matters that may be the subject of federal judicial power and vested in federal courts.93 Gummow and Hayne JJ expressly dismissed counter-arguments based on the convenience to litigants facilitated by the cooperative nature of the cross-vesting system.94 Kirby J issued a strong dissent, in the course of which he said:

If the arguments of the challengers are accepted, the foregoing would appear to consign those seeking to restore the benefits of cross-vesting legislation, enjoyed these past dozen years, to the highly problematic and expensive task of proposing and securing a formal amendment to the Australian Constitution. The inconvenience of such a rigid construction of Ch III is then shown in sharp relief.95

Re Wakim brought into ‘sharp relief’ the inconveniences of the second limb of Boilermakers’. It also appears unnecessary if we return to the justifications for the doctrine. How the conferral of State judicial power could undermine the independence, impartiality or status of the federal judiciary is not explained. The framers certainly believed that colonial courts, who traditionally exercised colonial judicial power (together with other powers) had the requisite independence, impartiality and status to

88 Family Law Act 1975 (Cth); Federal Court of Australia Act 1976 (Cth). 89 Jurisdiction of Courts (Cross-Vesting) Act 1987 (Cth). 90 (1999) 198 CLR 511. This was in fact the second challenge to the legislation, the first in Gould v Brown (1998) 193 CLR 346, being unsuccessful, the court splitting 3:3. 91 Anthony Malkoun, ‘Interview with Gavan Griffith’ (2009) 43(Summer) Young Lawyers Journal 12. 92 Constitution s 77(iii). 93 Re Wakim (1999) 198 CLR 511, 557. 94 Ibid 572. 95 Ibid 609.

278 University of Queensland Law Journal 2012

exercise federal judicial power when they became State courts. What the decision has done is continued the jurisdictional difficulties for litigants that the scheme was intended to counter.

C Military Justice: Lane v Morrison (2009)

Australian service men and women have historically been subject to the jurisdiction of courts martial. This is explained not generally as an exception to the Boilermakers’ doctrine but on the basis that the power exercised by courts martial is not judicial power of the Commonwealth referred to in Chapter III, but rather disciplinary jurisdiction supported by s 51(vi).96 This position is not unexpected, as courts martial would have been well known to the framers, and it would not have been likely they intended to invalidate them under the new Constitution.97 Courts martial had been subjected to criticism regarding their lack of independence, transparency and conformity with ordinary judicial process, thus undermining the equal rights of service men and women to a fair trial as guaranteed under international conventions.98 In response to these criticisms, in 2007 the Howard Government introduced the Australian Military Court (AMC).99 The AMC largely mirrored the safeguards and procedures of Chapter III courts, but the legislation was very clear that it remained a service tribunal.100 It was decided not to create a Chapter III court to try to retain the benefits of a specialist military body. The AMC was designated a ‘court of record’ made of ‘military judges’.101 By trying to improve the historic courts martial however, the government fell afoul of the first limb of Boilermakers’. In Lane v Morrison the High Court unanimously found that the AMC no longer fit within the historical exception.102 They said this was because, historically, courts martial had not exercised judicial power at all, because their decisions were not definitive of guilt, and their punishments not final. They were always subject to the chain of command.103 In contrast, the power exercised by the AMC, now independent from that chain of command, was dispositive,104 and therefore was judicial power of the Commonwealth. As such, it could not be vested in a service tribunal, even if the AMC now looked more like a Chapter III court than historic courts martial. In fact, it was this very attempt to bring the AMC towards the Chapter III model that led to its demise. French CJ and Gummow J explained the Parliament was attempting to borrow for ‘the AMC the reputation of the judicial branch for impartiality and non-partisanship’.105 After Lane v Morrison the military justice system reverted back to a more traditional form.106 In 2010, the government introduced a Bill that would have created

