Commonwealth v Hospital Contribution Fund of Australia

The matter at issue in the proceedings in the Supreme Court, being one in which a person being sued on behalf of the Commonwealth is a party, is one of a kind in which the High Court has original jurisdiction under s. 75(iii) of the Constitution. The Supreme Court of New South Wales was accordingly invested with federal jurisdiction in the matter by virtue of s. 39(2) of the Judiciary Act which, subject to an exception and to conditions and restrictions that it is unnecessary for present purposes to mention, provides, inter alia, as follows:

The several Courts of the States shall within the limits of their several jurisdictions, whether such limits are as to locality, subject-matter, or otherwise, be invested with federal jurisdiction, in all matters in which the High Court has original jurisdiction or in which original jurisdiction can be conferred upon it, . This provision was of course enacted pursuant to the power given to the Parliament by s. 77(iii) of the Constitution which provides as follows: With respect to any of the matters mentioned in the last two sections the Parliament may make laws -

(iii) Investing any court of a State with federal jurisdiction.

The Solicitor-General for the Commonwealth contended that the claim of the Commonwealth to privilege constituted the relevant matter that fell to be determined by the master in the exercise of federal jurisdiction. It is in my opinion unnecessary to consider whether a State court would be exercising federal jurisdiction if it were called upon to decide whether a subpoena directed to a Minister or officer of the Commonwealth should be set aside on the ground of Crown privilege, in a case not otherwise involving the exercise of federal jurisdiction. An example of such a case would be provided by an ordinary action for damages for negligence, brought by one resident of a State against another, in which one party wished to direct a subpoena to an officer of the Department of Social Security with a view to obtaining the production of documents disclosing the amount of an invalid pension paid to the plaintiff, and in which the officer claimed Crown privilege. It would be surprising if part of an action, otherwise within State jurisdiction, became transmuted into federal jurisdiction simply because a stranger to the action, subpoenaed as a witness, made a claim to privilege. But this question does not arise for decision here, because the matter at issue in the action instituted by H.C.F. can only be determined in the exercise of federal jurisdiction.

The argument on behalf of the Commonwealth was clear and simple: s. 39(2) of the Judiciary Act invests only a court of a State with federal jurisdiction; the master, although an officer of the Supreme Court of New South Wales, was not a part of that Court; therefore he was not invested with federal jurisdiction. The decisions of this Court in Kotsis v. Kotsis [1] and Knight v. Knight [2] were relied upon in support of that submission. In reply, the State of New South Wales submitted first, that under the Supreme Court Act 1970 N.S.W., as amended ("the Supreme Court Act "), a master exercising the powers of the Supreme Court under that Act and the Supreme Court Rules, as Master Allen was in the present case, does constitute the court, and, alternatively, that Kotsis v. Kotsis and Knight v. Knight were wrongly decided and that the reference to Courts in s. 39(2) of the Judiciary Act, and to court in s. 77(iii) of the Constitution, extends to the organization through which the jurisdiction of the court is exercised.

1. (1970) 122 C.L.R. 69. 2. (1971) 122 C.L.R. 114.

In New South Wales, masters are appointed by the Governor under s. 111 of the Supreme Court Act. The powers and functions of a divisional master - a master assigned to a Division of the court, as Master Allen was - are set out in s. 118 of the Supreme Court Act. The provisions of sub-ss. (2), (3), (4) and (5) of that section are as follows:

(2) A divisional master may exercise such powers of the Court in the Division as are, by or under this or any other Act, conferred upon a master assigned to the Division.

(3) A judgment directed to be entered or an order made by a divisional master in any Division may be set aside or varied by the Court.

(4) Subject to subsection three of this section, a judgment directed to be entered or an order made or direction given by a divisional master in any Division shall have effect as a judgment, or order or direction of the Court in the Division, whether or not the direction or order is within the powers mentioned in this section of the divisional master.

(5) A divisional master in any Division shall constitute the Court in that Division for the purpose of the exercise of the powers mentioned in subsection two of this section.

Similar provisions appear in s. 117A which deal with the powers of the master in the Court of Appeal. Section 48(4)(b), which deals with the powers of a master in relation to punishment for contempt in the face of the court or in the hearing of the court, or for breach of an undertaking given to the court, speaks of the case "where the Court is constituted by that master". On the other hand s. 25 provides: The Court shall be composed of a Chief Justice, a President of the Court of Appeal and such other Judges of Appeal and Judges as the Governor may from time to time appoint. Section 40 provides:

(1) All proceedings in any Division and all business arising out of proceedings in a Division shall be heard and disposed of before a Judge, who shall constitute the Court.

(2) Subsection one of this section does not affect the provisions of this Act and the rules concerning the hearing and disposal of proceedings and business before a master or before a registrar or other officer of the Court.

The proper conclusion to be drawn from these provisions is that the master is not a competent part of the court, in the sense that a judge is, but that he constitutes the court for the purpose of the exercise of the powers conferred upon him - that is, of course, not for all purposes. If Kotsis v. Kotsis and Knight v. Knight are correctly decided, a master in New South Wales cannot be invested with federal jurisdiction, because, according to those decisions, only a person who is a member of a State court may be invested with federal jurisdiction, and because although the master constitutes the court for limited purposes he is not a member of the court. The position may be different in Victoria and Western Australia, where the relevant statutes provide that the Supreme Court shall consist, not only of the judges, but also of the masters: see s. 75 of the Constitution Act 1975 Vict. and s. 7 of the Supreme Court Act 1935 W.A., as amended. In South Australia s. 7 of the Supreme Court Act 1935 S.A., as amended, provides that the court shall be constituted of a Chief Justice, the puisne judges and the masters and that provision also appears to make the masters part of the court. However, the question would remain whether in those States a Supreme Court whose membership is expanded to include masters would be a court within ss. 77(iii) and 39(2). Some of the reasoning in Kotsis v. Kotsis and Knight v. Knight would require a negative answer to that question.

In Kotsis v. Kotsis [3] this Court held that a registrar of the Supreme Court of New South Wales did not have power to make an order for payment of interim costs in a matrimonial cause. By s. 23(2) of the Matrimonial Causes Act 1959 Cth, as amended, which, like s. 39(2) of the Judiciary Act, was enacted pursuant to the power given by s. 77(iii) of the Constitution, "the Supreme Court" was invested with federal jurisdiction to hear and determine matrimonial causes. All the members of the Court held that the true effect of the laws of New South Wales under which the Supreme Court was constituted and registrars were appointed was that the registrar was not a member of the court [4] . The majority of the Court (Barwick C.J., and McTiernan, Menzies, Windeyer, Owen and Walsh JJ.) held that the registrar was not invested with federal jurisdiction. The case was decided on the effect of s. 23(2) of the Matrimonial Causes Act, but it followed from the reasoning, and in some instances was expressly said, that the constitutional provision had a similar effect, and that it was not constitutionally possible to invest the registrar with federal jurisdiction. I was unable to share the view of the majority. Within a few months in Knight v. Knight [5] a similar question arose, viz. whether the federal jurisdiction invested in the Supreme Court of South Australia by s. 23(2) of the Matrimonial Causes Act might be exercised by a master of that Court. It was held that under the legislation then in force in South Australia (which, as I have already indicated, has since been amended) the Supreme Court consisted only of judges and not of masters, and that accordingly a master could not exercise the federal jurisdiction invested in the court. I agreed with that decision, regarding myself as bound by Kotsis v. Kotsis.

1. (1970) 122 C.L.R. 69. 2. (1970) 122 C.L.R., at pp. 75, 81, 86, 90, 95, 97, 101, 103. 3. (1971) 122 C.L.R. 114.

The question then arises whether we should reconsider these decisions, and refuse to follow them if we disagree with them. Clearly we have power to take that course. This Court has always claimed the power to overrule its own decisions, and has exercised that power in a number of cases of great constitutional importance. But it is a power to be exercised with restraint, and only "after the most careful scrutiny of the precedent authority in question and after a full consideration of what may be the consequences of doing so": Queensland v. The Commonwealth [6] , per Stephen J. The authorities that have considered the circumstances in which this Court will reconsider an earlier decision of its own were fully discussed in the judgment of Aickin J. in Queensland v. The Commonwealth [7] . There are some general considerations that will assist the Court in deciding whether the previous decision should be overruled, as Aickin J. pointed out, but in the end the Court is required to make a discretionary judgment, and to do so with the caution, and the sense of responsibility, that the gravity of the matter requires.

1. (1977) 139 C.L.R. 585, at p. 602. 2. (1977) 139 C.L.R., at pp. 620-630.

