Mckain v R W Miller & Co (SA) Pty Ltd

[1991] HCA 56

HIGH COURT OF AUSTRALIA

Mason C.J., Brennan, Deane, Dawson, Toohey, Gaudron McHugh JJ.

McKAIN v. R.W. MILLER and COMPANY (SOUTH AUSTRALIA) PTY. LIMITED

(1992) 174 CLR 1

19 December 1991

Private International Law

Private International Law—Tort—Negligence—Act committed in South Australia—Personal injury—Three year limitation period under lex loci delicti—Action in another State beyond limitation period under lex loci but within period under lex fori—Choice of law—Applicable period—Statute of limitation—Character—Whether substantive or procedural—South Australian statute conferring power to extend limitation period—Whether court of another State may extend time in action governed by South Australian law—The Constitution (63 and 64 Vict. c. 12), s. 118—Limitation of Actions Act 1936 (S.A.), ss. 36(1), 48(1).

Decisions

MASON C.J. In this Court's recent decision in Breavington v. Godleman (1988) 169 CLR 41, a majority held that the common law conflict of laws rule in Phillips v. Eyre (1870) LR 6 QB 1, which was designed to resolve conflicts between the different tort laws of independent nations, should not be applied without qualification to resolve conflicts between the tort laws of the Australian States and Territories. In reaching that conclusion, the majority specifically departed from the Court's earlier decisions in Koop v. Bebb (1951) 84 CLR 629, and Anderson v. Eric Anderson Radio and T.V. Pty. Ltd. (1965) 114 CLR 20, where it had been accepted that the law of the forum is the applicable law to determine liability for torts committed in Australia.

2. In Breavington, the majority held that, in circumstances where the plaintiff commenced proceedings in the Supreme Court of Victoria for damages for personal injury, the plaintiff's cause of action having arisen in, and whilst the plaintiff and the defendant were residents of, the Northern Territory, the case should be determined in accordance with the law of the Northern Territory, the lex loci delicti and the law of the place having the closest connection with the parties and the cause of action. Only two Justices held that the rule in Phillips v. Eyre applied automatically so that the plaintiff was precluded by operation of the choice of law rules of the forum, the State of Victoria, from recovering certain types of damage (1988) 169 CLR, per Brennan J. at pp 118-119; Dawson J. at pp 144-146.

3. The case currently before the Court was removed pursuant to s.40(1) of the Judiciary Act 1903 (Cth) from the Supreme Court of New South Wales. It is not in dispute between the parties that the cause of action arises under and should be determined in accordance with the law of South Australia. What is in dispute between the parties is whether s.36 of the Limitation of Actions Act 1936 (S.A.) forms part of the substantive law of South Australia for the purpose of resolving the dispute between the parties.

4. The material facts may be briefly stated. The plaintiff, a resident of New South Wales, had worked as a merchant seaman since 1974 and, in that time, had been engaged on various Australian vessels. He was allocated to employment with Australian shipowners through the Sydney roster of the "Marine Cooks, Marine Stewards and Seamen's Engagement System", as provided for in Sched.(X) of the Maritime Industry Seagoing Award 1983 ("the Award"). In Sydney, that Engagement System requires all merchant seamen registered with the Department of Transport and Communications and not serving on a vessel or on ordinary leave to attend at a "pick-up centre", situated at the Mercantile Marine office in Sydney, on the day following the publication of their registration number in the Seamen's Engagement Notices in the Sydney Morning Herald. The names of all seamen present are recorded on a board in an order which reflects the length of time each seaman has spent on the roster without being allocated to a vessel.

5. On or shortly before 16 February 1984, the defendant, a company registered in South Australia, requested the "employers' representative" at the Mercantile Marine office, appointed in accordance with the Award, to arrange for the provision of a marine steward to work on the motor vessel "Troubridge" in South Australia from 17 February 1984. The "Troubridge" was chartered to the defendant company by its owners, the South Australian Government. Through the Sydney roster of 16 February 1984, the plaintiff was allocated to employment on this vessel from 17 February 1984. He flew to South Australia on that date to take up his position on the "Troubridge" and, on or about that date, entered into ship's articles of agreement with the defendant company in Port Adelaide. From that date until 22 February 1984, the plaintiff was accommodated on board and carried out his duties on the "Troubridge". On 22 February, according to his statement of claim, the plaintiff sustained injuries whilst on board the "Troubridge" as a result of a breach by the defendant of the duty of care owed to the plaintiff as employee. The following day the plaintiff consulted a medical practitioner in Adelaide and then, on 25 February, at the defendant's expense, flew back to Sydney where, he claims, he incurred medical expenses and sustained further damage.

6. The plaintiff, by statement of claim filed on 4 January 1990, sought damages in the Common Law Division of the Supreme Court of New South Wales. By its amended defence, the defendant alleged that any injuries sustained by the plaintiff were caused by his negligence. In the alternative, the defendant alleged that the plaintiff's claim is statute barred by virtue of s.36 of the Limitation of Actions Act.

7. On 21 September 1990 Sharpe J. ordered that, pursuant to Pt 31, r.2 of the Supreme Court Rules (N.S.W.), there be a separate trial on the question of law raised by the defendant's alternative defence, namely, whether the defendant is entitled to judgment by virtue of s.36 of the Limitation of Actions Act. It is that separate trial which has been removed to this Court.

8. At the outset of the hearing, Mr Ellicott Q.C. for the defendant applied to the Court for leave to further amend the defence by adding a further paragraph to the effect that the plaintiff's action is barred by s.82(2) of the Workers Compensation Act 1971 (S.A.). That sub-section relevantly provides:

"Where a worker has received or is entitled to receive compensation under this Act ... in respect of an injury, he shall not bring an action against the employer for damages in respect of the same injury unless he commences that action within three years from the day on which that injury occurred."

Section 36(1) of the Limitation of Actions Act provides:

"All actions in which the damages claimed consist of or include damages in respect of personal injuries to any person, shall be commenced within three years next after the cause of action accrued but not after."

9. There is no evidence that the plaintiff has received any compensation under the Workers Compensation Act but, for present purposes, I shall assume that he had an entitlement to receive such compensation. On that assumption, it may be observed that the date comprehended by the expressions "within three years from the day on which (the plaintiff's) injury occurred" Workers Compensation ACt, s. 82(2), and "within three years next after the cause of action accrued but not after" Limitation of Actions Act, s. 36(1) will be the same. It should be observed at this stage and for the sake of completeness that these two limitation periods have been construed as having a concurrent operation, Karasaridis v. Kastoria (1984) 37 SASR 345, per King C.J. at p 353. Whatever the juridical difference, if any, between the two limitation provisions, the relevant date before which, according to either statute, the plaintiff should have commenced his action so as to avoid the operation or potential operation of the limitation provision was 22 February 1987. As has been observed, the plaintiff in fact commenced his action on 4 January 1990.

10. In New South Wales, s.14(1) of the Limitation Act 1969 (N.S.W.) relevantly provides:

"An action on any of the following causes of action is not maintainable if brought after the expiration of a limitation period of six years running from the date on which the cause of action first accrues to the plaintiff or to a person through whom he claims -

(a) a cause of action founded on contract (including quasi contract) not being a cause of action founded on a deed;

(b) a cause of action founded on tort, including a cause of action for damages for breach of statutory duty".

Thus, it is evident that, if the New South Wales statute applies to the exclusion of the South Australian provisions, the plaintiff's claim would not be excluded as having been brought out of time and the two defences sought to be relied upon by the defendant based upon s.36(1) of the Limitation of Actions Act and s.82(2) of the Workers Compensation Act would not avail it.

11. Mr Wheelahan Q.C. for the plaintiff argues that, although the substantive question of liability should be determined according to South Australian law, limitation periods are matters of procedure and, accordingly, should be governed by the law of the forum, i.e., the law of New South Wales. This argument, he contends, is consonant with the traditional analysis applied in private international law to most statutes of limitation. Mr Wheelahan accepts that Breavington overruled Koop v. Bebb and concedes that Breavington decided that, "for torts committed within Australia, the law to be applied is the substantive law of the lex loci delicti".

12. Mr Ellicott contends that the plaintiff's decision to commence proceedings in New South Wales rather than South Australia was a case of forum shopping. The plaintiff's willingness to acknowledge that the law of South Australia should govern the substantive matters at issue between the parties and the fact that the plaintiff was a resident of New South Wales, was allocated to work on the "Troubridge" whilst in New South Wales and suffered damage and incurred medical expenses in that State detract from the force of that contention. However, Mr Ellicott further submits that, it being accepted that South Australia was the place of the wrong, full operation and effect should be given to all relevant laws of South Australia which define the plaintiff's entitlement to damages. He argues that the two South Australian statutes of limitation are such relevant laws. Moreover, he attacks the traditional private international law mode of analysis which characterizes "conventional" limitation provisions such as s.36(1) as denying the remedy without destroying the right on the ground that the analysis is both artificial and contrary to common sense. Such limitation provisions, he submits, should be classified as substantive rather than procedural.

13. The precise question, then, for this Court to consider is whether, in the context of a conflict of laws within Australia, the two South Australian statutes form part of the substantive law of South Australia or are merely procedural or adjectival. Underlying this question is the proposition, not questioned by the parties, that matters of substance are to be determined by the law of the cause (lex causae), in this case the law of South Australia, and matters of procedure by the law of the forum (lex fori), New South Wales law. Even those Justices in Breavington who placed special significance on s.118 of the Constitution as indicating which law should govern the outcome of the cause acknowledged impliedly or expressly that, as to procedural or adjectival matters, the law of the forum should apply (1988) 169 CLR, per Wilson and Gaudron JJ. at pp 98-99; Deane J. at pp 135-136. It will be necessary to consider the distinction drawn in private international law between "substantive" and "procedural" matters, the basis for this distinction and the question of what is the appropriate classification for statutes of limitation. Before embarking on that inquiry, however, it is necessary to examine how statutes of limitation have traditionally been classified in private international law.

14. In Australian Iron and Steel Ltd. v. Hoogland (1962) 108 CLR 471, at p 488 Windeyer J. observed that "(s)tatutory provisions imposing time limits on actions take various forms and have different purposes." Further, judicial opinions sometimes differ as to the juridical effect of the same statute. See the dissent of Kitto J. in Australian Iron and Steel Ltd. v. Hoogland, ibid., at p 480; and Byrnes v. Groote Eylandt Mining (1990) 19 NSWLR 13, per Mahoney J.A at pp 27-28. Some judges may regard a particular statute as substantive in effect while others will regard the same statute as merely procedural or adjectival. In Byrnes v. Groote Eylandt Mining, Kirby P observed (1990) 19 NSWLR, at p 24 that "(d)ifferences of opinion illustrate the imprecision inherent in the classification of particular limitation provisions as 'procedural' or 'substantive'."