96 See, eg, Re Tracey; Ex parte Ryan (1989) 166 CLR 518, 598 (Gaudron J). 97 Sawer, ‘The Separation of Powers in Australian Federalism’, above n 24, 180. 98 See, eg, Joint Standing Committee on Foreign Affairs, Defence and Trade Military Justice Procedures in the Australian Defence Force (1999); Senate Foreign Affairs, Defence and Trade References Committee, The Effectiveness of Australia’s Military Justice System, (2005). 99 Defence Legislation Amendment Act 2006 (Cth). 100 Defence Force Discipline Act 1982 (Cth) s 114(1) (repealed). 101 Ibid s 114(1A) and (2) (repealed). 102 (2009) 239 CLR 230. 103 Ibid 257 (Hayne, Heydon, Crennan, Kiefel and Bell JJ); 248 (French CJ and Gummow J). 104 The AMC’s finding and sentence took effect immediately, with no review from the chain of command. 105 Lane v Morrison (2009) 239 CLR 230, 238. 106 Military Justice (Interim Measures) Act (No 1) 2009; Military Justice (Interim Measures) Act (No 2) 2009.

Vol 31(2) Boilermakers’ and the Separation of Judicial Power 279

a new Military Court of Australia as a fully fledged Chapter III court.107 This lapsed, but has been reintroduced in 2012.108 For now, the strictness of Boilermakers’ has created a ‘frozen model of military justice’.109 It has forced the government to introduce a Chapter III Court for military justice if it wishes to develop that system in accordance with the expectations of modern society.110 This undermines the objectives of maintaining a separate military justice system for the purpose of dispensing a parallel system of military discipline as well as justice,111 composed of judges with specialist knowledge of military processes, culture and context, and the ability to be convened overseas during times of hostilities.112

D Rights Protection: Momcilovic v The Queen (2011)

In 2009, the National Human Rights Consultation Committee handed its report on the protection of rights in Australia to the federal government after wide community consultation. Its recommendations included the enactment of a federal Human Rights Act, based on the legislative bills of rights already operating in Victoria113 and the Australian Capital Territory.114 The model proposed by the Committee was what is commonly known as the ‘dialogue model’. One part of the model requires the judiciary to interpret laws consistently with the human rights contained in the statute, subject to inconsistency with the logic of the statute and parliamentary intent. If a law cannot be interpreted consistently with the rights in the statute, the court issues a declaration of incompatibility, notifying the Parliament of the inconsistency. The Parliament must respond to the declaration, but retains the discretion as to whether to amend the incompatible law. The perceived advantage of this model is it retains the supremacy of the Parliament in relation to decisions that may require delicate balancing of rights against other legitimate government objectives. However, recently a majority of the High Court in Momcilovic v The Queen, in the context of the Victorian Charter of Human Rights and Responsibilities Act, found the power to issue a declaration of incompatibility to be non-judicial.115 This was on the basis that it was not part of the disposition of the proceedings, and therefore was non-conclusive in nature.116 While the majority found that Victorian courts could exercise such a power without breaching the Kable principle, the finding precludes federal courts from doing so under the second limb of Boilermakers’. This has significantly narrowed the possibility of the enactment of an exhaustive human rights instrument at the federal level in Australia, and has certainly removed any possibility of a Human Rights Act that retains the supremacy of Parliament through the dialogue

107 Military Court of Australia Bill 2010. 108 Military Court of Australia Bill 2012. 109 Henry Burmester, ‘The Rise, Fall and Proposed Rebirth of the Australian Military Court’ (2011) 39 Federal Law Review 195, 196. 110 I do not necessarily think this is a negative outcome, but the Boilermakers’ decision has removed any flexibility for the government and the Parliament to establish a tribunal that has the necessary safeguards of independence and impartiality that courts provide, but may be able to better understand the military context of the matters that will come before it. 111 Burmester, above n 109, 210-11. 112 These reasons can be found in Commonwealth, Government Response to the Senate Foreign Affairs, Defence and Trade References Committee ‘Report on the Effectiveness of Australia’s Military Justice System’ (2005) 16. 113 Charter of Human Rights and Responsibilities Act 2006 (Vic). 114 Human Rights Act 2004 (ACT). 115 (2011) 245 CLR 1. 116 Ibid [89] (French CJ) (with whom Bell J agreed on this point, [661]; [146], [184], [187] (Gummow J) (with whom Hayne J agreed on this point, [280]; [457] (Heydon J); [584] (Crennan and Kiefel JJ);

280 University of Queensland Law Journal 2012

model (although the intent of the model could possibly be achieved without the conferral of the power to issue a declaration).