In the present case there are a number of circumstances that have led me to the conclusion that we should reconsider the correctness of Kotsis v. Kotsis and Knight v. Knight. It should be unnecessary to say that I am not about to recite a list of grounds any one of which, standing alone, would provide a justification for overruling an earlier decision. All the circumstances have to be considered, but for obvious reasons it is convenient to refer to some of them separately. In the first place, the decisions do not rest upon a principle that has been carefully worked out in a succession of cases. Although their correctness was perhaps assumed in Reg. v. Humby; Ex parte Rooney [8] , that assumption had nothing to do with the result in that case. They stand alone and to overrule them will not unsettle the law in other respects. Indeed, the decisions are in my respectful opinion opposed to the principle, established in Le Mesurier v. Connor [9] , that when the Parliament invests a State court with federal jurisdiction it has no power to alter or affect the constitution of the court or the organization through which its jurisdiction and powers are exercised; see also Adams v. Chas. S. Watson Pty. Ltd. [10] ; Peacock v. Newtown Marrickville and General Co-operative Building Society No. 4 Ltd. [11] ; and Russell v. Russell [12] . Secondly, there is an important difference between the reasoning of Barwick C.J. and that of the other justices who constituted the majority in Kotsis v. Kotsis. Barwick C.J., who spoke of s. 77(iii) as though it referred to "Supreme Courts" rather than simply to "courts", expressed the opinion [13] , that it is not entirely for a State to determine of whom the Supreme Court may consist for the exercise of federal jurisdiction. He said [14] : "If State law were to change this composition" [i.e., the composition of the Supreme Court] "in a radical way what the State continued to call the Supreme Court would not satisfy, in my opinion, the references in the Constitution to the Supreme Courts of the States." If Barwick C.J. intended to suggest that the reference to court in s. 77(iii) of the Constitution is only to a court as it was constituted before the commencement of the Constitution I am unable, with all respect, to accept the correctness of his suggestion. It is well established that the provisions of s. 39(2) of the Judiciary Act are ambulatory and operate upon State courts whether constituted before or after the commencement of the Judiciary Act: see The Commonwealth v. District Court of the Metropolitan District [15] , and Collins v. Charles Marshall Pty. Ltd. [16] . In the latter of those cases the court in question was established in 1937 - long after the commencement of the Constitution - but it was nevertheless held to be invested with federal jurisdiction. In Kotsis v. Kotsis, Menzies J. [17] stated the position in a way with which I would respectfully agree when he said that the Parliament "can do no more than invest a State Court, as it is or as it may become, with federal jurisdiction." Thirdly, the decisions in Kotsis v. Kotsis and Knight v. Knight achieve no useful result, but on the contrary lead to considerable inconvenience. On the one hand, to exclude the officers of a State court, who regularly exercise its powers, from the investiture of federal jurisdiction does nothing to protect the freedom of the individual or the independence of the judiciary, since a court composed of laymen, with no security of tenure, might effectively be invested with jurisdiction under s. 77(iii). On the other hand, to require the State courts to depart from their established organization in dealing with matters of federal jurisdiction has many disadvantages. I have expanded upon this aspect of the matter in Kotsis v. Kotsis [18] . Finally, a very important circumstance that led Stephen J. and myself in Queensland v. The Commonwealth [19] to refuse to overrule an earlier decision with which we disagreed was that the decision had been acted upon, and that a consequence of overruling it would be to deny representation in the Senate to inhabitants of the Territories to whom it had been conceded by the earlier decision. In the present case, the decisions in Kotsis v. Kotsis and Knight v. Knight have, it is true, in one sense, been acted upon, in that some States have enacted legislation for the obvious purpose of enabling masters to be invested with federal jurisdiction. However, the States have acted upon the decisions only for the purpose of endeavouring to circumvent them. No one will be adversely affected if the decisions are overruled.

1. (1973) 129 C.L.R. 231. 2. (1929) 42 C.L.R. 481. 3. (1938) 60 C.L.R. 545, at p. 554. 4. (1943) 67 C.L.R. 25, at p. 37. 5. (1976) 134 C.L.R. 495, at pp. 505, 515-517, 530, 535. 6. (1970) 122 C.L.R., at p. 75. 7. (1970) 122 C.L.R., at p. 77. 8. (1954) 90 C.L.R. 13, at pp. 20-21. 9. (1955) 92 C.L.R. 529, at p. 536. 10. (1970) 122 C.L.R., at p. 88. 11. (1970) 122 C.L.R., at p. 110. 12. (1977) 139 C.L.R. 585.

Having carefully reconsidered the question, I adhere to the views I expressed in Kotsis v. Kotsis. I need not repeat what I there said, but would add some further observations. The question is whether court in s. 77(iii) and Courts in s. 39(2) mean the persons of whom a court is composed (in the present case, the judges), or a court as an institution, "an organization for the administration of justice, consisting of judges and with ministerial officers having specified functions", to use the words of Windeyer J. in Kotsis v. Kotsis [20] . One would expect invested jurisdiction to be conferred on the court regarded as an entity, rather than on the individual persons who compose its membership, and some passages in the judgments in Silk Bros. Pty. Ltd. v. State Electricity Commission (Vict.) [21] , tend to support this view. If the investiture of State courts with federal jurisdiction which is made by s. 39(2) of the Judiciary Act is understood to be an investiture of the persons who compose the courts, the question will arise why that jurisdiction cannot be exercised by all or any of the members of the court, so that for example, a single judge might exercise the jurisdiction normally exercised by the Court of Appeal, or vice versa. The answer of course is that it is clear that the State law must determine the organization through which the powers and jurisdiction of the State courts are exercised. But if that is so, how is it possible to exclude officers of the court who are an integral part of its organization and who, under the State law, exercise its powers and jurisdiction? In Kotsis v. Kotsis Walsh J., who like other members of the majority accepted that the organization of the court is determined by State law, explained as follows his reasons for holding that nevertheless the registrar of the Supreme Court was not invested with jurisdiction [22] :

When it has been said that it is the State law which must determine the organization through which the jurisdiction and powers of the State court are exercised, it is to its jurisdiction and powers that reference has been made. A system created by State law under which some powers are exercised by the court and some are not exercised by it, but by someone else who may sometimes be described as its delegate, cannot be equated with "the organization" by which the powers of the court itself are exercised. With the greatest respect I accept the correctness of the first sentence in the passage that I have cited from the judgment of Walsh J., but not that of the second, at least in its application to the exercise by an officer of the court of powers confided to him by the Act and rules under which the court functions. The jurisdiction and powers of the court do not cease to be its jurisdiction and powers because they are exercised by an officer of the court, under the rules of the court. In the present case the jurisdiction and powers which Master Allen was called on to exercise were undoubtedly the jurisdiction and powers of the Supreme Court. He was the officer of the court by whom the jurisdiction and powers of the court in the matter in question were normally exercised, and an order made by him, if not set aside or varied by the court, would take effect as an order of the court. Although he was not a member of the court he was, in my respectful opinion, part of the organization through which the powers and jurisdiction of the court were exercised in matters of State jurisdiction, and through which they were to be exercised in matters of federal jurisdiction also, once the court was invested with federal jurisdiction. 1. (1970) 122 C.L.R., at p. 91. 2. (1943) 67 C.L.R. 1, at pp. 11, 21. 3. (1970) 122 C.L.R., at p. 99.

For the above reasons, I consider that Kotsis v. Kotsis and Knight v. Knight should be overruled.

I need only add that s. 39(2) of the Judiciary Act contains nothing to suggest that it was intended that the invested federal jurisdiction should be exercised only by judges. I accordingly would hold that the master was invested with federal jurisdiction to decide whether or not the claim to privilege should be allowed and the subpoena should be set aside.

I have had the advantage of reading the reasons for judgment of the Chief Justice, with which I am in agreement. I share the view of the law originally expressed by his Honour in Kotsis v. Kotsis [23] and agree with his Honour's further elaboration of that view in the present case. For the reasons given by his Honour I regard this as a proper case in which both Kotsis v. Kotsis and Knight v. Knight [24] should now be overruled.

1. (1970) 122 C.L.R. 69. 2. (1971) 122 C.L.R. 114.

In Kotsis v. Kotsis, Windeyer J., one of the majority, acknowledged that the argument (which he ultimately rejected) that court in s. 77(iii) of the Constitution must mean an institution as it exists from time to time, an organization for the administration of justice, consisting of judges and other ministerial officers having specified functions, was "formidable" [25] . And so it is. In my view it is to be preferred to the narrow interpretation which commended itself to the majority and which was later confirmed in Knight v. Knight, namely, that court in s. 77(iii) means the judges and judicial officers who are members of the court and that it does not include the organization and officers through which its powers and jurisdiction are exercised. I would be happy to accept the dissenting judgment of Gibbs J. in Kotsis as a sufficient and convincing statement of the reasons for adopting the broader view but for the circumstance that we are proposing to depart from a constitutional interpretation settled upon in two comparatively recent decisions by the entire Court with Gibbs J. alone dissenting. As it is I shall express my own reasons.