15. A distinction is generally drawn between those statutes which are said to take away the remedy available to a party without touching that party's underlying right and those statutes or limitation provisions which expressly or impliedly extinguish an underlying right. Australian Iron and Steel Ltd. v. Hoogland (1962) 108 CLR, at pp 488-489. In Williams v. Jones (1811) 13 East 439, at pp 450-451 (104 ER 441, at p 446, Lord Ellenborough C.J., when construing the statute of limitations applicable in India, stated:

"(H)ere there is only an extinction of the remedy in the foreign Court, according to the law stated to be received there, but no extinction of the right ... If it go to an extinguishment of the right itself, the case may be different."

In Harris v. Quine (1869) LR 4 QB 653 the statute of limitations of the Isle of Man was interpreted as barring only the remedy. Lush J. confirmed, however, that, had the Manx statute of limitations in fact extinguished the right after the nominated time, the case would have been different, for the Manx statute could have been pleaded as a good defence in the forum court, ibid., at p 658; and see per Blackburn J. at p 658. See also Phillips v. Eyre (1870) LR 6 QB, at p 29.

16. Statutes of limitation which operate to prevent the enforcement of independently existing rights of action, whether known at common law or, in some circumstances, created by statute, are typically described as denying a remedy while not destroying or extinguishing an underlying right. In certain contexts, this distinction is an important one. In Maxwell v. Murphy (1957) 96 CLR 261, at p 269 Dixon C.J. observed that "(t)here are rights in English law which have an existence and a purpose although the remedy be suspended or wanting." This proposition may be readily illustrated. In Spears v. Hartly (1800) 3 Esp 81 (170 ER 545) Lord Eldon held that, in an action of trover for a log of mahogany, the creditor defendant held a valid and subsisting lien over the wood as the debt owing to him had not been discharged even though his remedy by action was barred by the statute of limitations. That having been said, the distinction drawn between the denial of the remedy and the subsistence of the right has not been immune from criticism, especially in the context of the classification of statutes of limitation in private international law as procedural. The distinction has been described as both artificial and semantic, Cheshire and North, Private International Law, 11th ed. (1987) p 80; McLeod, The Conflict of Laws (1983) p 215, Sykes and Pryles state that, in the context of private international law, the distinction is not meaningful. Australian Private International Law, 3rd ed. (1991) p 258. Leflar et al. mordantly observe that a "right for which the legal remedy is barred is not much of a right". Leflar, McDougal and Felix, American Conflicts Law, 4th ed. (1986) p 349. To similar effect is Lorenzen's observation that a "right which can be enforced no longer by an action at law is shorn of its most valuable attribute". "The Statute of Limitations and the Conflict of Laws", (1919) 28 Yale Law Journal 492, at p 496.

17. Not all statutes of limitation, however, operate simply to deny a party a remedy while leaving a right in existence. Limitation provisions which can be seen as incidents of rights created, whether by the same or another related statute, have been typically construed as extinguishing those rights after the effluxion of the nominated period of time. As Windeyer J. put it in Hoogland (1962) 108 CLR at p 488 when describing such limitation provisions, "the limitation does not bar an existing cause of action. It imposes a condition which is of the essence of a new right." Earlier, in Maxwell v. Murphy, the Court considered that the statute there under consideration had conferred a new right in terms of remedy and had limited the remedy in terms of time. The Court therefore held in that case that the extinction of the remedy dictated the extinguishment of the right. In The Harrisburg (1886) 119 US 199, at p 214, the Supreme Court of the United States stated:

"The time within which the suit must be brought operates as a limitation of the liability itself as created, and not of the remedy alone. It is a condition attached to the right to sue at all. ... Time has been made of the essence of the right, and the right is lost if the time is disregarded. The liability and the remedy are created by the same statutes, and the limitations of the remedy are, therefore, to be treated as limitations of the right."

18. In some cases, a limitation provision will, by its express terms, extinguish the right of a person having a cause of action. See, e.g., s.63(1) of the Limitation Act (NSW); Commonwealth of Australia v. Dixon (1988) 13 NSWLR 601. In other cases, the conclusion that a limitation provision operates to extinguish a right will turn on the process of statutory construction. See e.g. Byrnes v. Groote Eylandt Mining. In the United States, where a statute both creates a new cause of action and provides for a limitation period concerning it, such limitation provisions have presumptively been held to be substantive. The Harrisburg; Davis v. Mills (1904) 194 US 451, at p 454. Such a statute and statutes of limitation which extinguish rights are classified in private international law as substantive.

19. On the other hand, statutes of limitation which only deny a remedy and leave intact and subsisting the underlying right have traditionally been classified as procedural or adjectival. Huber v. Steiner (1835) 2 Bing (NC) 202 (132 ER 80); Harris v. Quine; Alliance Bank of Simla v. Carey (1880) 5 CPD 429; John Robertson and Co. Ltd. v. Ferguson Transformers Pty Ltd (1973) 129 CLR 65, at p 92; The Commonwealth v. Verwayen (1990) 170 CLR 394, per Toohey J. at p 474. This classification was accepted and applied by this Court in Pedersen v. Young (1964) 110 CLR 162, at p 166, where Kitto J. described it as "the established doctrine of English law" and Menzies J. ibid., at p 166, referred to it as "a well-established principle". It has the consequence that, in a private international law context, where a statute of limitations forming part of the law of the cause is of a kind which extinguishes only the remedy but not the right, that statute will not apply in the forum in which the action is brought. Rather, the statute of limitations of that forum will apply, assuming it also falls within the "procedural" classification. The basis for this is said to be that "whatever relates to the remedy to be enforced, must be determined by the lex fori". Don v. Lippmann (1837) 5 Cl and Fin 1, per Lord Brougham at p 13 (7 ER 303, at p 307). An equation has thus been drawn between matters relating to a remedy and matters of procedure. See, e.g. Goodrich, Handbook on the Conflict of Laws, (1927) p 157.

20. Statutes or limitation provisions which extinguish rights are quite rightly classified as substantive. Their operation is central to the rights of the parties. They directly affect the outcome of any dispute. But the same may be said of a limitation provision which bars the remedy but not the right. Whether or not it follows as a logical corollary that, because a statute of limitations which extinguishes a right is classified as substantive, therefore a statute of limitations which does not extinguish a right but only denies a remedy must be classified as procedural is central to the question raised by this case.

21. As has been said, underlying the argument in this case is the proposition that, in the context of private international law, matters procedural or adjectival are to be determined according to the law of the forum. Notwithstanding the observation of Lord Wilberforce in Black-Clawson Ltd. v. Papierwerke AG. (1975) AC 591, at p 632 that "(f)or English law to abolish the distinction between substance and procedure ... might be an intelligible objective", it remains the fact, as one writer has put it, that:

"It is perhaps the most inveterate doctrine of the conflict of laws that all questions of procedure in a given instance are governed by the lex fori, or the law of the court invoked, regardless of the law under which the substantive rights of the parties accrued." Heavner v. Uniroyal, Inc. (1973) 305 A 2d 412, at p 415 et seq.

Not only is such a rule firmly rooted in precedent, it is soundly based in common sense. That the courts of the forum should apply their rules of procedure is both sensible and legitimate by reason of the judges' practical familiarity with those rules and because those rules, no doubt developed and refined over time, are designed to facilitate the process of litigation in a particular jurisdiction and to ensure that cases are heard efficiently and expeditiously. The fact that one party has chosen and the other party has submitted to a forum's jurisdiction indicates a willingness on the parties' part to litigate their cause in the courts of that forum, according to the ordinary way in which litigation in that forum is conducted.

22. However, while the proposition that matters of substance should be determined according to the law of the cause and matters of procedure according to the law of the forum is clear and relatively uncontroversial, its simplicity belies the difficulty of identifying just what is procedural and what is substantive. Indeed, academic debate in this area of law tends to be concerned neither with the basic distinction between substance and procedure nor with the rationale for that distinction, but rather with the scope given to "procedure". McLeod, op. cit., p 213.

23. Dicey, in his first edition, commented that English lawyers give "the widest possible extension to the meaning of the term 'procedure'". The Conflict of aws (1896), p 712. In Maxwell v. Murphy Dixon C.J. (1957) 96 CLR, at p 267 referred to the "inveterate tendency of English law to regard some matters as evidentiary or procedural which in reality must operate to impair or destroy rights in substance". Various historical explanations have been offered for this tendency. Lorenzen, op. cit., p 496. Obviously, it developed in the context of transnational rather than intranational disputes such as occur within a federation. The tendency developed at a time when the importance of international judicial comity may not have been given the same recognition it nowadays commands. See, e.g., The Abidin Daver (1984) AC 398, per Lord Diplock at p 411; Spiliada Maritime Corporation v. Cansulex Ltd (1987) AC 460, at p 477; Oceanic Sun Line Special Shipping Company Inc. v. Fay (1988) 165 CLR 197, per Wilson and Toohey JJ. at p 212; Deane J. at pp 250, 253-254, and when the notion of forum shopping was not considered as objectionable a practice as it now is. Breavington v. Godleman (1988) 169 CLR, at pp 88, 113, 161, 170; cf. The Atlantic Star (1973) QB 364, per Lord Denning M.R. at p 382. In contrast to the first edition, the eleventh edition of Dicey and Morris notes that, in general, the practice of giving a broad scope to the classification of a matter as procedural has fallen into disfavour because of its tendency to frustrate the purposes of choice of law rules. Dicey and Morris on the Conflict of Laws, 11th ed. (1987), vol. 1, p 173. Evidence of this change of practice is substantial. English Law Commission, Classification of Limitation in Private International Law, Report No. 114, (1982), pp 6-11; Foreign Limitation Periods Act 1984 (UK); s. 23A of the Prescription and Limitation (Scotland) Act 1973 (Scot.); Art. 10(1)(d) of the EEC Convention on the Law Applicable to Contractual Obligations. Moreover, the rationale behind the change - that choice of law rules should operate to fulfil foreign rights - is in conformity with the approach of the majority of this Court in Breavington in the context of interstate torts. In the circumstances, and notwithstanding Pedersen v. Young, it is appropriate to reconsider the important question whether statutes of limitation should be classified as procedural or substantive, at least in the context of a conflict of laws problem within the Australian federation. In that reconsideration we should bear in mind both the purposes of choice of law rules and the undesirability of the practice of forum shopping to secure a result otherwise not available in the forum of the law of the cause.

24. It is convenient to begin that reconsideration by observing that the analysis which maintains that most statutes of limitation merely deny a remedy while not touching a plaintiff's rights seems to overlook the fact that the effect of such statutes is to confer upon a defendant a very important right - the right to plead the limitation period as an absolute defence. Wolff, Private International Law, 2nd ed. (1950) p 232; The Commonwealth v. Verwayen (1990) 170 CLR, per Brennan J. at p 425; Dawson J. at p 455. The classification of statutes of limitation as procedural provides a ready means for a plaintiff to circumvent this right of the defendant by choosing to bring proceedings in a forum which has a longer limitation period than that prescribed by the law of the cause. Furthermore, when a classification of limitation provisions as procedural produces an outcome in the forum which differs from the outcome which would result from an adoption of the law of the cause, including the limitation period prescribed for the action under that law, that classification seems to give scant recognition to the substantive operation of limitation provisions. It has the undesirable consequence of frustrating the operation of choice of law rules. It also has the potential to frustrate the very object of statutes of limitation. Their purpose is to prevent stale claims and therefore to relieve potential defendants from the uncertainty that such claims may be brought against them. In circumstances where the limitation period of the law of the cause does not coincide with that of the forum in which the cause is heard, the classification of statutes of limitation as procedural serves to frustrate their aim. A defendant will be left uncertain as to his or her liability and injustice may result, especially in circumstances where the limitation period of the forum is longer than that allowed by the law of the cause. Wolff, op. cit., pp. 232-233; English Law Commission, op. cit., p 8.