IV PART III – A LESS IMPERFECT, LESS INCONVENIENT ALTERNATIVE

There have been a number of instances of judicial dissatisfaction with the Boilermakers’ doctrine, emphasising the benefits of a less strict test that focuses on the integrity of the judicial process.117 From outside the court, I am not the first to advocate a ‘reconception’ of the separation of judicial power in Australia.118 Extra-judicially Sir Anthony Mason has advocated the replacement of the second limb of Boilermakers’ with an incompatibility test: ‘where ... non-judicial functions could be reposed in a federal court unless they were incompatible with the functions of a court’.119 However, the test that I propose differs from that proposed by other commentators. These commentators have generally focussed their criticism on the second limb of the Boilermakers’ doctrine that prevents the conferral of non-judicial power on federal courts. I have argued in this article that it is both limbs of Boilermakers’ that suffer imperfections and produce inconveniences. I propose the following test, which is, like Boilermakers’, two limbed, but like that of Mason, focussed on incompatibility with the judicial process:

Limb 1: Non-judicial bodies cannot be conferred with the federal judicial power of the Commonwealth unless they exhibit the necessary independence, impartiality and status. Limb 2: Chapter III courts (federal and State) cannot be conferred with functions that would be incompatible with their independence, impartiality and status.

The two-limbed incompatibility test focuses on maintaining independence and impartiality of those who exercise judicial power, whether they are federal judicial officers, State judicial officers, or others. The focus on ensuring independence and impartiality reflects both the republican objectives of dispersal of power among the branches and ensuring judicial power is exercised independently, and the federalist position that requires an independent and impartial arbiter to determine the delineation of federal and State powers.

117 See, eg, R v Joske (1974) 130 CLR 87, 90 (Barwick CJ); see also 102 (Mason J); Boilermakers’ (1956) 94 CLR 254, 314-15 (Williams J), Williams J constructed an alternative test of inconsistency built upon the position of the High Court in the earlier case of R v Federal Court of Bankruptcy; Ex parte Lowenstein (1938) 59 CLR 556, 567 (Latham CJ, with whom Rich J agreed); 590 (McTiernan J). 118 Stellios, ‘Reconceiving the Separation of Judicial Power’, above n 24; Sawer, ‘The Separation of Powers in Australian Federalism’, above n 24. See critiques of Else-Mitchell reported at 193-4 and Bailey at 195-6 of this article. See also George Winterton, Parliament, the Executive and the Governor-General: A Constitutional Analysis (Melbourne University Press, 1983) 63, who asserted ‘the eventual overruling of Boilermakers appears to be only a question of time.’ 119 Mason, above n 7, 5. Others share this position: Zines, above n 24, 299; Fiona Wheeler, The Separation of Federal Judicial Power: A Purposive Analysis (PhD Thesis, 1999), 157; Stellios, ‘Reconceiving the Separation of Judicial Power’, above n 24; Winterton, Parliament, the Executive and the Governor-General, above n 118, 63-4; Melissa Perry, ‘Ch III and the Powers of Non-Judicial Tribunals: Breckler and Beyond’ in Adrienne Stone and George Williams (eds), The High Court at the Crossroads: Essays in Constitutional Law (Federation Press, 2000); Gerangelos, above n 8, 11-12, 13ff. Although note the defence of the separation in, for example, Walker, above n 71.

