Commonwealth v Verwayen

170 CLR 394
95 ALR 321

---

(Judgment by: MASON CJ)

Between: COMMONWEALTH
And: VERWAYEN

Court:

Judges:
Mason C.J.Brennan J.
Deane J.
Dawson J.
Toohey J.
Gaudron J.
McHugh J.

Subject References:
Estoppel

Judgment date: 5 September 1990

---

Judgment by:
MASON CJ

On the night of 10 February 1964 a collision took place between H.M.A.S. Voyager and H.M.A.S. Melbourne upon the high seas in the vicinity of Jervis Bay. The respondent was at the time a member of the Royal Australian Navy serving on board H.M.A.S. Voyager. He is one of a number of people who subsequently brought actions against the Commonwealth for damages for injuries sustained as a result of the collision, alleging that their respective injuries had been caused by the negligence of the officers and crew of one or both of the ships.

It is not necessary for present purposes to describe in detail the circumstances surrounding the collision. It is convenient instead to set out the history of the ensuing litigation between the respondent and the Commonwealth. The most significant aspect of that history is that the respondent did not issue a statement of claim until 2 November 1984. The reason for this delay appears to have been that legal opinion after the time of the collision was influenced by certain remarks made by Windeyer J., by way of obiter, in Parker v. The Commonwealth (1965) 112 CLR 295 at pp 301-302, to the effect that, for reasons of public policy, a member of the armed forces could not recover damages for the negligence of another member of the armed forces in the course of duty. Those remarks were disapproved by the Court in Groves v . The Commonwealth (1982) 150 CLR 113 at pp 118-119, 133-134, 136, 137. That decision broadened the ambit of the law of negligence in the context of the activities of members of the armed forces.

By its defence, dated 14 March 1985, the Commonwealth admitted the allegations made in the statement of claim, except that it did not admit that the respondent had been injured or had suffered loss or damage as a result of the collision. In this way, liability was admitted but the question of damages remained in issue. The Commonwealth did not plead that the action was barred by the Limitation of Actions Act 1958 (Vict.) ("the Act"). Nor did it plead that it owed no duty of care to the respondent by reason of the decision in Groves. It is conceded by the respondent that the time period specified in the Act in relation to his cause of action had expired before the statement of claim was issued. Whether or not the Commonwealth owed a duty of care to the respondent on the facts of the present case is one of the questions with which this appeal is concerned.

The Commonwealth's action in not pleading these two defences was preceded by correspondence between solicitors acting for other persons seeking damages as a result of the collision and representatives of the Commonwealth. The early correspondence concerned a plaintiff named Robert Palmer, whose solicitors had been advised on 21 November 1983 by the Deputy Crown Solicitor that the Commonwealth would not rely on the Act in his case. Indeed, the Commonwealth did not seek to invoke the Act in Mr Palmer's case. Nor, it seems, did the Commonwealth raise the Groves defence in his case.

Mr Palmer's solicitors came to act on behalf of many other survivors of the collision, including the respondent. On 6 September 1984 they wrote to the Secretary of the Department of Defence on behalf of the respondent and four other claimants requesting that he "waive the Statute" in their cases. The Australian Government Solicitor (formerly the Crown Solicitor) advised the solicitors on 25 January 1985 that the Commonwealth proposed to admit liability and to waive the Statute of Limitations defence; this advice was later confirmed in writing.

After the defence had been filed, the Australian Government Solicitor joined with the respondent's solicitors on several occasions in making applications for an expedited hearing of the damages question on the ground that liability was not in issue. Moreover, in a number of statements by Ministers of the Commonwealth, it was made clear that the Commonwealth had adopted a policy not to contest liability and not to plead the Statute of Limitations. Many of these statements were not addressed directly to the respondent or his solicitors, but on 27 November 1985 the Minister Assisting the Minister for Defence wrote to the respondent in relation to a proposed settlement of all claims, which his solicitors had suggested. The letter stated in part:

"As you have pointed out, the Commonwealth has admitted negligence and is not pressing the statutory limitation period as a defence. Nevertheless, it still expects claimants to show that they have suffered injury ... and to prove the extent of their injuries and resultant loss, in order to justify an award of damages."