As s. 77(iii) grants power to invest jurisdiction in courts, and does not deal with their composition, structure or organization, it supplies in itself no reason for restricting the content of the expression "any court of a State" to judges or judges and other judicial officers who comprise its membership. A more natural reading is to take the expression in its application to the possible recipients for federal jurisdiction as referring to the curial institutions of the State as they happen to be organized from time to time.

The majority view in Kotsis was partly founded on the proposition that a court traditionally is comprised of judges or judicial officers who are to be distinguished from other officers who form part of the court's organization. This is a well recognized distinction (Reg. v. Davison [26] ) which has importance in the interpretation and application of common form statutory provisions so expressed as to arm "the Court" with specific jurisdiction. The jurisdiction is then exercisable by those persons who comprise the court. But the distinction is by no means an absolute one. The statute may provide that the court is comprised not only by judges but also by masters, registrars or deputy registrars, either generally or for a particular purpose. As we have seen in this very case, some recent State statutes regulating the composition of, and the exercise of jurisdiction by, the State Supreme Courts contain provisions of this kind. The relevant jurisdiction is then exercisable not only by judges but also by the designated officers of the court.

Generally speaking, the Parliament of a State may in the exercise of its plenary legislative power alter the composition, structure, and organization of its Supreme Court for the purposes of the exercise of State jurisdiction. It is in the exercise of this power that provisions of the kind already discussed have been enacted. Chapter III of the Constitution contains no provision which restricts the legislative competence of the States in this respect. Nor does it make any discernible attempt to regulate the composition, structure or organization of the Supreme Courts as appropriate vehicles for the exercise of invested federal jurisdiction. It is therefore sensible and natural to read the expression any Court of a State in s. 77(iii) as referring to State courts in the sense explained by Gibbs J. in Kotsis.

His Honour there observed that the exercise of federal jurisdiction did not call for a curial organization different in kind from that established for the exercise of State jurisdiction [27] . In this situation there is every reason for supposing that the framers of the Constitution intended to arm the Parliament of the Commonwealth with a power to invest federal jurisdiction in a State court as it happened to be organized under State law from time to time. Although the Commonwealth Parliament has no power to alter the structure or organization of State courts, its freedom of action is completely preserved. It has the choice of investing State courts with federal jurisdiction or of establishing appropriate federal courts. Moreover, it may condition the investment of federal jurisdiction on the existence of a suitably structured State court - see, for example, s. 39(2) of the Judiciary Act 1903 Cth, as amended.

To me it seems highly improbable that the framers of the Constitution intended to restrict the power to invest federal jurisdiction in State courts so that the exercise of it was limited to an exercise by judges to the exclusion of any exercise by masters, registrars or deputy registrars, who are officers of State courts. I am unable to perceive any basis in any of the provisions of Ch. III for the implication of such a limitation on the power. For the most part the majority in Kotsis appear to have regarded the word court in s. 77(iii) as the source of an implication based on the traditional distinction between judges and officers forming part of the court organization. This approach, as I have sought to show, insufficiently takes account of the States' legislative competence to alter the structure and organization of State courts.

An exercise of the legislative power conferred by s. 77(iii) imposes an obligation on the State court to exercise the jurisdiction thereby invested. It would indeed be a great inconvenience to the States and their courts if the structure and composition of a State court for the exercise of federal jurisdiction is to differ from that selected by the State for the exercise of its similar non-federal jurisdiction. Yet this is the unfortunate consequence of the Kotsis interpretation. What is more it will have a coercive and restraining influence on the States' competence to organize their courts as they choose. If the States are compelled to employ their judges in the exercise of federal jurisdiction, even in unimportant matters, they may be impelled to use the same organization for the exercise of similar non-federal jurisdiction, to avoid having two organizations or to avoid drawing an unacceptable distinction between federal and non-federal business. As we know the exercise of federal jurisdiction may suddenly intrude into the exercise of non-federal jurisdiction without the court or the parties perceiving that a federal element has arisen. The consequence is that if the States adopt a curial organization for the despatch of non-federal business that differs from the Kotsis specification the intrusion of a federal element in the exercise of what is thought to be non-federal jurisdiction could make the hearing abortive. There is no reflection in the provisions of Ch. III of an intention to submerge the exercise of federal jurisdiction in problems of this kind or of an intention to constrain the States' freedom of action in the organization of their courts.

Although s. 77(iii) makes no mention of the State Supreme Courts, the majority judgments in Kotsis, notably that of Barwick C.J., deal with them to the virtual exclusion of other State courts. Indeed, Barwick C.J. [28] erroneously speaks of the reference in s. 77(iii) to the State Supreme Courts. The absence of any reference to particular State courts in s. 77 makes it impossible in my view to conclude that the provision contemplated particular State courts with the characteristics which they possessed in 1900. It is evident that s. 77(iii) empowers the Parliament to invest federal jurisdiction in a State court created after 1900. It is just as evident that the section permits the investment of federal jurisdiction in a State court whose characteristics, either in point of composition or organization, have been altered by State legislation after 1900.

Barwick C.J. drew attention to the reference in s. 73(ii) to the Supreme Courts of the States and to the provision which it made for appeals from those courts to this Court. This reference is quite inadequate to sustain the conclusion that a Supreme Court which is composed of masters, registrars and deputy registrars as well as judges ceases, or may cease to be, a Supreme Court. I can conceive that a State Parliament may so radically alter the structure of its system of courts that there is a difficulty in determining what is the Supreme Court of a State for the purpose of s. 73(ii). But I have no doubt that, notwithstanding such an alteration, a court or courts will be found in the particular State to answer the constitutional description. And in any case there is no reason for us to carry over from s. 73(ii) the hypothetical problems which may there arise to s. 77(iii) with a view to influencing its construction in a fashion certainly not contemplated by its language.

In Kotsis the spectre was raised of appeals being brought from orders made by masters, registrars or deputy registrars directly to this Court by virtue of s. 73(ii). It is not a prospect to be contemplated with equanimity. This Court would not be a convenient forum for such appeals. And s. 73 enables Parliament to prescribe appropriate exceptions to the right of appeal for which the section otherwise provides. The proviso would not inhibit an exclusion of appeals from orders made by persons other than judges.

The majority interpretation carries over into s. 77 the notion, derived from ss. 71 and 72, that the exercise of Commonwealth judicial power must be reposed in judges. Yet ss. 71 and 72 deal with the composition of this Court and the appointment and removal of justices of this Court and judges of other federal courts only. So much at least was recognized in Waterside Workers' Federation of Australia v. J. W. Alexander Ltd. [29] , where it was decided that the requirement, not expressed by the Constitution, but distilled by a process of judicial interpretation, that judges of federal courts be appointed for life had no application to State courts exercising federal jurisdiction. See also Shell Co. of Australia Ltd. v. Federal Commissioner of Taxation [30] ; Silk Bros. Pty. Ltd. v. State Electricity Commission (Vict.) [31] .

1. (1918) 25 C.L.R. 434. 2. (1930) 44 C.L.R. 530. 3. (1943) 67 C.L.R. 1, at pp. 9-10, 21.

The proposition that the exercise of Commonwealth judicial power by the High Court and other federal courts must be undertaken by judges alone is likewise not expressed by the Constitution. It has the support of some dicta - see Le Mesurier v. Connor [32] - and the decision of Long Innes J. in Re Grant; Ex parte Edgley [33] . But in Davison the Court did not decide the question. The decision there turned on the unique situation of the registrar and deputy registrars in bankruptcy. Kitto J. described them as officers of the executive government whose duties were prescribed by the Attorney-General or the Governor-General [34] . Dixon C.J. and McTiernan J. said that they were not officers of the court but functionaries placed under its control [35] , having earlier stated that it "must not be supposed" that they "are officers of the Federal Court of Bankruptcy or form part of the staff or organization of the court" [36] . Their Honours merely referred to the earlier dicta and to Grant without expressing a view as to their correctness [35] .

1. (1929) 42 C.L.R. 481, at pp. 511, 512, 522-525. 2. (1928) 29 S.R. (N.S.W.) 31. 3. (1954) 90 C.L.R., at p. 380. 4. (1954) 90 C.L.R., at p. 365. 5. (1954) 90 C.L.R., at p. 362. 6. (1954) 90 C.L.R., at p. 365.

Section 71 vests the judicial power in the High Court, the federal courts and such courts as may be invested with federal jurisdiction. It also provides that the High Court shall consist of a Chief Justice and so many other judges, not less than two, as Parliament prescribes. However, as I see it, the vesting of judicial power in a High Court consisting of a Chief Justice and justices should not necessarily exclude the exercise of some jurisdiction and powers by a master or registrar of the Court, whether as a delegate or otherwise, provided that the exercise is subject to review or appeal, more particularly now that the Court is autonomous by virtue of the High Court of Australia Act 1979 Cth. In the case of other courts created by Parliament, whose membership is not confined by s. 72 to judges, there is perhaps even less reason for denying that part of their jurisdiction and powers may be exercised by officers who are not judges, whether as delegates or otherwise, provided of course that they are officers who truly form part of the court's organization.