25. Moreover, it is simply no longer accepted that all matters which touch upon a remedy are necessarily to be treated as procedural or adjectival. Thus, the question of what heads of damage are recoverable is now treated as a substantive issue. Chaplin v. Boys (1971) AC 356, and a statute placing a limit on the amount of damages recoverable is not regarded simply as a matter concerning quantification of damage (which, on traditional analysis, has been treated as a procedural consideration) but rather as an express limitation on substantive liability. Livesley v. Horst Co. (1925) 1 DLR 159; Panozza and Co. Pty. Ltd. v. Allied Interstate (Q.) Pty Ltd (1976) 2 NSWLR 192; Breavington v. Godleman. In Breavington I rejected the notion that the principles according to which damages for personal injury are to be assessed is a matter of procedure and concluded that the measure of damages is plainly a question of substantive law (1988) 169 CLR at p 79. It follows that, even if it be correct to say that a statute of limitations only affects the availability or otherwise of a remedy, that circumstance, of itself, should not dictate that statutes of limitation be classified as procedural.

26. The equation drawn between matters relating to a remedy and matters of procedure has been criticized. Falconbridge, Essays on the Conflict of Laws, 2nd ed. (1954) p 308, has observed:

"On principle it is difficult to understand why a rule of law which denies a right of action should be construed as procedural. Even if a right of action is sometimes regarded as in the nature of a remedy, remedy is a wider concept than procedure."

Perhaps an explanation for the acceptance of an equivalence between remedy and procedure lies in the fact that, as a statute of limitations which extinguishes a right is considered to be substantive in effect, then it was thought, as a matter of symmetry, that statutes of limitation which only denied a remedy but which did not extinguish the underlying right were procedural. But considerations of symmetry should not be permitted to outweigh considerations of logic and reason. True it is that the operation of a statute of limitations does not touch the merits of a case. It does not follow from this, however, that the operation of such a statute may not affect the outcome of a case, i.e., that it may not have a substantive effect. Even Story, a proponent of the orthodox view that statutes of limitation should be classified as procedural, acknowledged that "(i)t is true, as Baldus contends, that the Statute of Limitations or Prescription does go to the decision of the cause". Commentaries on the Conflict of Laws, 7th ed. (1872) p 722. He went on to observe, however, that:

"that is not the question. The question is, whether it is a matter of the original merits, as, for example, a question of the original validity, or interpretation, or discharge of a contract, or whether it is a matter touching the time and mode of remedial justice, which is provided by law to redress grievances, or to prevent wrongs, or to suppress vexatious litigation."

Thus, Story appears to have drawn a line between those matters going to the original merits of the case and those going to the remedy, classifying the former as substantive and the latter as procedural. So understood, his classification of statutes of limitation as procedural is in conformity with the traditional approach described by Dicey, namely, the tendency to give a broad and liberal meaning to matters of procedure. However, a dividing line which is predicated upon whether a rule touches the merits or not tends to overlook the fact that a party may choose to bring an action in a particular forum in order to achieve a result, albeit "on the merits", which would not necessarily be available in the forum whose law governs the cause. A dividing line between substance and procedure which lends itself to manipulation in this way is not in harmony with the concerted effort demonstrated by the courts in other areas to guard against forum shopping.

27. True it is that all rules typically classified as procedural have the potential to affect substantive rights. Ailes, op. cit., p 408. See also Nygh, Conflict of Laws in Australia, 5th ed. (1991) p 217. But such an objection does not eliminate the need to draw a line between substance and procedure and to identify the characteristics which would enable us to distinguish between the two.

28. Professor Cook, in his article "'Substance' and 'Procedure' in the Conflict of Laws" (1933) 42 Yale Law Journal 333, at pp 343-344 argued that the line between substance and procedure "does not 'exist', to be discovered merely by logic and analysis" but may be drawn by reference to the answer to the question: "How far can the court of the forum go in applying the rules taken from the foreign system of law without unduly hindering or inconveniencing itself?" He contends that the forum court should apply the procedural rules of the law of the cause so long as it is not unduly inconvenient to apply them. In this way the procedural rules of the law of the forum would have a limited and largely residual operation.

29. Within the Australian federation, one should have thought that it would not be unduly inconvenient to apply the procedural rules of the law of the cause especially now that, in a slightly different context, there is a statutory precedent for so doing. Jurisdiction of Courts (Cross-vesting) Act 1987 (Cth) s. 11(1)(c). Certainly, in the case of statutes of limitation, it is difficult to see what inconvenience or hindrance would be caused to a forum court in giving effect to the limitation period prescribed by the law of the cause. Heavner v. Uniroyal, Inc. (1973) 305 A 2d, at p 415. The text of such a statute would be readily available and, so long as its application did not depend on knowledge of and reference to factors peculiar to the enacting State, its application by the forum court would be productive of no inconvenience. If a statute of limitations forming part of the law of the cause is classified as "substantive", the forum court will apply it in any event. This exercise does not appear to involve the courts in any undue inconvenience. But the criterion of "inconvenience" is too vague to serve as a definition of principle when the classification of a rule as procedural or substantive may materially affect the outcome of a case.

30. For the purposes of private international law, an appropriate criterion may be formulated by reference to the principal reason why it is necessary to draw a distinction between matters of substance and procedure. This reason, as has been seen, is associated with the efficiency of litigation. That efficiency is achieved by the adoption and application of the rules of practice and procedure and by the judges' practical familiarity with those rules. With this in mind, the essence of what is procedural may be found in those rules which are directed to governing or regulating the mode or conduct of court proceedings. In Breavington, Deane J. stated that "a procedural or adjectival law directed to regulating court proceedings will have its territorial operation in the territory in which those court proceedings take place" (1988) 169 CLR at p 136 (emphasis added). Sykes and Pryles have described rules which regulate court proceedings as "the mechanism of litigation" op. cit., p 257. Others have described these rules as the "machinery of litigation". The phrases "regulation of court proceedings" and mechanism or "machinery of litigation" describe rather than define the criterion which should determine what laws of the forum should apply in a case where the law of the cause is other than the law of the forum.

31. Can it be said that either s.36(1) of the Limitation of Actions Act or s.82(2) of the Workers Compensation Act forms part of the mechanism or machinery of litigation or is directed to the regulation of the mode or conduct of court proceedings? In my opinion this question must be answered in the negative. The limitation imposed by s.82(2) is directed to a potential plaintiff and, to adopt the words of Mahoney J.A in Commonwealth of Australia v. Dixon (1988) 13 NSWLR at p 619 (emphasis added):

"(T)he effect of such a limitation is to grant to a defendant a privilege, which he may invoke at his option, to defeat the plaintiff's right. If he invokes it, then the plaintiff's right is, to that extent, qualified or defeated."

Though s.36(1) is not, in terms, directed to a plaintiff as is s.82(2), its effect is the same. It is difficult to see how either provision, until invoked by a defendant, has any bearing on court proceedings and even when invoked how it may be said to be connected to the regulation of the mode or conduct of those proceedings.

32. Against this argument it might be contended that s.48 of the Limitation of Actions Act at least gives both s.36(1) and s.82(2), by extension, the necessary regulatory character to warrant them being classified as procedural. Section 48(1) provides that, where an Act, regulation, rule or by-law prescribes or limits the time for instituting an action, a court may extend the time so prescribed or limited to such an extent and upon such terms (if any) as the justice of the case may require.

33. Section 48(3)(b) states that the section does not:

"empower a court to extend a limitation of time prescribed by this Act unless it is satisfied -

(i) that facts material to the plaintiff's case were not ascertained by him until some point of time occurring within twelve months before the expiration of the period of limitation or occurring after the expiration of that period and that the action was instituted within twelve months after the ascertainment of those facts by the plaintiff;

or (ii) that the plaintiff's failure to institute the action within the period of the limitation resulted from representations or conduct of the defendant, or a person whom the plaintiff reasonably believed to be acting on behalf of the defendant, and was reasonable in view of those representations or that conduct and any other relevant circumstances,

and that in all the circumstances of the case it is just to grant the extension of time."

34. Can s.48, as it affects s.36(1) and s.82(2), be said to be directed to regulating the mode or conduct of court proceedings? An argument in favour of an affirmative answer to this question draws some support from the observation of Windeyer J. in Hoogland. That case concerned s.63(3) of the Workers' Compensation Act 1926 (N.S.W.). His Honour commented (1962) 108 CLR at p 489:

"That the sub-section does not impose an absolute bar on the expiration of a specified period, but provides for its extension, emphasizes its essentially procedural character."

However, Hoogland was not a case which was concerned with the classification of statutes of limitation in the context of a conflict of laws. Moreover, although it may be correct to say that the existence of a provision whereby a court may extend a limitation period means that a defence based on the expiry of a limitation period is not an absolute defence, that is not to deny its character as a substantive defence.

35. It is clear that, just as it is up to a defendant to plead a statute of limitations, it is up to a plaintiff to apply to the court for an extension of time, that is to say, the initiative of one of the parties is a prerequisite to any decision of a court to grant an extension of time. A court may not of its own motion grant an extension of time within which to commence an action pursuant to s.48. When an extension of time is granted, although s.48(1) gives the impression that a court has a wide discretion, that section is subject to sub-s.(3)(b) which prescribes definite criteria which are to govern the granting of any extension of time. The basis for granting an extension of time is not determined by reference to what a court considers is best for the efficient and expeditious disposition of litigation before it, but rather by reference to the position, circumstances and knowledge of the parties, particularly the plaintiff. A court's discretion to grant an extension of time will only be exercised when the court is of the view that, in all the circumstances of the case, it is just to grant an extension of time. It is difficult to relate such a discretion, so expressed, simply to the regulation of the mode or conduct of court proceedings.

36. In my opinion, therefore, s.48 does not dictate the result that s.36(1) of the Limitation of Actions Act or s.82(2) of the Workers Compensation Act must be classified as procedural. When account is taken of the circumstances and times in which the classification of statutes of limitation as procedural first occurred and it is accepted, as it must be, that the right/remedy distinction is artificial and that, in reality, statutes of limitation confer an important right or privilege on a defendant which, if exercised, is intended to have substantive consequences, then the conclusion is compelling that statutes of limitation should be classified as substantive and available to be pleaded as a good defence in circumstances where the statute of limitations forms part of the law of the cause applied by the forum court according to its choice of law rules. Such a conclusion also obviates the need to consider whether a given statute of limitations extinguishes both the right and the remedy or only withholds the latter.

37. Whether or not the plaintiff's decision to commence proceedings in New South Wales in this case could be described as "forum shopping" is not to the point. It is sufficient to note that the conclusion to which I have come provides an obvious disincentive to such a practice. It assists in ensuring that in Australia, at least, different legal consequences do not flow from a single set of events.