Vol 31(2) Boilermakers’ and the Separation of Judicial Power 281

The formulation of independence and impartiality is brought across from the High Court’s jurisprudence on the limits of the States’ power to undermine the ‘institutional integrity’ of State courts. The Kable principle started as one of incompatibility (similar to that developed in the persona designata jurisprudence), that is, State courts cannot be given powers that are incompatible with their status as receptacles of federal judicial power under Chapter III.120 The focus of the doctrine has changed over time, and the touchstone is now whether the States have impaired the institutional integrity of the court. In North Australian Aboriginal Legal Aid Service Inc v Bradley the focus of this inquiry shifted to determining whether a court that was capable of exercising federal judicial power is, and appears to be, ‘an independent and impartial tribunal’.121 In Baker v Commonwealth, Keane CJ and Lander J recently applied a judicial test of independence and impartiality to federal courts beyond its previous application as part of the persona designata doctrine.122 The case involved a challenge by a number of Federal Magistrates to their exclusion from the definition of a ‘judge’ in the Judges’ Pensions Act 1968 (Cth). Instead of receiving a judicial pension under that Act, Federal Magistrates receive contributions to their superannuation fund by the Commonwealth.123 One of the arguments made by the Magistrates was that the Constitution requires that justices appointed under Chapter III of the Constitution be, and be seen to be, independent of the other branches of government.124 Implicit in this argument is that it is not enough that the judges are appointed in compliance with s 72 of the Constitution, previously thought to be the touchstone of judicial independence in the federal courts. In my proposed tests I have added a criterion not seen in the State jurisprudence or the persona designata tests: the idea of ‘status’. Status requires consideration of the relative standing of the body as against the nature of the power conferred. By considering the standing of the institution, it also imports an examination of the processes adopted by it, thus ensuring the manner of the exercise of judicial power is not inappropriately interfered with.125 As with the concepts of ‘independence and impartiality’, this idea strongly correlates to the judicial test used to determine whether the Kable principle has been breached at the State level. In the recent State Chapter III cases,126 the High Court has focussed on whether incursions into the judicial process undermine the institutional integrity of the State courts, and has emphasised the importance of principles such as procedural fairness, the open court principle, and the requirement to give reasons.127 Having status as part of the test means, for example, the Anti-Discrimination Tribunal of Tasmania could be conferred with jurisdiction to finally determine disputes

120 Kable v Director of Public Prosecutions (NSW) (1996) 189 CLR 51 (‘Kable’), 105-6; 107-8 (Gaudron J); 116; 124 (McHugh J); 128, 144 (Gummow J). 121 (2004) 218 CLR 146, 163 (McHugh, Gummow, Kirby, Hayne, Callinan and Heydon JJ). 122 [2012] FCAFC 121. 123 This entitlement was established by the Governor-General pursuant to the Federal Magistrates Act sch 1, cl 8(1). 124 [2012] FCAFC 121, [28] (Keane CJ and Lander J). 125 George Winterton, ‘The Separation of Judicial Power as an Implied Bill of Rights’ in Geoffrey Lindell (ed) Future Directions in Australian Constitutional Law: Essays in Honour of Professor Leslie Zines (Federation Press, 1994) 185. 126 International Finance Trust Co Ltd v New South Wales Crime Commission (2009) 240 CLR 319 (‘International Finance Trust Co’); South Australia v Totani (2010) 242 CLR 1 (‘Totani’); Wainohu v New South Wales (2011) 243 CLR 181 (‘Wainohu’). 127 International Finance Trust Co (2009) 240 CLR 319, 354–5 [54]–[56] (French CJ); 367 [98] (Gummow and Bell JJ); 379–80 [141], 385 [155] (Heydon J); Totani (2010) 242 CLR 1, 43 [62], 49 [72] (French CJ); 62–3 [131] (Gummow J), 162 [443] (Kiefel J); Wainohu (2011) 243 CLR 181, 208 (French CJ and Kiefel J); 229-30 (Gummow, Hayne, Crennan and Bell JJ).