It is for the foregoing reasons, rather than by reference to earlier authority in this Court, that I favour the view taken by Gibbs J. in Kotsis. None the less the view conforms to statements made by members of this Court in the earlier cases. It has frequently been said that the Commonwealth must take the organization of the State court as it is found. Gibbs J. called attention to these statements in the decided cases and I need not repeat them.

Although it is well established that this Court is not bound in a constitutional case by its prior decisions, it is usually reluctant to depart from them. In this case the Court gave leave to the parties to present an argument to the effect that Kotsis and Knight were wrongly decided. Having granted that leave, the Court ought now to deal with the issue according to its merits. Obviously the Court will be slow to embark on a reconsideration of a previous decision, unless some good reason for doing so emerges. Here the correctness of the decisions on a pure question of Constitutional interpretation was open to very serious doubt; their effect was to radically restrict the mode in which invested federal jurisdiction could be exercised and to restrict it in such a way as to produce adverse and limiting consequences for the States and their courts; and there was a divergence in the reasoning of the justices who formed the majority in Kotsis, notably a divergence between Barwick C.J. and other members of the Court.

In conclusion I should say that I have assumed, without actually deciding, that the master will exercise federal jurisdiction when he considers and determines the Commonwealth's claim of privilege against production as an element in making or refusing an order for production of the documents. Whether this exercise of jurisdiction is attributable to the grant of jurisdiction effected by s. 39(2) of the Judiciary Act to try the action instituted by The Hospital Contribution Fund of Australia or whether it is attributable to a separate grant of jurisdiction effected by s. 39(2) to determine a claim by the Commonwealth for privilege against production of documents the subject of a subpoena I do not stay to consider.

In Reg. v. Hegarty; Ex parte City of Salisbury [37] I was of opinion that Kotsis v. Kotsis [38] was wrongly decided. Kotsis v. Kotsis (and Knight v. Knight [39] which followed it) should be overruled. They were an indirect result or extension of the doctrine of separation of powers stated in Reg. v. Kirby; Ex parte Boilermakers' Society of Australia [40] (the Boilermakers' Case) which, although deeply rooted in our constitutional law, can easily lead to great inconvenience and absurdity if pushed to extremes. The Constitution is to be read as the general framework of government and not narrowly and pedantically, as it so often is. The Constitution, in vesting, or providing for the investing of, jurisdiction in courts, should not be taken as allowing the exercise of the jurisdiction only by those who, in the strictest sense, constitute the court, such as justices of the High Court and other federal courts; it should be taken as permitting the exercise by officers who are under the supervision of those who constitute the court. The necessity for avoiding any narrow approach to these provisions is evident from the fact that the courts of the State which may be vested with federal jurisdiction under s. 77(iii) include not only the Supreme Courts (as Barwick C.J. seemed to think in Kotsis v. Kotsis) but also numerous courts having a variety of limited jurisdictions constituted with a variety of memberships. Some have members who are not lawyers, some have persons exercising jurisdiction who are not strictly members of the court. In investing any court of a State with federal jurisdiction (under the Constitution, s. 77(iii)) Parliament can authorize the exercise of that jurisdiction by a person who is not strictly a member of the court, provided that the person is under the real supervision and control of the court (as by review or appeals to a judge). Traditionally, officers such as prothonotaries and registrars have exercised judicial power in taking of accounts, inquiries into or assessments of damages, and other interlocutory or preliminary matters. If Parliament has not stated otherwise, investment of any State court with federal jurisdiction should be taken as investment of the court as it is organized from time to time under State law, so that not only its members, but its officers and other persons under its supervision and control may exercise federal jurisdiction in the same way as they would State jurisdiction.

1. (1981) 147 C.L.R. 617. 2. (1970) 122 C.L.R. 69. 3. (1971) 122 C.L.R. 114. 4. (1956) 94 C.L.R. 254.

There is no provision express or implied in the Constitution which would prevent exercise of the jurisdiction of the High Court (or other court created by the Parliament) by officers or others who are not justices but act under the supervision of the justices (in the sense that their decisions are subject to review or appeal). Thus in respect of any federal court created by the Parliament (see Constitution, ss. 71, 72), Parliament may authorize the exercise of its jurisdiction by officers or others who are not justices provided that these are under the real supervision and control of the justices of the court. The master of the Supreme Court of New South Wales is not strictly part of the Court, but acts under its supervision and control. Consistently with the Constitution, s. 77(iii), he was entitled to exercise the federal jurisdiction conferred on the court by the Judiciary Act 1903.

In this matter I have had the advantage of reading the reasons for judgment prepared by the Chief Justice. I agree with those reasons and the conclusion that it is proper in this case to overrule the decisions in Kotsis v. Kotsis [41] and Knight v. Knight [42] . The considerations referred to by the Chief Justice are in my opinion sufficient to warrant the Court taking the exceptional step of overruling those decisions.

1. (1970) 122 C.L.R. 69. 2. (1971) 122 C.L.R. 114.

There is one further consideration which I think should be mentioned. Covering cl. 5 of theCommonwealth of Australia Constitution Act Imp., 63 & 64 Vict. c. 12, provides that the laws made by the Parliament of the Commonwealth "shall be binding on the courts, judges, and people of every State". On the coming into force of the Constitution the only courts were those of the States and they administered the whole of the law within each State, including the laws made by the Parliament of the Commonwealth. It is plain enough that that Act took the State courts as it found them. The Constitution itself rests upon the assumption of an existing system of State courts in existence in 1901 but capable of being changed in structure and jurisdiction by the legislatures of the States.

It does not seem to me that the "autochthonous expedient" (Reg. v. Kirby; Ex parte Boilermakers' Society of Australia [43] , per Dixon C.J., McTiernan, Fullagar and Kitto JJ.) of empowering the Parliament to confer federal jurisdiction on State courts under s. 77(iii) of the Constitution required the conclusion that only State courts organized in a particular manner could exercise federal jurisdiction so conferred. I can see no reason for the conclusion that the intention to be found in s. 77 in relation to the conferring of "federal jurisdiction" on courts of the States is significantly different from the intention of covering cl. 5 which made the laws passed by the Parliament binding upon the courts and judges of the States. Each provision is appropriate to the conferring of judicial power upon the State courts as they may be established and organized from time to time.

This is the hearing of a demurrer by the State of New South Wales to a claim of the Commonwealth for declaratory relief related to the jurisdiction and powers of a master of the Supreme Court of New South Wales.

The problem arises in this way. In 1977, the Hospital Contribution Fund of Australia ("the Fund") instituted a suit in the Common Law Division of the Supreme Court of New South Wales against Mr. Hunt, the then Commonwealth Minister of State for Health. Its statement of claim alleged that the Minister had refused to approve a proposal by the Fund for changes in the contributions payable for the medical and hospital benefits which it provided to those persons who insured with it. It claimed that the decision of the Minister was unlawful, and sought damages and ancillary relief in respect of the harm and loss allegedly occasioned by it. A defence was entered to the action. Then in February 1981, the Fund caused to issue out of the Supreme Court a subpoena directed to Mr. MacKellar, the then Minister for Health requiring him to produce the documents specified therein. On the return of the subpoena before the Master, the Commonwealth claimed privilege in respect of certain of the documents which were referred to therein. More relevantly for present purposes, it also challenged the jurisdiction of the Master to make any order in respect of the production of the documents. The proceedings were adjourned so as to allow the question of jurisdiction to be determined, and thereafter the Commonwealth together with Mr. Hunt and Mr. MacKellar instituted these proceedings in this Court.

It is common ground that the original proceedings involve the exercise by the Supreme Court of New South Wales of federal jurisdiction invested in it by the Parliament pursuant to s. 77(iii) of the Constitution and s. 39(2) of the Judiciary Act 1903 Cth, as amended. In answer to the demurrer, the Solicitor-General of the Commonwealth relies on the decision of this Court in Kotsis v. Kotsis [44] in support of the proposition that the master of the Court cannot exercise the federal jurisdiction that is invested in the Court. That jurisdiction can be exercised only by the members of the Court, and it is the judges alone who satisfy that description. Even though a law of New South Wales may authorize the Master to exercise certain of the jurisdiction of the Supreme Court, he remains an officer of the Court and therefore beyond the reach of the constitutional power of the Commonwealth to vest federal jurisdiction in "any court if a State" (s. 77(iii)).