38. Having concluded that s.36(1) and s.82(2) should be given effect as part of the law of the cause, one question which remains is whether the plaintiff in this case could apply for an extension of the limitation period and, if so, to what court and under what Act could he apply. As the action has been commenced in the Supreme Court of New South Wales, the matter is not before the Supreme Court of South Australia and there would be no jurisdiction in that Court to grant an extension of time. The question then raised is whether the Supreme Court of New South Wales has jurisdiction to extend the limitation period. Although s.58 of the Limitation Act (N.S.W.) confers jurisdiction on the Supreme Court, in a cause of action founded on negligence for damages for personal injury, to order an extension of the limitation period for the cause of action, the limitation period in relation to which an extension may be made is that prescribed by the Limitation Act. As has been seen, that limitation period has no application to the plaintiff's cause of action.

39. Does the Supreme Court of New South Wales have jurisdiction to extend the time in which the plaintiff may commence his action under s.48 of the Limitation of Actions Act? In my view it does. In its terms, s.48(1) states that "where an Act ... prescribes or limits the time for ... instituting an action ..., a court may extend the time so prescribed or limited". Emphasis added. No territorial restriction is placed on the "court" to which s.48 refers. This is not to say that the Limitation of Actions Act has an extended or extraterritorial operation. Breavington v. Godleman (1988) 169 CLR, per Wilson and Gaudron JJ. at pp 98-99. Nor would I wish to be taken as implying that the jurisdiction of the Supreme Court of New South Wales to extend the time within which an action may be brought under s.48 depends upon the absence of such an express territorial restriction. Once it is accepted that s.36(1) and s.82(2) may be pleaded as substantive defences and have operation by virtue of the choice of law rule applicable in New South Wales, then it is appropriate that all South Australian provisions related to those defences should be applied by the forum court. Those provisions are picked up and applied as part of the relevant New South Wales law. There is no legitimate basis for picking up anything less than the entirety of the substantive law of the cause. To do so would be productive of injustice to a plaintiff. Picking up all provisions relating to substantive defences of the law of the cause ensures that the plaintiff is not denied the right he would have had had the action been brought in South Australia, namely, the right to apply for an extension of time.

40. This conclusion may also be justified by reference to s.118 of the Constitution which requires that "(f)ull faith and credit shall be given, throughout the Commonwealth to the laws ... of every State". When that section was considered in Merwin Pastoral Co. Pty. Ltd. v. Moolpa Pastoral Co. Pty. Ltd. (1933)48 CLR 565, a case in which the Supreme Court of Victoria refused to recognize a New South Wales statute on grounds of public policy even though the proper law of the contract under consideration was that of New South Wales, Evatt J. observed, ibid., at p 588 that "(i)n the United States the constitutional provision from which our sec.118 is taken has been regarded as prohibiting a refusal by the Courts of one State 'to give effect to a substantive defence under the applicable law of another State'". Bradford Electric Light Co. v. Clapper (1932) 286 US 145, at p 160. The notion of "giving effect" to a substantive defence involves giving full effect to it in circumstances where that defence, or more particularly the law of a State creating that defence, has been held, by choice of law, to apply in another State. As Dawson J. put it in Breavington (1988) 169 CLR at p 150, "(o)nce the choice is made, then full faith and credit must be given to the law chosen". In the context of this case that means that the limitation defence and all provisions relating to that defence (which necessarily includes those provisions such as s.48 which a plaintiff may seek to rely on to attempt to overcome what would otherwise be a good defence) should be given full effect by the forum court.

41. Although I have come to a conclusion in this case without reference, apart from the previous paragraph, to s.118 of the Constitution, because the scope of s.118 was the subject of considerable argument before the Court, I should repeat the view which I stated in Breavington, ibid., at pp 82-83. See also per Wilson and Gaudron JJ. at p 98; Brennan J. at pp 116-117; Dawson at p 150, namely, that s.118 is not a choice of law provision. The case currently before the Court does not involve a choice of law question. As was stated at the outset of these reasons, it is not in dispute between the parties that the cause of action arises under and should be determined in accordance with the law of South Australia. The applicable law of South Australia relevantly includes the defences afforded by s.36(1) of the Limitation of Actions Act and s.82(2) of the Workers Compensation Act. In so far as s.48 of the Limitation of Actions Act may affect those substantive defences, full faith and credit in the sense of full effect should be given to it by the Supreme Court of New South Wales in the event that the plaintiff seeks an extension of time.

42. As the separate trial which has been removed into this Court is confined to the defence raised under s.36 of the Limitation of Actions Act, there is a technical difficulty in granting to the defendant leave to amend so as to plead the defence under s.82(2) of the Workers Compensation Act. For present purposes it will be sufficient if the matter is dealt with on the footing that the defence pleaded under s.36 of the Limitation of Actions Act is a good defence, subject to any extension of time which may be granted in accordance with the provisions of s.48 of that Act. Should it become necessary to do so in order to resolve the litigation, an application could be made to the Supreme Court for leave to further amend the defence by pleading s.82(2) of the Workers Compensation Act, though I cannot presently foresee events which would make that course necessary or desirable, as s.82(2) and s.36 cover the same ground.

43. In the result I would make the following orders: 1. Declare that the defendant is entitled to judgment by virtue of

s.36(1) of the Limitation of Actions Act, subject to the grant of an extension of time under s.48 of that Act.

2. Remit the action to the Supreme Court of New South Wales to be

dealt with in accordance with this judgment.

3. Order that the plaintiff pay the defendant's costs of the

proceedings in this Court.

BRENNAN, DAWSON, TOOHEY AND McHUGH JJ. The circumstances which gave rise to the plaintiff's action in the Supreme Court of New South Wales are set out in the Chief Justice's reasons for judgment. It is unnecessary to repeat them. It suffices to note the following:

2. On or about 17 February 1984, the plaintiff entered into ship's articles of agreement with the defendant in Port Adelaide and was employed as a marine steward on the motor vessel "Troubridge" plying between South Australian ports. On or about 22 February 1984, when the vessel was tied up beside a wharf at Port Lincoln in South Australia, the plaintiff was allegedly injured in the course of lifeboat drill. On 4 January 1990, he commenced proceedings in the Supreme Court of New South Wales against the defendant claiming damages for common law negligence. The originating process was served on the defendant out of the State, presumably pursuant to r.1(10)(e) of Pt 10 of the Rules of the Supreme Court (N.S.W.) which authorizes service out of the jurisdiction when the proceedings are founded on damage suffered wholly or partly in New South Wales caused by a tort wherever occurring. By its amended defence the defendant pleaded that "the plaintiff's claim is statute barred by virtue of Section 36 Limitation of Actions Act, 1936 South Australia".

3. Section 36(1) of the Limitation of Actions Act 1936 (S.A.) reads as follows:

" All actions in which the damages claimed consist of or include damages in respect of personal injuries to any person, shall be commenced within three years next after the cause of action accrued but not after."

However, a power to extend time is provided by s.48(1) of the Limitation of Actions Act, the material part of which reads:

" Subject to this section, where an Act ... prescribes or

limits the time for - (a) instituting an action; ... a court may extend the time so prescribed or limited to such an extent, and upon such terms (if any) as the justice of the case may require."

4. Sharpe J. ordered that there be a separate trial of the question "whether the defendant is entitled to judgment by virtue of Section 36 of the Limitation Act 1936, South Australia". An order was made by this Court that "the separate trial ... ordered by Mr Justice Sharpe ... be removed into this Honourable Court pursuant to Section 40(1) of the Judiciary Act 1903". In this Court the defendant sought leave to amend its defence to add a further plea based on s.82(2) of the Workers Compensation Act 1971 (S.A.) which reads as follows:

" Where a worker has received or is entitled to receive compensation under this Act or under the repealed Act in respect of an injury, he shall not bring an action against the employer for damages in respect of the same injury unless he commences that action within three years from the day on which that injury occurred."

The plaintiff contends that he has not received and is not entitled to receive compensation "under this Act", but it is not necessary to determine that issue. In Karasaridis v. Kastoria (1984) 37 SASR 345, at pp 353-363, the Full Court of the Supreme Court of South Australia held that the provisions of the two Acts limiting time operated concurrently and that both provisions were subject to the power to extend time contained in s.48(2) of the Limitation of Actions Act. If this view be correct - and we assume that it is in the absence of contrary submission - the question whether one provision or the other is applicable can be put aside until it is decided whether a South Australian Act prescribing the time within which an action for damages for personal injuries must be brought bars the action commenced outside that time in the Supreme Court of New South Wales. The action was not barred by the Limitation Act 1969 (N.S.W.).

5. To answer the question whether the plaintiff's action in the Supreme Court of New South Wales is barred by a South Australian Act, it is necessary to consider three issues: 1. What is the system of law by reference to which the Supreme

Court of New South Wales is to determine the question?

2. What is the content of the applicable law? 3. What effect does the applicable law attribute to the South

Australian statute of limitations?

Separation of the elements of the problem in this way assists in avoiding an ambiguity that sometimes appears in the use of the term "lex fori". References to the lex fori to indicate the system of law which furnishes the conflict of law rules to be applied by a court should be distinguished from references to the lex fori when that term is used in reference to a specific rule, which is the forum's domestic law made applicable to a particular case by the conflict of law rules. The system of law.

6. It is axiomatic that the governing system of law which furnishes the conflict of law rules in this case is the law binding on the Supreme Court of New South Wales: the lex fori in the former sense. The system can be more precisely described by reference to the source of the relevant New South Wales body of law. Prior to Breavington v. Godleman (1988) 169 CLR 41, it was generally understood that the rules for resolving conflicts of law in matters litigated in the forum were furnished by the common law of the forum: see, for example, Koop v. Bebb (1951) 84 CLR 629, at pp 641-642, 644; Pedersen v. Young (1964) 110 CLR 162, at pp 169-170; Pozniak v. Smith (1982) 151 CLR 38, at pp 49,52; Anderson v. Eric Anderson Radio and T.V. Pty. Ltd. (1965) 114 CLR 20, at pp 30-31. In Breavington v. Godleman, however, differing views were expressed. Wilson and Gaudron JJ. held that s.118 of the Constitution required that a uniform legal consequence should attach to any set of facts occurring in a State wherever an action is brought and that that imperative can be satisfied only by an inflexible rule that liability for extraterritorial acts occurring in Australia be determined exclusively by the substantive law of the State "in which the events occurred" (1988) 169 CLR at p 98. The constitutional imperative favoured the adoption of a corresponding common law rule to be applied to events occurring in a Territory, ibid., at p 98. Their Honours, ibid., at p 99, explained how the constitutional imperative (and the corresponding common law rule) operate:

"(Section) 118 dictates a consequence to which State choice of law rules must conform. In so far as the law of a State or Territory is thus applied in another State or Territory this occurs not by reason of extended or extraterritorial operation of that State or Territory law, but by operation of the choice of law rule applicable in the State or Territory where the matter falls for adjudication."