282 University of Queensland Law Journal 2012

that arise between individuals and the Commonwealth under the Anti-Discrimination Act 1998 (Tas). The ADT’s status as a quasi-independent tribunal relative to the power conferred (which while it may be a final adjudication it does not include, for example, the adjudication of criminal guilt or the power to order detention) is reasonably related. It also adopts an appropriately independent process, one that conforms to the requirements of natural justice,128 the open court principle,129 and the requirement to give reasons on request,130 to afford it the necessary status to exercise this type of power. The focus on maintaining the independence and impartiality of the judicial process but allowing the limited conferral of federal judicial power on non-judicial bodies will overcome the inconveniences that arose after the Brandy decision. It will allow the conferral of judicial power on tribunals that have many characteristics of a court but not sufficient characteristics as to make them ‘federal courts’ for the purpose of the Constitution because they may not satisfy s 72 of the Constitution (including, for example, the Inter-State Commission). It would also allow for the creation of a military court that does not satisfy all of the requirements of Chapter III, but meets contemporary standards of independence, impartiality and status. The first limb accords with British history in which many institutions which may not meet the strict requirements of s 72 of the Constitution were vested with judicial power without concern. Sawer argued for example that the Commonwealth Court of Conciliation and Arbitration, while found not to be a ‘court’ in Boilermakers’, by English standards of the nineteenth century, would have been readily accepted as such.131 The first limb of my proposed test also accords with the framers use of State courts as vessels capable of exercising judicial power. John Quick and Robert Garran noted that the autochthonous expedient was made acceptable to the framers because of ‘[c]onfidence in the integrity and impartiality of the Bench’, making ‘it possible to contemplate without misgiving the exercise of federal jurisdiction by State courts’.132 This was of course despite the absence of constitutional guarantees of tenure in State courts that mirrored those granted by s 72 of the Constitution. The other advantage of the proposed test is that it will streamline the tests among the federal and State judiciary, which accords strongly with McHugh and Gaudron JJ’s assertion in Kable that in our federation there are no two grades of justice.133 My alternative two-limbed test introduces flexibility, overcomes practical inconveniences associated with the ‘pure or strict’ strict separation of judicial power doctrine,134 and avoids the need of relying on legal fallacies such as persona designata to do so. It also allows the federal judiciary to be vested with powers such as state judicial power (under a cross-vesting regime) or the power to issue declarations of incompatibility under a dialogue model of a human rights instrument. The test facilitates greater efficiency in government,135 although recognises that government efficiency is not an absolute objective, maintaining safeguards for the independence, impartiality and status of bodies exercising judicial power. It would continue to prevent such incursions into good government as a Bill of Attainder; the interference in judicial processes by the political branches of government; the engagement by the judiciary in political processes, or those who exercise judicial power from doing so.

128 Anti-Discrimination Act 1998 (Tas) s 87(4)(a). 129 Ibid s 85. 130 Ibid s 93. 131 Sawer, ‘The Separation of Powers in Australian Federalism’, above n 24, 179-80. 132 Quick and Garran, above n 13, 804. 133 Kable v DPP (NSW) (1996) 180 CLR 51, 105 (Gaudron J); 115 (McHugh J). 134 R v Trade Practices Tribunal; ex parte Tasmanian Breweries (1970) 123 CLR 361, 389. 135 Sawer, ‘The Separation of Powers in Australian Federalism’, above n 24, 177.