On the other hand, the Solicitor-General of New South Wales advances two submissions in support of the demurrer. First, she points to changes in the law of New South Wales since Kotsis which in her submission renders that decision inapplicable. Secondly, she asks the Court to re-examine the decisions in Kotsis, and in Knight v. Knight [45] , a decision which applied Kotsis, with a view to overruling them. In these submissions, she is supported by the States of Victoria, Queensland and Western Australia.

The Supreme Court Act 1970 N.S.W., as amended ("the Act") has introduced significent changes into the law of the State. In the respect which bears closely on the present problem, the Act speaks in several places of the Supreme Court being constituted by the Master. This may be compared with s. 4(4) of the Matrimonial Causes Act 1899 N.S.W., as amended, which read:

The registrar, when exercising the jurisdiction and powers conferred upon or delegated to him under this Act, shall be deemed to be exercising the jurisdiction and powers of the Supreme Court. This was held by all members of this Court in Kotsis to fall short of making the registrar a constituent part of the Court. The Court consisted of the judges of the Court, and officers remained officers of the Court notwithstanding that under the law of the State they exercised part of the jurisdiction of the Court. A declaration that when exercising the powers of the Court a master "shall constitute the Court" (ss. 117A(5) and 118(5) of the Act; see also, s. 48(4)) may be to a different and stronger effect than the deeming provision with which it may be compared. Nevertheless the slender ground on which the Solicitor-General is forced to stand in this regard is revealed when the declaration is placed alongside s. 25 of the Act, which reads: The Court shall be composed of a Chief Justice, a President of the Court of Appeal and such other Judges of Appeal and Judges as the Governor may from time to time appoint. So long as the strict conception of the word "court" favoured by the justices who formed the majority in Kotsis remains authoritative, then it seems to me that s. 25 of the Act dictates a similar conclusion in the present case. Notwithstanding that in certain circumstances for the purposes of State law a master may constitute the Supreme Court, he is not a member of it for the purposes of the exercise by the Court of invested federal jurisdiction.

The situation would appear to be otherwise in Victoria and Western Australia. In each of those States the constitution of the Supreme Court has been changed since Kotsis in an attempt to meet the problems which flowed from that decision. In Victoria, the Supreme Court shall consist of judges and "the Prothonotary, the Registrar of Probates and the Masters of the Court" (s. 75(2), Constitution Act 1975 Vict.). In Western Australia, the Supreme Court "shall consist of" judges and the master (s. 7(1), Supreme Court Act, 1935-1979 W.A.).

Notwithstanding the expedient which has been adopted in some States, though not in others, serious questions may remain. In Knight, Gibbs J. (as the Chief Justice then was) referred to the further questions which might arise if a State law were to make the master a member of the Supreme Court. He said [46] :

It might then be necessary to consider whether a Supreme Court whose membership was expanded to include persons other than judges would be a court within s. 77(iii.) of the Constitution, and whether, if so, the Commonwealth Act nevertheless discloses an intention that the federal jurisdiction in matrimonial causes with which the Supreme Courts are invested should only be exercised by judges and, if it does, whether the Constitution permits the Parliament to impose conditions of that kind when effecting an investiture of federal judicial power. These questions were debated but not decided in Kotsis v. Kotsis [47] . 1. (1971) 122 C.L.R., at pp. 131-132. 2. (1970) 122 C.L.R. 69.

I do not find it necessary to examine these questions, although I have no doubt of the substantial difficulties that they would present if the determination of this case required their resolution. For example, the Kotsis doctrine requires that the words any court in s. 77(iii) comprehend only those persons who are members of the court. When the Parliament invests a court with federal jurisdiction without any qualification, how is it that the master, notwithstanding his limited functions and subordinate status, does not share in that investiture to the same extent as any other member of the court? Such a result can only come about if it be the fact that an unqualified investiture of federal jurisdiction may nevertheless become qualified in its exercise by reason of the operation of State law. But it seems to me that such a conclusion may well be opposed to the Kotsis doctrine. As I understand it, that doctrine is that only those persons who compose the court can exercise the invested jurisdiction, and in the absence of any special conditions imposed by the Parliament there is no room for any differentiation between them. Cf. Knight [48] , per Walsh J.

I think it follows that Miss Gaudron's submission runs into difficulty on two fronts. First, notwithstanding that in limited specified circumstances the master may "constitute the court", on the proper construction of the Act as a whole the master is not a member of the Court and therefore on the authority of Kotsis may not exercise the federal jurisdiction invested in the Court. Secondly, even if it were conceded that the master is a member of the Court, the problem remains of reconciling, consistently with the Kotsis construction of s. 77(iii), the limited jurisdiction which under State law he is authorized to exercise with the plenitude of the invested federal jurisdiction. Section 39(2) of the Judiciary Act supplies the measure of that plenitude for present purposes by reference to the jurisdiction of the court which receives the investiture.

The question then is whether this Court should review the decision in Kotsis. There are two aspects to the question, namely, whether the decision is considered to be correct and if not, whether it should be overruled. To those questions I now turn.

In Kotsis, the Court was concerned with the effect of s. 23(2) of the Matrimonial Causes Act 1959-1966 Cth. That provision invested without limitation the Supreme Court of each State with federal jurisdiction to hear and determine matrimonial causes instituted under that Act. A majority of the Court (Barwick C.J., McTiernan, Menzies, Windeyer, Owen and Walsh JJ.) were of the opinion that the federal jurisdiction invested in the Supreme Court could be exercised only by persons who are members of that Court, and "not by persons who, although they may have had authority under State laws to perform certain judicial functions, are not part of that Court" [49] : per Walsh J. Gibbs J. dissented, being of the view that any court of a State in s. 77(iii) of the Constitution includes its organization as provided by the State law. He said [50] : "The federal jurisdiction may therefore be exercised not only by the judges of the court but also by those officers of the court who form part of the organization through which it exercises jurisdiction in analogous cases under State law". The opposing contentions are fully discussed in the reasons for judgment of the members of the Court, and no useful purpose is to be served by my going over the ground again in any detail. In the circumstances, I find that I can express my conclusion as to the correctness of the decision quite shortly. In my opinion, with all respect to the views of the majority in that case, the decision presents too limited a concept of "any court" as those words are used in s. 77(iii). There is much to be said for the distinction which Miss Gaudron drew between the fact of conferring jurisdiction on a court and the manner of its exercise. Jurisdiction is vested in a court, and in that sense the word "court" identifies an institution consisting both of the persons who compose the court, and the officers and procedures through which its judicial function is performed. On this view, there is no problem in giving effect to an exercise by the Parliament of the power to invest federal jurisdiction in a Supreme Court. That invested jurisdiction in the absence of any qualification attached to the investiture is intended by the Parliament to be exercised in the same manner as the state jurisdiction of the court. In my view, these principles find support in the majority decision of the Court in Le Mesurier v. Connor [51] ; on the other hand, the emphasis in Kotsis on the membership of the court to the exclusion of its organization bears resemblance more to the dissenting judgment of Isaacs J. in that case [52] . I do not think that any decision of this Court prior to Kotsis supplies any authority to the contrary. I agree so fully with the dissenting judgment of Gibbs J. in Kotsis that I would gratefully adopt the opinions expressed therein as my own. It follows that I am satisfied that the decision of the Court in Kotsis and the consequent decision in Knight, were erroneous.

1. (1970) 122 C.L.R., at p. 99. 2. (1970) 122 C.L.R., at p. 109. 3. (1929) 42 C.L.R. 481, at pp. 495-496. 4. (1929) 42 C.L.R., at p. 510.

It remains to consider whether the decisions should be overruled. In Queensland v. The Commonwealth [53] the principles of stare decisis in relation to constitutional cases were considered by several members of the Court. Aickin J. [54] under-took a comprehensive review of earlier cases. He said [55] :

The cases to which I have referred above show that some general considerations have emerged which assist in the determination of the question whether a previous constitutional decision regarded as erroneous may, or should, be overruled. The first is that there should be no inhibitions about overruling a decision, the error of which has been made manifest by later cases which however have not directly overruled it. The second is that the Court will be slow to overrule, or should refuse to overrule, cases which "go with a definite stream of authority" and do not "conflict with well established principle". The third is whether the prior decision can be confined as an authority to the precise question which it decided or whether its consequences would extend beyond that question. The fourth is whether the prior decision is isolated "as receiving no support from prior decisions and as forming no part of what in one metaphor is called a stream of authority and in another a catena of cases". The fifth is whether it concerns "so fundamental a provision of the Constitution", or involves a question of such "vital constitutional importance", that its consequences are likely to be far reaching even though not immediately foreseeable in detail. It is obvious that this list cannot be regarded as exhaustive and that the factors overlap to some extent. It will be seen at once that there is nothing in the application of these principles to prevent the court from departing from the two decisions. Indeed, the third, fourth and fifth considerations combine to encourage such a step. Nevertheless, the power is not one which is ever to be exercised lightly. The doctrine of precedent plays an important part in preserving a stable legal framework and promoting respect for law. 1. (1977) 139 C.L.R. 585. 2. (1977) 139 C.L.R., at pp. 620-630. 3. (1977) 139 C.L.R., at p. 630.