On this approach, the conflict of law rules which would otherwise be applied by a court of the forum are modified (or replaced) in their application to torts occurring within Australia in consequence of the effect attributed to s.118 of the Constitution. Deane J. held that the Constitution had been effective to create "a unitary national legal system", ibid., at p 124, which attributes to the laws of the respective States the function of regulating "conduct and things within (or with sufficient relevant nexus with) their respective territories", ibid., at p 135. The laws of the Territories, like the laws of the States, are part of the national legal system and are subject to like limitations on their territorial operation, ibid., at pp 137-138. On this approach, the courts of the forum are not troubled by conflicts of law in cases of intra-national torts for there are no conflicts. In a particular case, the same relevant part of the national law alone applies to determine substantive liability, the law of the forum contributing only the procedural or adjectival law applicable to the proceedings, ibid., at p 83.

7. These were the only judgments which purported to depart from the established view that the governing system of law in cases of extraterritorial tort is the common law of the forum. Mason C.J. rejected the notion that the Constitution furnishes a solution to conflict of laws problems within Australia, ibid., at p 83. Noting that "the relevant American law is not dictated by the provisions of the Constitution", ibid., at p 82; see also the majority opinion in Sun Oil Co. v. Wortman (1988) 486 US 717 (100 L Ed 2d 743) which denies that the full faith and credit clause dictates the content of conflict of law rules (at p 727 (p 755)) and the minority view that the clause merely restrains the forum State from applying unfair and arbitrary rules (at pp 740-742 (pp 764-765)), his Honour concluded, ibid., at p 83:

"It is preferable that Parliament should provide a solution by an exercise of legislative power, if that be legitimate, than that the Court should spell out a rigid and inflexible approach from the language of s.118."

Brennan J., ibid., at pp 107,118, Dawson J., ibid., at p 150 and Toohey J., ibid., at pp 157-158, each accepted, either expressly or implicitly, the position that the Supreme Court of Victoria (the Court of the forum) was bound to apply conflict of law rules to be found in the common law applied by that Court.

8. In the absence of any overriding constitutional or statutory provision, a State or Territorial court exercising jurisdiction in a claim for damages for a tort which has occurred outside the jurisdiction, but within Australia, must apply the common law rules governing conflict of laws. (The problem for federal courts exercising Australia-wide jurisdiction and bound by s.79 of the Judiciary Act 1903 (Cth). See Robinson v. Shirley (1982) 149 CLR 132, at p 136; Pozniak v. Smith (1982) 151 CLR at p 54, raises additional questions for consideration which need not delay us here.) The common law rules, being part of the municipal law binding on the court, are prima facie amenable to variation by the forum legislature.

9. The power of a State to enact laws governing the procedure of its courts is unquestioned, but a constitutional imperative that the courts of a State should apply only the substantive law of another part of the Commonwealth in determining a claim for damages for a tort occurring outside the State but within Australia would deny the forum State an important legislative power. If there were such a constitutional imperative, a State law which prohibits the bringing of an action of a particular kind (for example, an action for damages for personal injury where the injury is not serious) or which creates a particular defence (say, in defamation) would be applied by the courts of the State to claims arising from intra-jurisdictional torts but would be constitutionally ineffective in respect of claims arising from torts occurring outside the State but within Australia. Although it would be wrong to deny that an implication properly drawn from the text of the Constitution can limit the legislative power of a State as effectively as an express limitation such as s.92, we do not find in the Constitution sufficient indicia to raise such an implication. The majority in Breavington v. Godleman rejected such an implication and we adhere to that view.

10. To describe the States, as Windeyer J. once described them, Pedersen v. Young (1964) 110 CLR at p 170, as "separate countries in private international law" may sound anachronistic. Yet it is of the nature of the federation created by the Constitution that the States be distinct law areas whose laws may govern any subject matter subject to constitutional restrictions and qualifications. The laws of the States, though recognized throughout Australia, are therefore capable of creating disparities in the legal consequences attached in the respective States to the same set of facts unless a valid law of the Commonwealth overrides the relevant State laws and prescribes a uniform legal consequence. That may or may not be thought to be desirable, but it is the hallmark of a federation as distinct from a union. Far from eliminating the differential operation of State laws, s.118 commands that all the laws of all the States be given full faith and credit: the laws of the forum are to be recognized as fully as the laws of the place where the set of facts occurred. Section 118 would not be obeyed by refusing recognition to the laws of a forum State and by applying only the laws of the part of Australia in which the set of facts occurred. A disparity in legal consequences attached to a set of facts cannot be eliminated by refusing recognition to laws of the forum which create the disparity. In our respectful opinion, s.118 does not prescribe the selection of the lex loci delicti or other extraterritorial body of law as the exclusive body of law governing liability for extraterritorial torts. See also per Mason C.J. in Breavington v. Godleman (1988) 169 CLR at pp 81-83. The selection of the applicable rules governing liability is the function of the common law; s.118 provides for recognition by the courts of the forum of the rules so selected. This was explained by Dawson J. in Breavington v. Godleman, ibid., at p 150:

" In my opinion, the requirement that full faith and credit be given to the laws of a State, statutory or otherwise, throughout the Commonwealth, affords no assistance where there is a choice to be made between conflicting laws. Once the choice is made, then full faith and credit must be given to the law chosen but the requirement of full faith and credit does nothing to effect a choice. Nor is it to the point to say that the full faith and credit requirement assumes the applicability of a single law. No doubt that is so, but it is to say no more than that where there is a conflict of laws upon a given question a selection must be made before the question can be answered. The conflict rules are based upon the same assumption but they, unlike the full faith and credit requirement, provide a basis upon which the selection can be made. Section 118 of the Constitution is not directed to a conflict of laws; where there is a conflict it makes no choice or, to put it another way, does not require the application of a law which is not otherwise applicable."

11. The problem, then, is to ascertain the content of the rules of the common law by which the choice of law is made. Since at least Varawa v. Howard Smith Co. Ltd. (1911) 13 CLR 35, at p 69, rules of the common law have been applied for that purpose by courts in all parts of Australia, subject to local legislation. The content of the common law rules.

12. The rule in Phillips v. Eyre (1870) LR 6 QB 1, its consideration by this Court in a line of authoritative cases, the aberration introduced by Machado v. Fontes (1897) 2 QB 231, and the extirpation of that aberration by Chaplin v. Boys (1971) AC 356, are sufficiently described in judgments in Breavington v. Godleman (1988) 169 CLR see per Mason C.J. at pp 71-73; per Brennan J. at pp 108-111; per Dawson J. at pp 141-146; per Toohey J. at pp 155-158. Among the Justices who rejected a constitutional solution to the problem of conflict of laws, Brennan and Dawson JJ. substantially followed the rule in Phillips v. Eyre. Mason C.J., ibid., at pp 77-80, observing that the Phillips v. Eyre conditions "have little to offer and present a needless complication" in relation to foreign torts where the parties may have no substantial connection with the law of the place of the tort, contrasted the situation of parties resident in Australia suing in relation to a tort which occurred in another part of Australia. His Honour saw no reason (at least in the circumstances of that case) "to depart from the lex loci delicti as the primary or basic law to be applied", ibid., at p 79; and note a similar approach to the exercise of a discretionary power to remit in Pozniak v. Smith and Robinson v. Shirley. Provided effect be accorded to statutes which are part of the lex fori, this approach is similar in its operation to that favoured by Brennan and Dawson JJ. But his Honour's reference to the lex loci delicti as the primary law to be applied suggests that his Honour regarded the underlying principle, applicable to torts occurring outside as well as to torts occurring inside Australia, as having some degree of flexibility, sufficient to accommodate a case where the parties had no substantial connection with the lex loci delicti. We do not wish to foreclose. See per Dawson J. in Breavington v. Godleman (1988) 169 CLR, at pp 147-148, future consideration of the question whether the Phillips v. Eyre rule (as we state it in this judgment) is suitable for application to actions arising from torts occurring outside Australia but, mindful of the freedom of intercourse throughout this country and the general similarity of the laws in force in the various parts of Australia (to which the Chief Justice referred in Breavington v. Godleman, ibid., at p 78, the overwhelming desideratum in a rule for intra-national torts is certainty of application or, more accurately, as much certainty as the subject matter admits.

13. In Breavington v. Godleman, Toohey J., ibid., at p 162, was prepared to accept the general rule in Phillips v. Eyre but subject to the flexible exception enunciated by Lord Wilberforce in Chaplin v. Boys. His Honour was attracted to that view as being "less parochial and (capable of being) applied to give appropriate significance to the lex loci delicti and the lex fori in all the circumstances". However, the flexible exception would apply only in special circumstances where "the lex loci delicti has no real connexion with the proceedings", ibid., at p 163. As such circumstances would be out of the ordinary in cases of intra-national torts, and as it is now preferable to state the common law rules for application by Australian courts in cases of Australian torts in terms to which a majority of this Court assent, we think it right to adopt rules expressed in more categorical terms.

14. Those rules were formulated by Brennan J. in Breavington v. Godleman as follows, ibid., at pp 110-111:

"A plaintiff may sue in the forum to enforce a liability in respect of a wrong occurring outside the territory of the forum if - 1. the claim arises out of circumstances of such a character that, if they had occurred within the territory of the forum, a cause of action would have arisen entitling the plaintiff to enforce against the defendant a civil liability of the kind which the plaintiff claims to enforce; and 2. by the law of the place in which the wrong occurred, the circumstances of the occurrence gave rise to a civil liability of the kind which the plaintiff claims to enforce. This restatement is narrower in expression than the traditional formulation of the Phillips v. Eyre conditions which speak of 'a character that ... would have been actionable' and 'justifiable'. It defines more precisely the issues which are referred for determination to the lex fori and the lex loci respectively."

The reasons for judgment of Dawson J. in Breavington v. Godleman, ibid., esp. at pp 146-147, and in the associated case of Perrett v. Robinson (1988) 169 CLR 172, at p 186, accord with this formulation. It is, we think, the appropriate formulation in respect of intra-national torts though the present case suggests a need for a qualification or refinement next to be mentioned.

15. The common law rules founded on Phillips v. Eyre are part of the common law of the forum, prescribing the elements of a cause of action justiciable in the courts of the forum. But that view of the common law rules, coupled with a rejection of the vested rights theory, gives rise to a logical anomaly, for it postulates as many causes of action as there are forums in which the rules apply. The anomaly is more apparent than real. The causes of action arising in the several law areas of the Commonwealth are dependent on, and incorporate as an element, the existence of a civil liability under the lex loci delicti. Once any of the causes of action merges in a judgment, the original civil liability must be taken to be merged in the judgment and to have been extinguished. It is no longer available to found any other cause of action. It follows that the civil liability to which the law of the place in which the wrong occurred gave rise must be a continuing liability; if that civil liability be extinguished, the cause of action conferred by the common law of the forum is extinguished too. That qualification or refinement is of importance in the present case, for the same principle must apply whether the civil liability under the lex loci is extinguished by merger in a judgment, by accord and satisfaction, or by statute.

16. In argument, the plaintiff conceded that the substantive law of the place of the wrong is "imported into the forum" and that concession is consistent with the second part of the conflict of law rules as we have stated them. The defendant's contention, however, attributes to s.36(1) of the South Australian Limitation of Actions Act the effect of a substantive rule which denies the continuing availability of civil liability under South Australian law to found a cause of action under the law of New South Wales. The common law rules and extraterritorial statutes of limitation.