In an earlier letter to another Member of Parliament, the Minister had stated that the Commonwealth had "waived the absolute defence open to it" under the Statute of Limitations in relation to the claims of the various plaintiffs.

At some time in or around November 1985, the Commonwealth decided to reconsider its policy in relation to the claims arising from the collision. From that point the Commonwealth began to plead defences based upon the Act and upon the argument that no duty was owed to the plaintiffs because of the decisions in Parker and Groves. This action was taken pending a final decision as to the position which the Commonwealth should adopt. Ultimately it was decided that both defences should be raised in each case.

Accordingly, the Commonwealth sought and was granted leave to amend its defence to the respondent's claim so as to raise both defences. The Master ordered that the Commonwealth pay the respondent's taxed costs of the application for leave to amend and the costs, if any, rendered abortive by reason of his order. Costs were otherwise reserved for the trial judge. An amended defence was delivered on 29 May 1986. At that time the trial would have been shortly due to commence. The respondent delivered a reply on 5 June 1986 in which he maintained that no defence was disclosed pursuant to the decisions in Parker and Groves and that the Act did not apply to the Commonwealth. In the alternative, he asserted that the Commonwealth had waived any such defences. The respondent also claimed that the Commonwealth was estopped from relying on either defence.

On 10 September 1986 in the Supreme Court of Victoria the matter came before Vincent J. who ordered that various issues of fact and law arising out of the pleadings be disposed of before the trial of the action. The respondent successfully appealed against this order and the matter came on for trial before O'Bryan J. on 4 November 1987. His Honour required counsel to argue the questions of law raised by the amended defence and the reply before a jury was empanelled, upon the basis that the relevant facts were not in issue or would not be seriously disputed at the trial.

O'Bryan J. held that the public policy defence based on Groves was not available to the Commonwealth on the established facts. However, he also held that, if the Commonwealth had waived the limitation defence, such waiver was revocable and indeed had been revoked. Further, no estoppel operated because the parties had not been in a relevant pre-existing legal relationship and the respondent had not been materially disadvantaged because any legal costs incurred would be recoverable by an appropriate costs order. Judgment was therefore entered for the Commonwealth. His Honour's consideration of the concept of pre-existing legal relationship took place prior to this Court's decision in Waltons Stores (Interstate) Ltd. v. Maher (1988) 164 CLR 387 . Since that case, if a pre-existing legal relationship is needed in order to found a promissory estoppel, it is clearly present in the plaintiff-defendant relationship in this case. However, this is not sufficient to dispose of the appeal.

The respondent appealed to the Full Court of the Supreme Court which allowed the appeal by majority (Kaye and Marks JJ.; King J. dissenting). The Court unanimously rejected the respondent's argument that the Commonwealth had waived its defences, but the majority held that the Commonwealth was estopped from resiling from its promise not to plead the Statute. King J. would have held that the respondent was entitled only to receive an amount representing the out-of-pocket costs and expenses incurred by him as a result of the change of mind by the Commonwealth. Kaye and Marks JJ. also dismissed a cross-appeal by the Commonwealth against O'Bryan J.'s decision that the public policy defence was not available to the Commonwealth. Accordingly, it was ordered that judgment for the Commonwealth be set aside and that the proceedings be remitted for trial.

In this Court, Mr Black Q.C., for the Commonwealth, argued that the majority in the Full Court had erred in granting to the respondent a remedy on the grounds of estoppel which was disproportionate to the detriment resulting from the respondent's reliance upon the Commonwealth's representation that it would not rely on the Act or the Groves defence. He also contended that, on grounds of public policy flowing from the decision in Groves, there was no duty of care owed towards the respondent in the circumstances of the case. The Groves issue does not arise if Mr Black's first argument is accepted. Moreover, the failure of the first argument would prevent the Commonwealth from relying upon the Groves defence even if it would otherwise have been available on the facts of the case. It is convenient therefore to concentrate first upon the effect of the Commonwealth's statements as to the defences it would not plead .

Mr Thomson Q.C., for the respondent, contended that the majority in the Full Court had correctly applied the law relating to estoppel. In the alternative, he argued, contrary to the decision of the Full Court, that the Commonwealth had voluntarily and irrevocably waived the benefit of a statutory right, namely its right to plead the Act as a complete defence. He also contended that O'Bryan J. and the Full Court had been correct in holding that the Commonwealth could not avail itself of the public policy defence based on Groves. Although the respondent did not file a notice of contention giving notice of his intention to rely upon the second of these arguments, the Commonwealth did not object to the course which the argument took. In any event, the dividing line between waiver and estoppel is, to say the least of it, by no means clear cut.