These decisions occupy what might be thought to be a special category. They were made in the face of united opposition from both the Commonwealth and State Governments. They represented a departure from past practice in all States. They have caused significant practical difficulties in the administration of courts exercising federal jurisdiction. Several State legislatures have attempted to overcome the effect of the decisions by radical changes to the composition of their Supreme Courts, but without deriving therefrom any assurance that the exercise of federal jurisdiction by those courts is safe from the threat of invalidity which the decisions present. It is not merely that the decisions render the exercise of federal jurisdiction by State courts "less flexible and more costly than the exercise of State jurisdiction" (per Gibbs J. in Kotsis [56] ). A further disability is the uncertainty that attends the conduct of proceedings by reason of the fact that, in the course of an action which in its original form attracts the exercise of State jurisdiction, a federal element may intrude quite unexpectedly whereupon the proceedings continue in the exercise of federal jurisdiction (cf. Felton v. Mulligan [57] ). It is highly desirable that, subject to any special limitation which may be imposed on the investiture by the Parliament (as, for example, Judiciary Act, s. 39(2)(d); cf. Queen Victoria Memorial Hospital v. Thornton [58] ), the organization of the State court relative to the exercise of its jurisdiction be the same regardless of whether it be state or federal in origin. In my view this is a most persuasive consideration in favour of a departure from these decisions.

1. (1970) 122 C.L.R., at p. 110. 2. (1971) 124 C.L.R. 367. 3. (1953) 87 C.L.R. 144.

A further consideration is that there are no later decisions which depend upon Kotsis and Knight, and no present patterns of procedure which will be invalidated by their reversal. They were restrictive in their effect, and if they were now to be overruled there would be a liberation from that restriction.

After weighing all the circumstances I am of the opinion that the scales tip firmly in favour of overruling Kotsis [59] and Knight [60] . I would now do so.

1. (1970) 122 C.L.R. 69. 2. (1971) 122 C.L.R. 114.

It remains to consider the fate of the demurrer. The Supreme Court is invested with federal jurisdiction in relation to the suit instituted by the Fund against the Commonwealth Minister for Health. The organization and structure of the Court is relevantly determined by the Act, and in accordance with that Act the master is authorized to exercise judicial power in respect of various matters including the return of a subpoena duces tecum. In my opinion, therefore, it is competent for the Master in the case in question to deal with the return of the subpoena and if he thinks fit to release the Commonwealth from the obligation to produce the documents in question. The exercise of such authority is an exercise by the Supreme Court of the federal jurisdiction vested in it by the Parliament pursuant to s. 77(iii) of the Constitution and s. 39(2) of the Judiciary Act.

The autochthonous expedient adopted by the framers of the Constitution of investing the courts of the States with federal jurisdiction (Constitution, s. 77(iii)) was both novel and familiar. Its novelty lay in empowering the Parliament of the Commonwealth to invest a court established and maintained by the law of a State with jurisdiction which that court is bound to exercise (cf. British Medical Association v. The Commonwealth [61] ). The Parliament was not empowered to affect or alter the constitution or organization of the State court, which has to be taken as it is or, as Gibbs J. pointed out in Kotsis v. Kotsis [62] , as it is from time to time. The immunity of the courts of the States from federal legislative affection of their constitution and organization is firmly established: see Le Mesurier v. Connor [63] ; Adams v. Chas. S. Watson Pty. Ltd. [64] ; Peacock v. Newtown Marrickville and General Co-operative Building Society No. 4 Ltd. [65] ; Russell v. Russell [66] .

1. (1949) 79 C.L.R. 201, at p. 236. 2. (1970) 122 C.L.R., at p. 109. 3. (1929) 42 C.L.R. 481, at pp. 496, 498. 4. (1938) 60 C.L.R. 545, at p. 554. 5. (1943) 67 C.L.R. 25, at p. 37. 6. (1976) 134 C.L.R. 495, at pp. 516, 530, 535, 554.

Though the investing of State courts with federal jurisdiction was a novel expedient, the practice of investing new jurisdiction in an existing court was well established, and the incidents of the practice were known. In Smeeton v. Collier [67] , Alderson B. said:

I take it to be clear, that, where the legislature gives the Court any powers in general terms, and without any express limitation, it is the same as if those powers were given by the common law. The legislature is aware of the powers the Court is accustomed to exercise; and when fresh powers are given by the legislature, they are to be exercised in the usual and ordinary way. When, therefore, special limitations are intended to be imposed, the legislature express themselves to that effect Where any power is given to the Court in the usual way, the Court may exercise it in the ordinary and usual way in which the Court is accustomed to exercise its powers. Griffith C.J., speaking for this Court, expressed the same principle in Parkin and Cowper v. James [68] : Accordingly, it has been held by Judges of great authority that when a Statute confers powers upon the Court in general terms, and without any limitation either express or to be inferred from the context, they are to be exercised in the ordinary and usual way in which the Court is accustomed to exercise powers of an analogous nature, and that, if the powers in question are such as are ordinarily exercised in Chambers, they may be so exercised: Smeeton v. Collier. Medical Board of Victoria v. Meyer [69] added to the line of authority, Dixon J. citing the judgment of Griffith C.J. in Parkin v. James. The jurisdiction with which a court of a State is invested is therefore to be exercised in the ordinary and usual way in which that court exercises powers of an analogous nature, unless some contrary provision be made. 1. (1847) 1 Ex. 457, at pp. 464-465 [154 E.R. 194, at p. 198]. 2. (1905) 2 C.L.R. 315, at pp. 343, 344. 3. (1937) 58 C.L.R. 62.

The "ordinary and usual way" in which a court exercises its powers, the cursus curiae, is to be distinguished from the substantive matter in respect of which its powers are invoked. "The right to bring an action", said Lord Westbury L.C. in Attorney-General v. Sillem [70] "is very distinct from the regulations that apply to the action when brought, and which constitute the practice of the Court in which it is instituted". Lush L.J. said in Poyser v. Minors [71] that "practice" and "procedure" are convertible terms which denote "the mode of proceeding by which a legal right is enforced, as distinguished from the law which gives or defines the right", and Latham C.J. after citing this passage in Minister for Army v. Parbury Henty & Co. [72] added:

"Practice" in the common or ordinary sense of the word denotes "the rules that make or guide the cursus curiae, and regulate the proceedings in a cause within the walls or limits of the Court itself" - Attorney-General v. Sillem [73] . 1. (1864) 10 H.L.C. 704, at p. 724 [11 E.R. 1200, at p. 1209]. 2. (1881) 7 Q.B.D. 329, at p. 333. 3. (1945) 70 C.L.R. 459, at p. 489. 4. (1864) 10 H.L.C. 704, at p. 723 [11 E.R. 1200, at p. 1209].

The powers of the Parliament under s. 77(iii) extend to prescribing the practice and procedure to be followed by a State court in exercising invested jurisdiction (Baxter v. Commissioners of Taxation (N.S.W.) [74] ; Lorenzo v. Carey [75] ; The Commonwealth v. Limerick Steamship Co. Ltd. [76] ; Russell v. Russell [77] ) including the number of judges by whom invested jurisdiction is to be exercised (s. 79 of the Constitution; and see Aston v. Irvine [78] and Minister for Army v. Parbury Henty & Co. [79] ). Section 79 of the Judiciary Act 1903 Cth applies the laws of each State or Territory, including those relating to procedure, to all courts exercising federal jurisdiction in that State or Territory, and thus prescribes the procedure which governs the exercise by a State court of invested jurisdiction. The "ordinary and usual way" acquires the force of a federal law unless it is modified by another and particular federal law. In the exercise of federal jurisdiction invested by s. 39 of the Judiciary Act, the procedure of courts of a State is modified by only one provision, s. 39(2)(d), which relates to State courts of summary jurisdiction. No other provision affects or alters the "ordinary and usual way" in which invested jurisdiction is to be exercised. And so Latham C.J. in Minister for Army v. Parbury Henty & Co. said [80] :

In my opinion, both the single judges and the Full Courts of the Supreme Courts of the States have been invested with Federal jurisdiction by s. 39 and the Commonwealth Parliament has left that jurisdiction to be exercised in accordance with the ordinary cursus curiae, which has been left unchanged: Cf. Dale's Case [81] : "If a new jurisdiction is given to an existing Court - that is to say, a jurisdiction to deal with some new matters in a different mode and with a different procedure - if that jurisdiction be so given to a well-known court, with well-known modes of procedure, with well-known modes of enforcing its orders, it must, unless the contrary be expressed or plainly implied, be given to that court to be exercised according to its general inherent powers of dealing with the matters which are within its cognizance." 1. (1907) 4 C.L.R. 1087, at p. 1145. 2. (1921) 29 C.L.R. 243, at p. 253. 3. (1924) 35 C.L.R. 69, at p. 105. 4. (1976) 134 C.L.R., at pp. 518, 519, 554. 5. (1955) 92 C.L.R. 353. 6. (1945) 70 C.L.R., at pp. 488-489. 7. (1945) 70 C.L.R., at pp. 488-489. 8. (1881) 6 Q.B.D. 376, at pp. 450-451.