17. Traditionally, a distinction has been drawn between substantive law (the subject of the conflict of law rules) and procedural law which is governed by the lex fori alone. Though the dividing line is sometimes doubtful or even artificial, the need to distinguish between substantive law and procedural law is clearly recognized for a number of forensic purposes. Frankfurter J. drew attention to the variety of purposes for which the distinction is made in Guaranty Trust Co. v. York (1945) 326 US 99, at p 108:

" Matters of 'substance' and matters of 'procedure' are much talked about in the books as though they defined a great divide cutting across the whole domain of law. But, of course, 'substance' and 'procedure' are the same keywords to very different problems. Neither 'substance' nor 'procedure' represents the same invariants. Each implies different variables depending upon the particular problem for which it is used. ... And the different problems are only distantly related at best, for the terms are in common use in connection with situations turning on such different considerations as those that are relevant to questions pertaining to ex post facto legislation, the impairment of the obligations of contract, the enforcement of federal rights in the State courts and the multitudinous phases of the conflict of laws."

In this case it is necessary to determine whether s.36(1) of the South Australian Limitation of Actions Act is a procedural law or a substantive law for the purpose of applying the conflict of law rules governing extraterritorial torts. The precise question is whether s.36(1) leaves the civil liability under the law of South Australia intact or whether it is annihilative, extinguishing the civil liability and destroying the cause of action. The two classes of statute are well described by Jackson J. in Chase Securities Corp. v. Donaldson (1945) 325 US 304, at p 313:

" Statutes of limitations always have vexed the philosophical mind for it is difficult to fit them into a completely logical and symmetrical system of law. There has been controversy as to their effect. Some are of opinion that like the analogous civil law doctrine of prescription limitations statutes should be viewed as extinguishing the claim and destroying the right itself. Admittedly it is troublesome to sustain as a 'right' a claim that can find no remedy for its invasion. On the other hand, some common-law courts have regarded true statutes of limitation as doing no more than to cut off resort to the courts for enforcement of a claim."

18. For the purposes of applying conflict of law rules, English courts have long adopted the distinction that a true statute of limitation, which does no more than cut off resort to the courts for enforcement of a claim, is a procedural law, while a statute which extinguishes a civil liability and destroys a cause of action is a substantive law. The British Linen Company v. Drummond (1830) 10 B and C 903 (109 ER 683); Huber v. Steiner (1835) 2 Bing (NC) 202, at pp 210-211 (132 ER 80, at p 83); Lafond v. Ruddock (1853) 13 CB 813 (138 ER 1422); Don v. Lippmann (1837) 5 Cl and Fin 1, at p 13 (7 ER 303, at pp 307-308); Harris v. Quine (1869) LR 4 QB 653; Black-Clawson Ltd v. Papierwerke AG (1975) AC 591. The same distinction has been made in Scotland. Higgins v. Ewing's Trustees (1925) SC 440, and in the United States sometimes for purposes uniquely arising under the United States Constitution. Davis v. Mills (1904) 194 US 451, at p 454; Central Vermont Railway v. White (1915) 238 US 507, at p 511; Chase Securities Corp. v. Donaldson; Guaranty Trust Co. v. York (1988) 486 US (100 L Ed 2d), esp. at pp 726-727 (p 755). In Black-Clawson Ltd. v. Papierwerke AG. (1975) AC at p 630 Lord Wilberforce affirmed the existence of -

"the distinction made in English private international law between matters of substance and matters of procedure, and within that, the classification of limitation as a matter of procedure. Classification of limitation as procedural means that in proceedings in an English court, English law, as the lex fori, will apply its domestic law as to limitation and will not apply foreign limitation provisions even if the foreign law is the proper law, unless, at least, they extinguish the right."

19. In this Court the same distinction has been made. In Pedersen v. Young, where the question was whether a Queensland statute of limitations barred an action for damages for personal injury commenced in this Court in its diversity jurisdiction, Kitto J. said (1964) 110 CLR at p 166:

"It is the established doctrine of English law ... that statutes of limitation expressed to relate to the institution of proceedings do not provide substantive defences. They 'affect procedure, not substance': Cheshire, Private International Law, 6th ed. (1961) p.686. Cases such as Huber v. Steiner, Harris v. Quine and Alliance Bank of Simla v. Carey (1880) 5 CPD 429 are conclusive of the point."

And Menzies J. said (1964) 110 CLR at p 166:

" It is a well-established principle that statutes of limitation, except where title is affected, are rules of procedure only and form part of the lex fori. The reason why such statutes are so regarded is that they relate to the remedy and not the right. Since 21 Jac. I c. 16 the usual form of a statute of limitation has been to provide that an action shall not be brought except within the time stated after the cause of action arose. A statute in this form, notwithstanding any generality of expression, must of necessity be construed as applying to the commencement of actions in courts subject to the power of the legislature enacting it and to no other courts."

See also per Windeyer J., ibid., at p 169. Of course, if a statute of limitations which is part of the lex fori not only bars the bringing of an action after the expiration of a limitation period but extinguishes the civil liability which would otherwise exist under the lex fori as in s. 63 of the Limitation Act (NSW), the courts of the forum will give effect to the statute according to its tenor, that is, both as a procedural law and as a law having a substantive operation.

20. In John Robertson and Co. Ltd. v. Ferguson Transformers Pty. Ltd. (1973) 129 CLR 65, the effect of s.37 of the Limitation of Actions Act was considered. Section 37 is framed in terms identical with the terms of s.36(1): "All actions ... shall be commenced within ... years next after the cause of action accrued, but not after." Mason J. defined the effect of s.37 thus, ibid., at pp 92-93; see also pp 75,80,83 and 89:

" The section of the South Australian Act is a provision which bars the remedy, not the right. It is therefore a rule of procedure only (see Harris v. Quine; Alliance Bank of Simla v. Carey). As such it is a rule which applies to actions brought in the courts of South Australia and not elsewhere."

His Honour then cited the passage from the judgment of Menzies J. in Pedersen v. Young set out above. His Honour's construction of s.37 accords with the construction placed on the similarly worded provision of the Queensland Act considered in Pedersen v. Young (1964) 110 CLR see per Kitto J. at p 165; and see Bargen v. State Government Insurance Office (Q.) (1982) 154 CLR 318. We see no reason to depart from this view of s.37 of the South Australian Act or to refrain from applying it to s.36(1). It is a view which not only accords with a consistent line of authority in this Court. John Robertson and Co. Ltd v. Ferguson Transformers Pty Ltd (1973) 129 CLR at pp 75, 80, 92-93; Australian Iron and Steel Ltd v. Hoogland (1962) 108 CLR 471, esp. at pp 488-489; Noske v. McGinnis (1932) 47 CLR 563, at pp 586, 589, 590, 592-593; Maxwell v. Murphy (1957) 96 CLR 261, at pp 286-287; Fielding v. Doran (1984) 59 ALJR 511; 60 ALR 342, but it gives effect to a principle which, in 1932, Starke J. regarded as "established beyond doubt". Noske v. McGinnis (1932) 47 CLR, at p 589; see also per Lord Wilberforce in Black-Clawson (1975) AC at p 631.

21. A different kind of statute is one which creates a right of limited duration so that, after the expiry of the time prescribed, the right ceases to exist for any purpose. Such a statute is substantive in nature. The distinction between the two classes of statutes was stated by Windeyer J. in Australian Iron &Steel Ltd. v. Hoogland (1962) 108 CLR at p 488:

"It may be that there is a distinction between Statutes of Limitation, properly so called, which operate to prevent the enforcement of rights of action independently existing, and limitation provisions annexed by a statute to a right newly created by it. In the latter case the limitation does not bar an existing cause of action. It imposes a condition which is of the essence of a new right."

And in The Commonwealth v. Verwayen (1990) 170 CLR 394, at p 497 McHugh J. said:

"There is, of course, a fundamental difference between a true statute of limitation ... which bars stale claims and a limitation period annexed by a statute to a right which it creates. In the latter class of case, the limitation period will generally be of the essence of the right".

An example of a statutory right of limited duration can be found in Maxwell v. Murphy. There, a majority held that s.5 of the Compensation to Relatives Act 1897 (N.S.W.), which, following Lord Campbell's Act, conferred a right to sue within a prescribed time in respect of a death negligently caused, operated at the expiry of that time to extinguish the cause of action which it created. Fullagar J. dissented, construing the time provision of the Act as a true statute of limitation, barring a remedy but not extinguishing the right. Pointing out (1957) 96CLR at p 290, that true statutes of limitation are "classed as 'procedural' statutes, and therefore, for purposes of private international law, to be applied as part of the lex fori", his Honour added:

"Such a view may be open to criticism, but there is a good deal to be said for it. Such statutes do deal with remedies as such, and not with rights as such, and the distinction involved is not meaningless or unreal."

The meaning and reality of the distinction can be perceived more clearly where the statute bars an action to enforce a contractual proprietary, or possessory, right, for example, a right to a possessory lien; Spears v. Hartly (1800) 3 Esp 81 (170 ER 545), than where it bars an action to enforce a claim for damages in tort. Nevertheless, there is a real distinction between a statute which extinguishes a right to damages and a statute which bars an action to enforce a right to damages, as the decision of this Court in The Commonwealth v. Verwayen illustrates. At least the continued existence of a right to damages will furnish consideration for an agreement to settle the plaintiff's claim. But, whether or not a distinction between a statute extinguishing a right and a statute barring an action to enforce the right be thought desirable, it is firmly and clearly established as a principle of law. As the distinction has operated in practice free of injustice, there is no warrant for discarding it.

22. Section 36(1) of the Limitation of Actions Act should be held to be procedural in nature, as s.37 was held to be in John Robertson &Co. Ltd. v. Ferguson Transformers Pty. Ltd. Section 48 of that Act, which authorizes an order extending the time for bringing action, confirms that s.36(1) is procedural only and does not extinguish the civil liability which a plaintiff might bring action to enforce. See per Windeyer J. in Australian Iron and Steel Ltd. v. Hoogland (1962) 108 CLR at p 489. Being procedural, s.36(1) has no effect on the action brought by the plaintiff in the Supreme Court of New South Wales to enforce the cause of action existing under the common law applicable in that State. Although the terms of s.36(1) are not the same as the terms in which s.82(2) of the Workers Compensation Act is drawn, the difference does not suggest that the latter provision would operate differently from the former. That being so, it is appropriate to allow the defendant to amend its defence further to raise s.82(2), and to amend the question which Sharpe J. ordered to be decided separately to cover s.82(2). The amended question should be answered as follows:

12. It follows from what has been written above that ss.36(1) and 48 of the Limitation Act and s.82(2) of the Compensation Act are all applicable, as part of the substantive law of South Australia, to govern the rights and liabilities of the parties arising out of the South Australian accident in which the plaintiff sustained his injuries. Conversely, New South Wales limitation provisions, including provisions conferring power to extend applicable limitation provisions, are substantive laws of New South Wales which are inapplicable to proceedings by the plaintiff against the defendant in the present case. There remains for consideration the question whether the Supreme Court of New South Wales possesses jurisdiction, in those proceedings, to extend the limitation period prescribed by s.36(1) of the Limitation Act (and, if pleaded by the defendant, s.82(2) of the Compensation Act) pursuant to s.48 of the Limitation Act. I agree, for the reasons which he gives, with the conclusion of the Chief Justice that it does.