Putting estoppel to one side for the moment, it is desirable to consider, as Mr Thomson invited, the existence of a doctrine of "waiver of the benefit of a statutory right". Undoubtedly, some statutory rights are capable of being extinguished by the person for whose benefit they have been conferred: Sandringham Corporation v. Rayment (1928) 40 CLR 510 at p 527; Wilson v. McIntosh (1894) AC 129 at pp 133-134. However, some statutory rights may also operate as a condition precedent to a court's jurisdiction: Park Gate Iron Co. v. Coates (1870) LR 5 CP 634; Kammins Co. v. Zenith Investments (1971) AC 850. More importantly, some rights may be conferred for reasons of public policy so as to preclude contracting out or abandonment by the individual concerned: see Lieberman v. Morris (1944) 69 CLR 69 . It is therefore necessary to examine the relevant statutory provision in this case in order to ascertain whether it is susceptible to extinguishment in this way.

Section 5(6) of the Act provided:

"No action for damages for negligence ..., where the damages claimed by the plaintiff consist of or include damages in respect of personal injuries to any person, shall be brought after the expiration of three years after the cause of action accrued."

This sub-section was repealed by s 3(c) of the Limitation of Actions (Personal Injury Claims) Act 1983 (Vict.). However, s 11(2) of that Act provided:

"The Acts amended by this Act shall apply as in force immediately before the commencement of this Act to a cause of action arising more than six years before the date of commencement of this Act."

That Act was proclaimed to commence on 11 May 1983.

Although the terms of s 5(6) are such that it is susceptible of being read as going to the existence of the jurisdiction of a court to hear and determine an action of the kind described, limitation provisions similarly expressed have not been held to limit the jurisdiction of courts. Instead, they have been held to bar the remedy but not the right and thus create a defence to the action which must be pleaded: Dawkins v. Lord Penrhyn (1878) 4 App Cas 51 at pp 58-59; The Llandovery Castle (1920) P 119 at p 124; Dismore v. Milton [1938] 3 All ER 762 ; Ronex Properties v. John Laing (1983) QB 398; Ketteman v. Hansel Properties (1987) AC 189 at p 219.

On the footing that the right to plead the statute as a defence is a right conferred by statute, the respondent's contention that the right is capable of waiver hinges on the scope and policy of the particular statute: Admiralty Commissioners v. Valverda (Owners) (1938) AC 173 at p 185. The issue is not whether the relevant provisions are beneficial to the public, but whether they are "dictated by public policy" and enacted "not for the benefit of any individuals or body of individuals, but for considerations of State": at p 185. Although, in one sense, all statutes give effect to some public policy (see Lieberman v. Morris at pp 82, 84), the critical question is whether the benefit is personal or private or whether it rests upon public policy or expediency: Brown v. The Queen (1986) 160 CLR 171 at p 208.

In this case there is the public policy that there should be finality in civil litigation. However, the Parliament has seen fit to implement this policy, not by imposing a jurisdictional restriction, but by conferring on defendants a right to plead as a defence the expiry of the relevant time period. In these circumstances and having regard to the nature of the statutory defence, I conclude that the purpose of the statute is to confer a benefit upon persons as individuals rather than to meet some public need which must be satisfied to the exclusion of the right of access of individuals to the courts. On that basis, it is possible to "contract out" of the statutory provisions, and it is equally possible to deprive them of effect by other means such as waiver. Put differently, the provisions are procedural rather than substantive in nature, which suggests that they are capable of waiver: Admiralty Commissioners v. Valverda, at p 185.

But, granted that some statutory rights can be waived, the mere existence of cases in which statutory rights have been held to be susceptible to waiver does not signify that those cases are all exemplifications of one concept or doctrine. As often as not, the term "waiver" is used to describe the result of the application of various principles rather than to designate a particular legal concept or doctrine. The consequence is that the expression "waiver" has been the subject of robust criticism, notably by Dr Ewart in his work Waiver Distributed, (1917); see also Bysouth v. Shire of Blackburn and Mitcham (No. 2) (1928) VLR 562 at p 579; Larratt v. Bankers and Traders' Insurance Co. (1941) 41 SR (NSW) 215 at p 226; Kammins, per Lord Diplock at pp 882-883. This is because "waiver" is an imprecise term capable of describing different legal concepts, notably election and estoppel.