It follows that in matters of practice and procedure there can be no objection to a State court invested with federal jurisdiction under s. 39 pursuing its ordinary and usual way, though that way may require the exercise by non-judicial officers in the court's organization of powers reposed in them by the procedural rules of that court even though those powers are of a judicial character. The Parliament, taking the court as it finds it, must be taken to intend that the ordinary incidents of the procedure of that court are attached to the exercise by that court of its invested jurisdiction: seeElectric Light and Power Supply Corporation Ltd. v. Electricity Commission (N.S.W.) [82] and the cases there cited following National Telephone Co. Ltd. v. Postmaster-General [83] .

1. (1956) 94 C.L.R. 554, at p. 559. 2. [1913] A.C. 546.

When a law, enacted in reliance on s. 77(iii), invests with federal jurisdiction a court composed of judicial officers and does not affect or alter the practice and procedure of that court, there is no reason to assume that the Parliament intended to forbid the making of orders in matters of practice and procedure by non-judicial officers in the ordinary and usual way. There is nothing in s. 71 or s. 77(iii) which requires an order in a procedural matter to be made by the judges (or by one of the judges) who compose the court invested with the substantive jurisdiction. For that matter, I see nothing in Ch. III which requires orders in procedural matters to be made by the justices of this Court or by the justices of other courts created by the Parliament if the procedural rules of the respective courts provide that non-judicial officers of those courts may hear and dispose of applications in matters of practice and procedure. But that is because courts have an inherent power to define their own practice (Cocker v. Tempest [84] ; Scales v. Cheese [85] ; Reg. v. Justices of Denbighshire [86] ). That matter does not now require consideration.

1. (1841) 7 M. & W. 502, at p. 503 [151 E.R. 864, at p. 865]. 2. (1844) 12 M. & W. 685, at p. 687 [152 E.R. 1374, at p. 1375]. 3. (1846) 15 L.J.Q.B. 335.

In Kotsis v. Kotsis, which we are invited to reconsider, the order was not procedural. It was common ground that the order there in question - an order to pay interim costs - was made in a matrimonial cause as defined by par. (d) of the definition of that term in s. 5(1) of the Matrimonial Causes Act 1959 Cth. The substantive jurisdiction to hear and determine matrimonial causes, as defined, was conferred by that Act upon "the Supreme Court of each State" (s. 23(2)). The first question which arose was whether the Supreme Court mentioned in s. 23(2) comprehended the deputy registrar of the Supreme Court of New South Wales in his exercise of the function of making the order in question (per McTiernan J. [87] ). This question was resolved without dissent. The Supreme Court was composed of the judges who formed it, the registrars being but officers of the Court. That opinion was common to all the judgments. In the absence of material statutory amendment, it is not now open to reconsideration.

The next question was whether federal jurisdiction to make an order for interim costs had been validly delegated to the deputy registrar who was not a member of the Court. That question arose because there had been an earlier delegation of State jurisdiction. State jurisdiction to make an order as to costs under the Matrimonial Causes Act 1899 N.S.W. had earlier been conferred upon the Supreme Court (s. 47(1)), and s. 91(2) of that Act had conferred upon that Court a power of delegation. The judges had been authorized to delegate to the registrar "any such authority and jurisdiction as is now exercised by the Court except in respect of the hearing of suits or in respect of matters relating to the liberty of the subject". The judges had delegated the jurisdiction to make an order for a sum on account of costs to the registrar (r. 198(1)) and, in his absence, to the deputy registrar (r. 198(12)). Gibbs J. (as the Chief Justice then was) stated the effect of these provisions [88] : "It is clear from the words of s. 91(2) and r. 198 that a deputy registrar exercised the authority conferred on him by these provisions not as a member of the court but as its delegate." Although both the Commonwealth and State Acts had conferred jurisdiction upon "the Supreme Court", only the State statute had conferred a power of delegation. And thus the question was whether there was, under State law, a delegation to the deputy registrar of the federal jurisdiction with which the Supreme Court had been invested. Gibbs J. gave an affirmative answer to the question, distinguishing between delegation to a stranger and delegation to an officer of the court. He referred [89] to what Rich and Dixon JJ. had said in Bond v. George A. Bond & Co. Ltd. and Bond's Industries Ltd. [1] :

Speaking obiter, Rich and Dixon JJ. did suggest that Parliament probably could not validly authorize a court to delegate part of its judicial power. These remarks in their context no doubt refer to a delegation to a person who is a stranger to the organization of the court, but whether or not that is so they do not relate to the present case in which Parliament has not purported to authorize any delegation by the State courts which it has invested with federal jurisdiction. The delegation here in question is made under the authority of a State law which governs the organization of the State court. Accordingly, Gibbs J. [2] held that "federal jurisdiction may be exercised not only by the judges of the court but also by those officers of the court who form part of the organization through which it exercises jurisdiction in analogous cases under State law." His Honour was the sole dissentient. The majority denied the applicability to invested jurisdiction of the power to delegate which the New South Wales Act had authorized the judges to exercise with respect to a claim for substantive relief. Walsh J. said [3] : Nobody can be brought in, as it were, from outside the existing Court by the Commonwealth law and directed by it to exercise part of the jurisdiction with which the Commonwealth invests the State court. Likewise the investing cannot be taken to have the effect of conferring authority to exercise federal jurisdiction upon someone who is not a part of the State court which is being invested with that jurisdiction. A conclusion that the Act has this last-mentioned effect cannot be supported, in my opinion, by saying that that other person is part of the organization through which the powers and jurisdiction of the court are exercised. 1. (1970) 122 C.L.R., at p. 103. 2. (1970) 122 C.L.R., at p. 108. 3. (1930) 44 C.L.R. 11, at p. 21. 4. (1970) 122 C.L.R., at p. 109. 5. (1970) 122 C.L.R., at p. 99.

I respectfully agree. There is a difference between the procedures which govern the exercise by a court of jurisdiction conferred upon it and the delegation of that jurisdiction to another, albeit to an officer in the court's organization. A court constituted by judges does not exercise its jurisdiction to hear and determine applications for substantive relief if it delegates that jurisdiction to an officer of its organization. On the other hand, a court may exercise its jurisdiction though it requires its officers to inquire into and report upon issues or otherwise perform duties of the kind which Windeyer J. described as "ancillary" or "subservient" to adjudication [4] . Delegation of a power to adjudicate is to be distinguished from the delegation of duties ancillary or subservient to adjudication; in the former case the power is exercised by the delegate, in the latter it is retained for exercise by the court. Delegation of a power to determine substantive applications is not a way of exercising that power; it is a way of avoiding the necessity for its exercise by the primary repository of the State power. The deputy registrar in Kotsis v. Kotsis had been empowered to exercise the State jurisdiction delegated to him in relief of the duty which would otherwise have rested upon the Supreme Court to exercise the jurisdiction originally conferred upon it. By delegation, the Court had conferred that power upon a person who was not a member of that Court. The majority in Kotsis v. Kotsis denied and the minority affirmed that a consequence of the delegation by the Court of its State jurisdiction was that analogous federal jurisdiction would be delegated in like manner if it should be conferred on that Court.

I do not understand the reference to analogous cases inSmeeton v. Collier [5] , Parkin v. James [6] and Medical Board of Victoria v. Meyer [7] to do more than point to the practice and procedure which is to be taken to regulate the cursus curiae of proceedings under newly-invested jurisdiction. The ordinary and usual way in which a court is accustomed to exercise powers in a particular class of case is to be followed when new jurisdiction gives rise to an analogous class of case. And thus analogy may identify the way in which judicial powers of hearing and determining an application for substantive relief should be exercised; but I do not understand the earlier cases to suggest that analogy might identify those judicial powers which follow the path of delegation.

1. (1847) 1 Ex. 457 [154 E.R. 194]. 2. (1905) 2 C.L.R. 315. 3. (1937) 58 C.L.R. 62.