13. There are two further matters which should be briefly mentioned. The first is that it is unnecessary for the purposes of the present case to consider whether a State law whose practical effect is to preclude proceedings in some or all of the State's courts after the expiry of a designated period might be so framed that it is, for relevant purposes, properly to be characterized as merely procedural. It may, for example, be arguable that a law which was merely concerned with defining the jurisdiction of some or all of the courts of the State in a way which would indiscriminatingly protect them from the delays and complications of stale claims would properly be so characterized if it neither directly barred the remedy or the bringing of proceedings in a particular class or particular classes of case nor purported to enable proceedings to be brought after the remedy or proceedings were barred by some applicable substantive law, be it the law of the particular State or the law of some other State. It suffices, for present purposes, to say that none of the relevant limitation provisions of the States of South Australia and New South Wales is a law of that character.

14. The other further matter is that, as I acknowledged in Breavington, the Constitution's resolution of conflict or competition between the laws of different States of the Commonwealth does not preclude uncertainty nor render the rules of private international law irrelevant in all circumstances. In a case where there is a substantive nexus with the territory of more than one State, the determination of predominant territorial nexus may well involve a discretionary weighing of competing factors, including considerations of what is fair and just, in which the rules of private international law may be important by way of analogy. Any uncertainty or flexibility involved in that process is, however, an aspect of the determination of the content of the applicable law in the circumstances of the case. It does not undermine or destroy the essential unity of the Australian legal system since the reference point - i.e. predominant territorial nexus - will be the same regardless of the place within the Commonwealth in which the proceedings are brought. Nor does the Constitution's resolution of conflict or competition between the laws of different States preclude State legislation to avoid the existence of such conflict or competition by providing that, in certain circumstances, the law of another State (or Territory) should be adopted and applied, as part of the legislating State's own law, to matters within its territorial competence.

15. I agree with the orders proposed by the Chief Justice.

GAUDRON J. In Breavington v. Godleman (1988) 169 CLR 41, a majority of this Court decided, contrary to what had been said in Koop v. Bebb (1951) 84 CLR 629 that the law of the forum does not automatically apply to determine liability for torts committed in Australia. That conclusion was reached by different paths, Mason C.J. (1988) 169 CLR at p 78, resting his conclusion on the view that the rules developed in relation to "the international scene" are not necessarily appropriate in Australia which is "one country and one nation", Wilson J. and I, ibid., at p 99, and Deane J., ibid., at pp 126-130, concluding that the approach taken in Koop v. Bebb was inconsistent with the mandate in s.118 of the Constitution.

2. This case raises a further question in the interjurisdictional area considered in Breavington. The primary question is whether a South Australian limitation period applies to an action commenced in the Supreme Court of New South Wales in respect of an employment accident in South Australia. If it does, a further question arises whether the Supreme Court of New South Wales may extend that period pursuant to s.48 of the Limitation of Actions Act 1936 (S.A.).

3. The oral arguments of the parties and interveners in this case accepted the authority of Breavington in so far as it overruled Koop v. Bebb. And the arguments proceeded on the basis that Breavington requires that, in the circumstances of this case but subject to the question whether the action is statute barred, the liability of R.W. Miller and Company (South Australia) Pty. Limited ("the employer") to pay damages to its injured employee, William McKain, is to be determined by application of the law of South Australia. Indeed, it was put by Mr Wheelahan, senior counsel for Mr McKain, that Breavington decided that, "for torts committed within Australia, the law to be applied is the substantive law of the lex loci delicti". That being so, this case must be decided on the basis that liability is to be determined by application of the substantive law of South Australia and not that of New South Wales.

4. Although the arguments accepted the authority of Breavington in so far as it overruled Koop v. Bebb, the parties and the interveners sought to use the decision in that case in different ways in relation to the issues in this case. And, a number of the arguments directly challenged the view that s.118 of the Constitution is concerned with the issue that arose in Breavington or with the issues in this case.

5. I remain convinced that the Constitution makes provision with respect to the issue that arose in Breavington. And I consider that, in the final analysis, the issues in this case must also be answered by reference to the Constitution. However, a reconsideration of Breavington leads me to think that some aspects of what Wilson J. and I then said might conveniently be stated in different terms.

6. In Breavington the issue was identified in terms of "conflict of laws" and resolved in terms of "choice of law". Those terms can be misleading. The term "conflict of laws" suggests that different legal regimes might simultaneously govern the one set of facts. The term "choice of law" suggests that the legal consequences which attach to a set of facts may be determined by choosing between different legal regimes, including a regime which had no application to the facts when they occurred, but which is treated as applicable because it is the regime of the jurisdiction in which proceedings are brought. The suggestions implicit in the terms "conflict of laws" and "choice of law" may obscure the essential nature of the solution provided by the Constitution in cases like Breavington.

7. The constitutional solution operates at two stages. At the first stage, it eliminates "conflict of laws". More precisely, it brings about a situation such that, as between the States, the Territories and the Commonwealth, there is only one body of law which applies to any given set of facts. That is achieved by covering cl.5 and by ss.106, 107, 108, 109 and 118 in Ch.V of the Constitution which, when taken together, leave no room for the notion that the one set of facts might, within Australia, simultaneously be subject to different legal regimes.

8. The second stage of the constitutional solution eliminates "choice of law". The Constitution does not permit of the possibility that the legal consequences attaching to a set of facts occurring in Australia might be determined other than by application of the body of law governing those facts. That is the immediate and direct effect of covering cl.5 which provides that the Constitution and the laws of the Commonwealth are "binding on the courts, judges, and people of every State and of every part of the Commonwealth, notwithstanding anything in the laws of any State" and s.118 which requires that "(f)ull faith and credit ... be given, throughout the Commonwealth to the laws, the public Acts and records, and the judicial proceedings of every State."

9. Save to the extent that s.109 provides with respect to inconsistency between a law of a State and a law of the Commonwealth, the Constitution does not, in terms, expose the body of law applicable in any particular case. However, and leaving aside the situation in which Commonwealth law applies, because the States have identifiable territorial limits, the applicable body of law will ordinarily be that of the State in which the events in question occurred. Of course, there may be difficulties even in determining that issue, particularly where legal consequences are said to flow from a failure to do something or where acts occur partly in one State and partly in another. As to these problems in an international context, see Voth v. Manildra Flour Mills Pty Ltd (1990) 171 CLR 538, at pp 567-569. And, there may be cases in which the place of some act is entirely fortuitous or, for some reason, some other State has a more substantial connection with it than the State in which it occurred. These are matters to be taken into account in identifying the governing body of law.

10. As already indicated, it is not in issue that liability is to be determined in accordance with the law of South Australia. But it is argued on behalf of Mr McKain that the question whether his action is statute barred raises an issue of a different kind. The issue, it is said, is not liability but whether his action can proceed in the Supreme Court of New South Wales. And that distinction is treated as equivalent to or based upon the distinction between "procedural" and "substantive" laws - a distinction which, so long as it is applied to distinguish laws which are "directed to regulating court proceedings". Breavington v. Godleman (1988) 169 CLR, per Deane J. at p 136, from those which operate to identify and define rights, is both convenient and longstanding. It has always been accepted that, no matter that a court is required to apply the substantive law of some other jurisdiction, it should apply its own practices and procedures. See, for example, De La Vega v. Vianna (1830) 1 B and Ad 284, at p 288 (109 ER 792, at p 793); Huber v. Steiner (1835) 2 Bing (NC) 202, at pp 210-211 (132 ER 80, at p 83); Leroux v. Brown (1852) 12 CB 801, at pp 803, 827 (138 ER 1119, at p 1121, 1130); Phillips v. Eyre (1870) LR 6 QB 1, at pp 29-30; Chaplin v. Boys (1971) AC 356, at pp 378-379, 394. And that is so even in the Australian federal context. Breavington v. Godleman (1988) 169 CLR at pp 135-136, Pedersen v. Young (1964) 110 CLR 162, at pp 166-167, 169, 170; and John Robertson and Co Ltd v. Ferguson Transformers Pty Ltd (1973) 129 CLR 65, at pp 92-93. Were it otherwise, it would be as productive of inconvenience, for a general discussion of inconvenience, see Cook, "'Substance' and 'Procedure' in the Conflict of Laws", (1933) 42 Yale Law Journal 333, at pp 343-344. See also Heavner v. Uniroyal Inc (1973) 305 A 2d 412, at p 415, in that context as in the international context.

11. A distinction has commonly been drawn between limitation provisions which are said to "bar the remedy" and those which are said to "bar the right". See, for example, Harris v. Quine (1869) LR 4 QB 653; Maxwell v. Murphy (1957) 96 CLR 261, at pp 267-268, 279-280, 286-288; Australian Iron and Steel Ltd v. Hoogland (1962) 108 CLR 471, at pp 476, 488-489. Provisions of the former kind have consistently been classified as procedural, Harris v. Quine (1869) LR 4 QB at pp 657, 658; Alliance Bank of Simla v. Carey (1880) 5 CPD 429; Black-Clawson Ltd v. Papierwerke AG (1975) AC 591, at pp 630-631, including in this Court. Pedersen v. Young (1964) 110 CLR at pp 166, 169-170; John Robertson (1973) 129 CLR at pp 80, 83, 92-93; Bargen v. State Government Insurance Office (Q.) (1982) 108 CLR 318; and The Commonwealth v. Verwayen (1990) 170 CLR 394, at pp 473-474. Provisions of the latter kind are clearly substantive in that they operate directly to identify and define rights.

12. Limitation laws "take various forms and have different purposes". Australian Iron and Steel Ltd v. Hoogland (1962) 108 CLR, per Windeyer J. at p 488. However, "(s)ince 21 Jac. I c.16 the usual form ... has been to provide that an action shall not be brought except within the time stated after the cause of action arose." Pedersen v. Young (1964) 110 CLR, per Menzies J. at pp 166-167. A provision in that form neither deprives a court of jurisdiction nor operates automatically to prevent judgment being given on the claim. The Commonwealth v. Verwayen (1990) 170 CLR, at pp 405, 425, 456, 473, 486-487. It affords protection to a defendant to the extent that it is called in aid and made out, ibid., at pp 405, 425-426, 456, 473-474, 486-487, only incidentally does it protect the courts against stale claims.

13. Two South Australian provisions have been identified in the argument made on behalf of the employer, it being asserted that one or the other is applicable in this case. Both provisions take the "usual form": s.82(2) of the Workers Compensation Act (1971) (S.A.) relevantly provides that "a worker ... shall not bring an action ... unless he commences that action within three years from the day on which that injury occurred"; s.36(1) of the Limitation of Actions Act relevantly provides that "(a)ll actions ... shall be commenced within three years ... after the cause of action accrued but not after."