Vol 31(2) Boilermakers’ and the Separation of Judicial Power 283

Under the two-limbed incompatibility test, each of the imperfections of Boilermakers’ identified in Part I of this article is corrected; and each of the inconveniences identified in Part II of the article could be avoided. The flexible, non- absolute nature of the test will also allow for the evolution of the separation of powers in accordance with changing legal and social expectations. The separation of powers is a principle that has been in evolution for centuries since its first articulation, and it is not true to its history for it to be applied using inflexible rules. That is not to say that my alternative test is perfect and is always going to produce convenient and efficient results. The two-limbed incompatibility test has its own imperfections. The first limb is inconsistent with the strong implication from s 71 of the Constitution that the judicial power of the Commonwealth is only vested in the High Court, federal courts and state courts (although note my criticism of this implication above). The capacity of non-judicial tribunals, that is, bodies that are not courts, to exercise federal judicial power also appears contrary to the intention of s 73 of the Constitution. This sets out the appellate jurisdiction of the High Court, which gives it jurisdiction to hear appeals from (i) any justice or justices exercising the original jurisdiction of the High Court; (ii) any other federal court, or court exercising federal jurisdiction; or of the Supreme Court of any State, or of any other court of any State from which at the establishment of the Commonwealth an appeal lies to the Queen in Council; (iii) the Inter-State Commission, but as to questions of law only. Appeals do not lie from non-judicial bodies exercising federal jurisdiction. Allowing such bodies to exercise federal jurisdiction may thus undermine the intention of the framers, through s 73, to ensure there was always an avenue of appeal from the exercise of federal jurisdiction to the High Court as a ‘General Court of Appeal’.136 It has the potential to undermine the creation by the Court of the uniform common law of Australia.137 The inclusion of the Inter-State Commission in s 73(iii) strengthens this implied intention, preventing ‘any exception being made to that uniform interpretation of the law which it is the aim of the Constitution to ensure’.138 However, it must be remembered that s 73 anticipates exceptions to the appellate jurisdiction, its opening words state the High Court’s appellate jurisdiction is ‘with such exceptions and subject to such regulations as the Parliament prescribes’. Further, federal tribunals vested with federal judicial power will be subject to the guaranteed oversight of the High Court under s 75(v), and State tribunals exercising federal jurisdiction will be subject to the oversight of the State Supreme Courts after the decision in Kirk v Industrial Court of New South Wales.139 The federal Parliament would also have the power to vest federal judicial power in non-judicial tribunals (State and federal) subject to appeals to a federal court.140 Alternatively, as occurred in Brandy, the final order of the tribunal may be deemed to be the order of a federal court, thus allowing for appeals to lie through the ordinary court hierarchy. My proposed two-limbed incompatibility test also suffers from imperfections in application. The Privy Council in Boilermakers’ dismissed incompatibility as a criterion that was ‘vague and unsatisfactory’.141 Gerangelos points out the problem with any form of functionalism in the separation of powers doctrine:

136 Quick and Garran, above n 13, 735. 137 Lange v Australian Broadcasting Corporation (1997) 189 CLR 520, 563. 138 Quick and Garran, above n 13, 743. 139 (2010) 239 CLR 531. 140 Stellios, The Federal Judicature, above n 54, 379. 141 Attorney-General (Cth) v The Queen (1957) 95 CLR 529, 542.

284 University of Queensland Law Journal 2012

[T]he problem with functionalism still remains that of providing sufficiently objective criteria by which to determine the limits to be imposed on the intermingling of functions which it allows.142

So, just like the definition of judicial power in Boilermakers’, the two-limbed incompatibility test requires the tackling of difficult, ill-defined concepts and standards. However, Zines argues that an incompatibility test is preferable to the strictness of Boilermakers’ and its reliance on the definition of judicial power:

The incompatibility doctrine has the merit of focussing attention directly on the policy of the separation of powers and on the actual functions conferred in the context of the Court’s primary judicial powers and functions. The mind is directed (rightly in my view) to the reasons for having a separation of powers rather than to the issue of mere classification.143