It has been doubted that waiver exists as a defence or answer in any case except where it is used as an alternative designation for some other defence or answer, for example, election, estoppel or new agreement: Bysouth, per Lowe J. at p 579. Generally speaking, as Jordan C.J. pointed out in Larratt (at pp 226-227), an existing legal right is not destroyed by mere waiver in the sense of an express or implied intimation that the person in whom the right is vested does not intend to enforce it: see Mulcahy v. Hoyne (1925) 36 CLR 41 per Isaacs J. at pp 55-56; Atlantic Shipping and Trading Co. v. Louis Dreyfus and Co. [1922] 2 AC 250 per Lord Sumner at pp 261-262. In these cases, unless consideration is present, something in the nature of an election or an estoppel is required.

According to its strict legal connotation, waiver is an intentional act done with knowledge whereby a person abandons a right by acting in a manner inconsistent with that right: Craine v. Colonial Mutual Fire Insurance Co. Ltd. (1920) 28 CLR 305 at p 326; Grundt v. Great Boulder Pty. Gold Mines Ltd. (1937) 59 CLR 641 at p 658. However, the better view is that, apart from estoppel and new agreement, abandonment of a right occurs only where the person waiving the right is entitled to alternative rights inconsistent with one another, such as the right to insist on performance of a contract and the right to rescind for essential breach: see Kammins, at p 883. This category of waiver is an example of the doctrine of election.

Another category of waiver is one in which a person is prevented from asserting, in response to a claim against him, a particular defence or objection which would otherwise have been available. Here waiver is said to arise when the person agrees not to raise the particular defence or so conducts himself as to be estopped from raising it: see Kammins, at p 883.

In these circumstances, the authorities dealing with waiver of statutory rights do not call for special consideration. They speak with different voices, sometimes in the language of election, at times in that of estoppel and at other times in terms of unconscionability: see, for example, Ward v. Raw (1872) LR 15 Eq 83; Phillips v. Martin (1890) 11 NSWLR 153 at pp 157-158, 159; Wilson v. McIntosh. Quasi-estoppel by acquiescence is another approach which has found favour: Willmott v. Barber (1880) 15 Ch D 96, at pp 105-106; Kammins at pp 884-885. The old references to unconscionability may be taken today as forerunners of the modern principles of estoppel, now that prevention of unconscionable conduct has been identified as the driving force behind equitable estoppel: Waltons Stores, at pp 404, 419, 450.

It is necessary to consider whether, first, the doctrine of election and, secondly, the principles of estoppel (including quasi-estoppel by acquiescence) apply in the present case. The broad principles of election are not in doubt. They were formulated by this Court, under the title of waiver, in Craine v. Colonial Mutual, at p 326; see also O'Connor v. S.P Bray Ltd. (1936) 36 SR (NSW) 248 at pp 257-264. In Sargent v. A.S.L. Developments Ltd. (1974) 131 CLR 634 Stephen J. explained (at p 641):

"The doctrine only applies if the rights are inconsistent the one with the other and it is this concurrent existence of inconsistent sets of rights which explains the doctrine; because they are inconsistent neither one may be enjoyed without the extinction of the other and that extinction confers upon the elector the benefit of enjoying the other, a benefit denied to him so long as both remained in existence."

See also per Mason J. at p 656.

The respondent contends that the Commonwealth was required to elect between two inconsistent rights, namely, the right to plead and the right not to plead. The defences could be either pleaded or not pleaded; and a choice needed to be made as to which course to follow. But that is by no means the end of the matter because it is not clear that one of these "rights" could not be enjoyed without the extinction of the other. Indeed, the Commonwealth argues that, subject to the grant of leave to amend the pleadings, there was nothing to prevent it from adopting the right which the respondent claims was extinguished. The essential preliminary question is therefore whether or not the Commonwealth was required to make an irrevocable choice between two alternative positions; if it was not, then the two cannot be said to have been relevantly inconsistent, and the doctrine of election would not come into play.