The chain of reasoning upon which Gibbs J. relied contains a link which was not forged in earlier cases, namely, that a delegation of judicial power under State law by a State court to its own officers alters or affects its organization. With respect, it seems to me that the organization of the Supreme Court of New South Wales was not altered by the delegation of certain State judicial powers to its registrar or deputy registrar. The delegation took effect within an unaltered organization which was taken as it was found when it was invested with federal jurisdiction. There was no alteration to its organization as that term was used in Le Mesurier v. Connor [8] and in Bond's Case [9] . Moreover, it seems that Gibbs J. did not hold the view in Kotsis v. Kotsis that the delegation of State judicial power to the deputy registrar effected such an organizational alteration that it was beyond the reach of federal legislative power. After reference to Queen Victoria Memorial Hospital v. Thornton [10] , and the recognition which was there given to the validity of s. 39(2)(d) of the Judiciary Act, his Honour said [1] :

It may follow that it would similarly be open to Parliament to prescribe that the federal jurisdiction invested in a State court should be exercised only by a judge and not by a master or registrar and that such a prescription would not be regarded as an interference with the organization of the State court. If the power of delegation is a part of the organization of a State court, it differs from those other aspects of the organization which are characteristically immune from federal interference. 1. (1929) 42 C.L.R. 481. 2. (1930) 44 C.L.R. 11. 3. (1953) 87 C.L.R. 144, at p. 152. 4. (1970) 122 C.L.R., at p. 111.

If delegation under a State law of a State court's jurisdiction to its officers is held to affect the organization of the court so that federal jurisdiction of an analogous kind conferred upon the court is likewise delegated, federal jurisdiction, non-delegable by federal law, becomes both delegable and delegated by virtue of the antecedent delegation of State jurisdiction. It is beyond the legislative power of the Parliament under s. 77(iii) to confer a power of delegation upon a State court, and it is equally beyond the competence of a State legislature to confer upon its courts a power of delegation of federal jurisdiction with which they are invested by federal law. It would be a curious result if a power to delegate federal jurisdiction with which a State court is invested cannot be directly conferred upon that court by either the Commonwealth or the State legislature, yet the exercise of a power to delegate State jurisdiction works a delegation of analogous federal jurisdiction with which the State court is subsequently invested.

It follows that I would not depart from Kotsis v. Kotsis. I find the coincident reasoning of the majority to be compelling. Nor would I depart from Knight v. Knight [2] . The importance of the judgments in Kotsis v. Kotsis and Knight v. Knight for the operation of the autochthonous expedient have led me to examine the principles therein expressed, but those authorities do not furnish the answer to the problem raised in the present case. Each of those cases concerned substantive claims for relief, that is, claims which originated matrimonial causes as defined in s. 5(1) of the Matrimonial Causes Act 1959 Cth. In one case, a claim for interim costs; in the other, a claim for maintenance pending suit. Neither case was concerned with mere matters of practice and procedure. Nor did either application seek the performance of some duty ancillary or subservient to adjudication.

The present case relates to proceedings for substantive relief pending in the Supreme Court of New South Wales between the Hospital Contribution Fund of Australia as plaintiff and Ralph James Dunnet Hunt as defendant. Mr. Hunt was the Minister of State for Health of the Commonwealth in November 1977 and the action was brought to challenge the lawfulness of a decision made by him. The action appears to be framed in tort. The question which has arisen for determination in that action follows upon the issue out of the Supreme Court upon the request of the plaintiff of a subpoena directed to Mr. MacKellar, the then Minister of State for Health of the Commonwealth, to produce documents which are in his possession by reason only of his being the Minister of State for Health. The question which the Commonwealth wishes to have determined in the Supreme Court is whether it is entitled not to produce certain of the documents referred to in the subpoena upon the ground that their production would be injurious to the public interest.

The subpoena was returned before the master. Pursuant to Pt 60, r. 1 of the Supreme Court Rules 1970 N.S.W., the master was authorized to exercise the power of the Court not to compel the production of a document. Part 36, r. 13(1) provides:

Where the Court, by subpoena or otherwise, orders any person to produce any document or thing, and any person makes and substantiates sufficient lawful objection to production on grounds of privilege, the Court shall not compel production of that document or thing except production to the Court for the purpose of ruling on the objection. A master is restricted to the making of an exempting order. He has no jurisdiction to enforce an obligation to produce documents by entertaining proceedings for contempt (see Supreme Court Act 1970 N.S.W., s. 124(4); Supreme Court Rules, Sch. D, Pt 3 Item 6). The jurisdiction of the master was therefore limited to exempting the Commonwealth from an obligation to produce documents according to the exigency of the subpoena. However, the statement of claim endorsed upon the writ by which the Commonwealth, Mr. Hunt and Mr. MacKellar commenced the action in this Court, appears to assume that the master's power is not merely to exempt, for it alleges that the master "is not invested with federal jurisdiction pursuant to the Judiciary Act 1903 or otherwise and that any order made by him for the production of the documents would be void and of no effect". Of course, if a master's order overrules a claim of privilege, the obligation to produce is enforceable so long as the order stands (Field v. Beaumont [3] ), but a master's order under Pt 36, r. 13(1) is not in form or substance an order for production. 1. (1818) 1 Swans. 204, at p. 209 [36 E.R. 358, at p. 360].

The obligation to produce a document is antecedent to any order that might be made on the return of the subpoena. The obligation flows from the power of the Court to secure the attendance of witnesses either to testify or to produce documents, a power which has long been recognized as an essential adjunct to adjudication: see Amey v. Long [4] and Field v. Beaumont. The Commonwealth and Mr. MacKellar do not assert an immunity from the subpoena issued by the Supreme Court. Therefore the action in this Court should be determined on the footing that Mr. MacKellar is bound by the subpoena to produce to the Supreme Court the documents therein specified (see Sankey v. Whitlam [5] ) unless he is exempted from producing those documents on the ground of privilege.

1. (1808) 9 East. 473, at p. 484 [103 E.R. 653, at p. 658]. 2. (1978) 142 C.L.R. 1, at pp. 32, 93.

Neither the Commonwealth nor Mr. MacKellar is a party to the action. If the Commonwealth were a party to the action, its obligation to discover documents would flow from s. 64 of the Judiciary Act (The Commonwealth v. Miller [6] ). Questions with respect to discovery between parties to an action are matters of procedure (Griffin v. South Australia [7] , per Isaacs A.C.J.). Although the application for exemption made by the Commonwealth is not made by a party, it is made in proceedings pending in the Supreme Court between the parties, the Hospital Contribution Fund and Mr. Hunt. Applications in a proceeding are not restricted to those made by a party to the proceeding (R. v. Investors' Review Ltd.; Ex parte Wheeler [8] ).

1. (1910) 10 C.L.R. 742. 2. (1924) 35 C.L.R. 200, at p. 206. 3. [1928] 2 K.B. 644.

An objection to production of documents under a subpoena duces tecum is incidental to the proceedings in which the subpoena is issued; it is incidental to the securing of evidence for the purpose of the hearing of those proceedings. A claim of privilege is so much an incident of the securing of evidence that counsel for a person to whom a subpoena was directed has been refused audience to argue the objection: Doe d. Rowcliffe v. Earl of Egremont [9] . No doubt that case would not prevent a modern court from hearing counsel when it seems appropriate to do so, but it illustrates the incidental nature of the exempting power. Its exercise is no more than a procedural incident in the securing of evidence.

I do not say that a claim for a declaration made against the party who caused the issue of the subpoena would be procedural. Much depends upon the form of the proceeding as well as the question for determination. But in the present case, the application before the master is purely a matter of procedure in the proceedings pending in the Supreme Court. Making the assumption, which was adopted in argument here, that the Supreme Court was exercising federal jurisdiction in hearing and determining the proceedings pending before it, the application for exemption from production, being procedural, is to be determined by the master in the ordinary and usual way in which he determines questions of that kind. The master's jurisdiction is therefore to be determined by reference to the Supreme Court Rules. Pursuant to Pt 60, r. 6 the master may determine the application himself or, if the question for determination appears to warrant referring it to the Court, he may refer it.

The statement of claim in this Court is demurrable. However, the first ground of the demurrer of the State of New South Wales seeks to revive the first proposition which was put to rest in Kotsis v. Kotsis [10] . The demurrer alleges:

A divisional master of the Supreme Court of New South Wales constitutes that court for the purpose of exercising the jurisdiction conferred on that court pursuant to Section 39(2) of the Judiciary Act 1903, to the extent of the powers conferred on such master by the Supreme Court Act 1970 N.S.W. and the Rules made thereunder. I would not allow the demurrer on that ground. But I would allow it on the ground that the master is empowered to determine the question sought to be raised by the Commonwealth in the action between the Hospital Contribution Fund of Australia and Mr. Hunt. The statement of claim delivered in the action in this Court should be struck out and the plaintiffs in this Court should pay the costs. 1. (1970) 122 C.L.R. 69.