The High Court has ever-broadening experience in applying the criterion of incompatibility. It is the touchstone for constitutionality under the persona designata doctrine, and also for the constitutionality of conferring non-judicial power on State courts under the Kable principle,144 which has enjoyed somewhat of a resurgence in the last few years.145 If Baker v Commonwealth is correct, the judiciary will also have to consider the test more broadly in the federal context. Accordingly, my two-limbed incompatibility test focuses attention on what really gives confidence in the judicial process rather than the formalistic protections of independence given by s 72. Section 72 may provide a normative statement of independence, but it is not what guarantees independence at a practical level. Indeed, it is highly unlikely that litigants, when choosing the forum in which to bring a case, actively consider the ‘additional guarantees’ of tenure and remuneration enjoyed by federal judges over their State counterparts. There is increasing acceptance that the independence of a public office is not guaranteed by life tenure. In New South Wales the previous life tenure of crown law officers, including prosecutors and the Solicitor-General, was removed by statute in 2007.146 The Opposition opposed the move, claiming that limited terms and the desire to obtain reappointment would leave officeholders vulnerable and undermine the independence of the office.147 Greg James, whose review had formed the basis of the amending legislation, had anticipated these arguments, and asserted that other mechanisms sufficiently protected the independence of these officeholders, including professional and ethical obligations relating to independence and detachment, and protection against improper dealings by government through the variety of integrity offices including, for example, the Ombudsman.148 The list of other factors that protect independence closely mirrors those mechanisms that the Commonwealth argued protected the independence and impartiality of the Federal Magistrates in Baker v Commonwealth. While one was the constitutional protections of tenure, they also included:

142 Gerangelos, above n 8, 13. 143 Zines, above n 24, 299. 144 First seen in Kable v DPP (NSW) (1996) 180 CLR 51. 145 See further Gabrielle Appleby and John Williams ‘A New Coat of Paint: Law and Order and the Refurbishment of Kable’ (2012) 40 Federal Law Review 1. 146 Crown Law Officers Legislation Amendment (Abolition of Life Tenure) Act 2007 (NSW). 147 New South Wales, Parliamentary Debates, Legislative Council, 23 October 2007, 3030 (John Ajaka and Lee Rhiannon). 148 As quoted in New South Wales, Parliamentary Debates, Legislative Council, 23 October 2007, 3030 (John Ajaka); access to the full report was sought through a freedom of information request but it was denied on the basis of cabinet confidentiality.

Vol 31(2) Boilermakers’ and the Separation of Judicial Power 285

[T]he judicial oath, the culture of professionalism and ethics of Australian lawyers and judges, the open-court principle, the requirement that judges provide reasons for their decision, the susceptibility of inferior courts to superior court review, the apprehended bias rule, the fact that salary and related benefits are determined by an independent tribunal, the fact that salaries and superannuation are well above community norms and the existence of professional rules which regulate a return to practice from the judiciary.149

The focus on the professional obligations and ethics of officeholders as legal practitioners guaranteeing independence shifts the focus of independence guarantees from formalistic statutory provisions, which are often easily undermined by those intent on doing so or by subtle influences, to the integrity and capacity of the individuals involved. Hence my two-limbed incompatibility test requires an inquiry into whether the necessary standards of independence, impartiality and status are met, an inquiry relevant for other accountability mechanisms (for example, the Auditor-General, the Australian Human Rights Commission, the Information Commissioner, the Administrative Appeals Tribunal, the Ombudsman and the Australian Crime Commission). The preoccupation with the separation of judicial power as the most fundamental check on executive and parliamentary power has meant that other mechanisms of accountability are often undervalued, understudied and under-provided for in regulatory regimes.150 Boilermakers’ preoccupation with the judiciary has come at the expense of considering the matrix of other institutions and mechanisms that operate to check and bring transparency and accountability to executive and legislative action.151 A recent example illustrates the current preoccupation at the expense of other important mechanisms. In 2012, allegations were made by a federal Member of Parliament that an investigation into his conduct by Fair Work Australia (FWA), an independent tribunal established under the Fair Work Act 2009 (Cth), was tainted by misconduct on behalf of one of its members. The President of FWA, Justice Iain Ross of the Federal Court, argued he had no power to investigate or handle complaints against FWA members.152 He called for the tribunal to be covered by the Courts Legislation Amendment (Judicial Complaints) Act 2012, a new federal system to deal with complaints against judicial officers. At present, tribunals are not covered by the scheme, nor are they covered by schemes aimed at ensuring integrity and accountability in the administrative arm of the state such as the Ombudsman or the Public Service Commission. As with many of the accountability mechanisms of the modern administrative state, the FWA sits between the executive and the judiciary and has therefore been overlooked by those schemes directed at one or the other. Under my proposed test, the judiciary becomes one part of a larger whole devoted to accountability. The shared problems amongst these mechanisms to ensure independence and impartiality, including questions about handling complaints,