14. Two things should be noted with respect to the South Australian provisions. First, they operate by reference to "an action" or "actions". Secondly, they speak with perfect generality in the sense that there is nothing in their terms to limit their operation to actions brought in South Australia or in the courts of South Australia.

15. The course of judicial decision in cases such as the present assumes that limitation provisions should be classified as "substantive" or "procedural" and in no other way. See, for example, Pedersen v. Young (1964) 110 CLR, at pp 165-166, 166-167, 169-170; Maxwell v. Murphy (1957) 96 CLR, at pp 267-268, 277-278; and Australian Iron and Steel Ltd v. Hoogland (1962) 108 CLR, at pp 480, 488-489. That is a curious assumption, particularly as provisions in the usual form are directed to "actionability" - another concept which often assumes importance in cases in this area. Phillips v. Eyre (1870) LR 6 QB, at pp 28-29.

16. In Breavington Wilson J. and I said in relation to s.118 of the Constitution that "refusal to entertain a matter actionable in the State in which events happened but not actionable in the State in which action is brought, may constitute a refusal of full faith and credit to the laws of the State in which the events occurred" (1988) 169 CLR at p 99. And, in my view, that would necessarily be the case if the court of one State were to apply its shorter limitation period to prevent the determination of a matter governed by and actionable under the law of another State.

17. If the present position were reversed so that New South Wales had a limitation period of three years and South Australia had six years, application of the New South Wales period in this case would constitute a refusal of full faith and credit to the law of South Australia, for it would involve a denial of faith and credit to that law as it applied to the events in question and as it applies to determine their legal consequences. And faith and credit are given in full measure, as required by s.118, only if legal consequences are assigned by application of that body of law which both governs the events in question and determines their legal consequences.

18. Ordinarily, cases in this area are concerned with the situation in which the limitation period in the State where the events occurred is shorter than in the State where the action is brought. And, leaving aside any extension of time, that is the present case. In considering the effect of s.118 in that situation, it is convenient, for the moment, to view the South Australian limitation provisions as though they stood alone. And to do that it is necessary to ascertain their proper meaning.

19. Limitation provisions of the kind with which this case is concerned have usually been construed as applying only to actions commenced in the enacting State or Territory. Pedersen v. Young (1964) 110 CLR, per Kitto J. at p 165, per Taylor J. at p 166; Fielding v. Doran (1984) 59 ALJR 511, at p 514; 60 ALR 342, at p 346; Bowtell v. The Commonwealth (1989) 63 ALJR 465, at p 466; 86 ALR 31, at pp 32-33; Bargen v. State Government Insurance Office (Q.) (1982) 154 CLR at pp 322-323, or in its courts. Pedersen v. Young (1964) 110 CLR, per Menzies J. at pp 167-168, per Owen J. at pp 170-171; John Robertson (1973) 129 CLR, at pp 80-81, 82, 87, 92-93. Thus in Pedersen v. Young Menzies J. (1964) 110 CLR at p 167, said of a provision in the same form as the provisions involved in this case:

"A statute in (that) form, notwithstanding any generality of expression, must of necessity be construed as applying to the commencement of actions in courts subject to the power of the legislature enacting it and to no other courts."

I see no necessity of the kind there stated. Nor do I see the necessity described by Kitto J. in that case, ibid., at p 165, namely, to construe limitation provisions as applying only to actions commenced in the enacting State. Indeed, the statutory context of Pedersen v. Young invites a very different approach.

20. The issue in Pedersen v. Young was whether a Queensland limitation period could be pleaded in an action arising out of events in Queensland but commenced in the New South Wales Registry of this Court. There was not then, and there is not now, Commonwealth legislation providing for the limitation of actions brought in courts exercising federal jurisdiction. Instead, then as now, and indeed since 1903, s.79 of the Judiciary Act 1903 (Cth) relevantly provided:

"The laws of each State ... shall, except as otherwise provided by the Constitution or the laws of the Commonwealth, be binding on all Courts exercising federal jurisdiction in that State ... in all cases to which they are applicable."

21. In the statutory context provided by s.79 of the Judiciary Act, it was never "necessary" to construe a limitation provision in the "usual form" as "applying to the commencement of actions in courts subject to the power of the legislature enacting it and to no other" or as applying only to actions commenced in the enacting State. On the contrary, a construction of that kind - a construction which involves reading words into the provision that are not there - creates its own special problems in matters falling within federal jurisdiction. And, cases such as the present will often fall within federal jurisdiction: they may be matters between residents of different States or between a State and a resident of another State. Constitution, s. 75(iv), or they may be matters relating to the same subject-matter claimed under the laws of different States (1964) 110 CLR, at p 167.

22. In Pedersen v. Young the view expressed by Menzies J.(174) that limitation provisions apply only to actions commenced in the court of the enacting State led Taylor J., ibid., at p 166, to question whether the Queensland provision would have applied even if the action had been commenced in the Queensland Registry of this Court. But Menzies J., ibid., at p 168, thought that ss.79 and 80 of the Judiciary Act might operate "to make applicable in this Court some State statutes which, upon their true construction, apply of their own force only to courts governed by the laws of the State". The view then tentatively expressed by Menzies J. prevailed in John Robertson (1973) 129 CLR, per Menzies J. at pp 80-81, per Gibbs J. at p 88, and per Mason J. at p 95. However, it is difficult to resist the conclusion that, no matter that it achieved a practical result, the solution there adopted is, in the words of Windeyer J. in Pedersen v. Young (1964) 110 CLR, at p 170, not "an answer that is logically satisfying".

23. In Pedersen v. Young the view that limitation provisions in the usual form apply only to actions commenced in the enacting State led Kitto J. to doubt whether the Queensland provision would apply even if the action were heard and determined in Queensland. In his Honour's view, the Judiciary Act could not give the Queensland provision "a new meaning, converting it into a provision limiting the time for the commencement of actions outside Queensland", ibid., at pp 165-166. See also Taylor J. at p 166, Owen J. at p 171; cf. Menzies J. at p 167, and Windeyer J. at p 170. The view there expressed was accepted as correct in Fielding v. Doran (1984) 59 ALJR per Dawson J. at p 514; 60 ALR, at p 346, and in Bowtell v. The Commonwealth (1989) 63 ALJR, per Toohey J. at p 466; 86 ALR, at p 32.

24. If limitation provisions are construed as applying only to actions commenced in the enacting State, those provisions are circumvented by commencing the action elsewhere. And, as a matter of logic, that will be so even if the action is within federal jurisdiction and that jurisdiction is exercised (whether by remitter or otherwise) in the State where the events in question occurred. And the same logic would lead to the conclusion that, in those circumstances, there is no applicable limitation period. However, this is not an area in which logic has always prevailed. And it may be that different considerations apply in the case of a remitted action. In any event, that seems to be the effect of the decision in Commonwealth of Australia v. Dixon (1988) 13 NSWLR 601.

25. In Commonwealth of Australia v. Dixon Hope J.A, ibid., at pp 603-604 described the problems involved in the application of limitation provisions in proceedings instituted in this Court in one State but remitted for hearing to another as "unnecessarily complex". That is plain understatement. And, so far as limitation periods are concerned, that case is but one of several confusing situations that can arise if an action is commenced in a court other than a court of the jurisdiction in which the events in question occurred. That confusion largely results from treating limitation provisions, notwithstanding that they are expressed to operate generally, as confined to actions commenced either in the enacting State or in its courts.

26. So long as it was thought that cases like Pedersen v. Young were to be determined in accordance with the law of the State in which action was brought or, under s.79 of the Judiciary Act, in accordance with the law of the State where federal jurisdiction was exercised, it would have been absurd to apply the limitation period of some different State, even if it were the State where the events in question occurred. That is the only basis on which it is possible to make sense of the decision in that case. That basis was removed by Breavington and, as already indicated, the oral arguments in this case accepted that situation. Without that basis Pedersen v. Young serves no useful purpose and should no longer be followed. In particular, it should no longer be seen as authority for the proposition that limitation provisions which are expressed in general terms are to be construed as confined to actions commenced either in the enacting State or in its courts. Qualifications of that kind deny the plain meaning of the words in which the provisions are expressed. And nothing in the federal arrangements of this country requires that they be construed in that way. On the contrary, qualifications of that kind generate unedifying confusion within the federal system.

27. Stripped of qualifications of the kind asserted in Pedersen v. Young, a limitation provision expressed in the usual form, and hence in general terms, is properly to be seen as applying to actions brought with respect to events governed by the law of the enacting State. That construction recognizes the accepted effect of provisions of that kind which, as earlier noted, directly afford protection for a defendant but only incidentally protect the courts from stale claims. And when construed in this way, a provision of that kind is part of the general body of law which determines the legal consequences attaching to an act or event. For the purposes of s.118, no distinction is to be drawn between a law of that kind and a law which provides as to the legal character of an act or event or, as is more commonly said, which governs that act or event. As already indicated, to give effect to one and not the other is to give something less than full faith and credit. That being so, failure on the part of the Supreme Court of New South Wales to apply the relevant South Australian limitation provision in this case would involve a denial of full faith and credit to the law of South Australia.

28. It should be noted that the South Australian provisions, when construed in the way I have indicated, are not properly characterized as "procedural" laws. On the basis that there is a complete dichotomy between "procedural" and "substantive" laws, they are to be characterized as "substantive". And, quite apart from s.118, the authority of Breavington, which was accepted by the parties and interveners in so far as it overruled Koop v. Bebb, requires that the relevant provision be applied to this action, even though brought in New South Wales.

29. Once the South Australian limitation provisions involved in this case are construed as applying to actions brought to determine the legal consequences of events governed by the law of South Australia, it follows, as a matter of construction, that a court applying one or other of those provisions is in the same position under s.48 of the Limitation of Actions Act with respect to the extension of time as a South Australian court. And s.118 operates to require that court, if so requested, to consider whether the period should be extended in accordance with s.48. That is because failure to consider whether the period should be extended would involve giving something less than full faith and credit to the law of South Australia.

30. I agree with Mason C.J. that it is sufficient if the present matter is determined on the basis that the relevant limitation provision is that contained in s.36 of the Limitation of Actions Act. Although for somewhat different reasons, I have reached the same conclusion as his Honour and I agree with the orders which he proposes.

Orders

Allow the defendant's application to amend its defence to raise section 82(2) of the Workers Compensation Act 1971 (S.A.) and amend the question ordered by Sharpe J. to be separately tried to read as follows: Question: Is the defendant entitled to judgement by virtue of

section 36 of the Limitation of Actions Act 1936 (S.A.) or by virtue of section 82(2) of the Workers Compensation Act 1971 (S.A.)?

Order that the question be answered:

The defendant is not entitled to judgment by virtue of section 36 of the Limitation of Actions Act 1936 (S.A.) or by virtue of section 82(2) of the Workers Compensation Act 1971 (S.A.).

Remit the matter of the separate trial in proceeding No. 10023 of 1990 to the Supreme Court of New South Wales and direct that it be determined in accordance with the answer given by this Court.

Order that the defendant pay the plaintiff's costs of the proceedings in this Court.