149 Baker v Commonwealth [2012] FCAFC 121, [34]. The judgment of Keane CJ and Lander J (at [51]) accept the proposition that many mechanisms exist to ensure judicial independence, quoting from the judgment of Gummow, Hayne and Crennan JJ in Forge v ASIC (2006) 228 CLR 45, [84]-[85]. 150 See, eg, similar comments in John McMillan, ‘Re-Thinking the Separation of Powers’ (2010) 38 Federal Law Review 423. 151 See broader conceptions of public and constitutional law in Martin Loughlin, The Idea of Public Law, (Oxford University Press, 2003) 44; Mark Tushnet, A New Constitutional Order (Princeton University Press, 2003) 1; Suri Ratnapala, Australian Constitutional Law: Foundations and Theory (Oxford University Press, 2007) 4. 152 See Senate Estimates Committee, Statement of Iain Ross (28 May 2012) [29].

286 University of Queensland Law Journal 2012

allocation of funding,153 the proper selection of candidates (to ensure the necessary individual integrity and capacity),154 and appropriate accountability mechanisms, can be addressed coherently. Practice and ultimately convention can develop together, strengthening all of the institutions. Some argue that the strict separation of judicial power is good constitutional practice, or constitutional prudence. Gerangelos explains:

If the rigours of this approach are ameliorated, even for the best of policy reasons, these values will be threatened by the gradual yet inexorable erosion of the essential boundaries, even though this may not be apparent in a particular case.155

Thus the strictness and inflexibility of Boilermakers’ ensures against any potential temptation to undermine the independence and integrity of the judicial branch and the exercise of the judicial power. Two observations can be made about that slippery slope argument.156 First, if we accept that the only way to ensure independence is to remove any potential temptation to undermine it, we must then accept that the more contemporary accountability mechanisms, which are often used instead of the courts (and such use is increasingly encouraged by the government), lack this safeguard. Secondly, whilst the Boilermakers’ doctrine focuses on institutional integrity and separation, there is no constitutional safeguard in relation to the individual competency and capacity of judicial officers.157 Structural and institutional mechanisms that focus on independence at this more general level fail to consider the potential of subtle influences on individuals, or in the more extreme cases, the conscious efforts of individual members of the judiciary. If constitutional prudence is the question, Boilermakers’ is not a complete answer either.

V CONCLUSION

I, like other critics of the Boilermakers’ doctrine, understand and endorse the raison d’être of the separation of judicial power. But, as a human construct, its current formulation is imperfect and inconvenient. My proposal, for its replacement with a two-limbed incompatibility test, is perhaps advocating replacing one imperfect and inconvenient human construct for another. I would conclude by observing that it is also the natural lot of mortals to strive for betterment.

153 Stephen Parker, ‘The Independence of the Judiciary’ in Brian Opeskin and Fiona Wheeler (eds) The Australian Federal Judicial System (Melbourne University Press, 2000) 62, 79-81. 154 Ibid 89. 155 Gerangelos, above n 8, 3. See also Walker, above n 71, 163. 156 Referred to as such in Saunders, above n 8, 201. 157 Sawer, ‘The Separation of Powers in Australian Federalism’, above n 24, 177.