The respondent's contention is that, by filing a defence omitting reference to the Act and the Groves defence, the Commonwealth had irrevocably elected not to plead the defences available to it. The immediate difficulty with that contention is that it does not account for the possibility that leave may be granted to amend the pleadings for the purpose of including the defences. The respondent could not point to any authority in which the filing of a defence was said to amount to an election in this way. Indeed, there is support for the contrary view, based upon the following passage from the judgment of Viscount Simon LC. in United Australia Ltd. v. Barclays Bank Ltd. (1941) AC 1 at pp 18-19:

"No doubt, if the plaintiff proved the necessary facts, he could be required to elect on which of his alternative causes of action he would take judgment, but that has nothing to do with the unfounded contention that election arises when the writ is issued. There is nothing conclusive about the form in which the writ is issued, or about the claims made in the statement of claim. A plaintiff may at any time before judgment be permitted to amend. ... At some stage of the proceedings, the plaintiff must elect which remedy he will have. There is, however, no reason of principle or convenience why that stage should be deemed to be reached until the plaintiff applies for judgment."

See also per Lord Atkin at pp 29-31. That case involved the quite different question whether the initiation of proceedings against one defendant precluded the bringing of an action against a second defendant on the basis of the same facts. Of course, the decision provides no support for the view that a defendant will not need to elect whether or not to plead a defence until the plaintiff seeks judgment. Nonetheless, the general statements in relation to the plaintiff's opportunities to amend the pleadings have at least equal force when applied to a defendant.

This is not a case in which it could be said that the defendant was required by a certain point in time to elect whether or not to plead the defences. If there was no need to make an election when the defence was first filed, there is no reason why the comparatively insignificant proceedings which followed gave rise to such a need and precluded the reversal of the previous decision. If the facts give rise to a conclusion that the Commonwealth's decision was irrevocable, then the reason is not to be found in the principles of election.

That brings me to estoppel, a label which covers a complex array of rules spanning various categories. There are the divisions between common law and equitable estoppel, between estoppel by conduct and estoppel by representation, and the distinction between present and future fact. There are titles such as promissory estoppel, proprietary estoppel and estoppel by acquiescence. Yet all of these categories and distinctions are intended to serve the same fundamental purpose, namely "protection against the detriment which would flow from a party's change of position if the assumption (or expectation) that led to it were deserted": Waltons Stores, per Brennan J. at p 419. See also per Mason C.J. and Wilson J. at p 404; Grundt, at pp 674-675.

At common law the principle of estoppel by conduct or representation ("estoppel in pais") provided that protection by preventing the party estopped from unjustly departing from an assumption of fact which his conduct had caused another party to adopt or accept for the purpose of their legal relations: Grundt, at pp 657, 674; Thompson v. Palmer (1933) 49 CLR 507 at p 547; Waltons Stores, at pp 397-399, 413-415, 443, 458. But it was well established that, in order to support an estoppel by conduct, the representation (or assumption) must be a representation of an existing fact , a promise or representation of intention to do something being insufficient for that purpose: Yorkshire Insurance Co. v. Craine [1922] 2 AC 541 at p 553.

The principle of estoppel by conduct or representation applied in equity, as at common law, though in equity the principle was known as equitable estoppel: Jorden v. Money (1854) 5 HLC 185 at pp 210, 212-213 (10 ER 868 at pp 880, 881); Thompson v. Palmer, at pp 519-520, 547, 558; Waltons Stores, at pp 447-448. And in equity it was also well settled that the representation (or assumption) must be of an existing fact, not of future fact or mere intention. That is what Jorden v. Money decided, despite the fact, as Bowen LJ. pointed out in Edgington v. Fitzmaurice (1885) 29 Ch D 459 at p 483, that "the state of a man's mind is as much a fact as the state of his digestion". This limitation upon the principle of estoppel was seemingly founded upon the notion that to hold a person to an assumption which his conduct has caused another to adopt or accept was tantamount to enforcing a voluntary promise in the absence of consideration. The need to avoid this consequence was an important aspect of the majority reasoning in Jorden v. Money.

However, neither the decision nor the reasoning in that case can now be sustained. Promissory estoppel, recognized by this Court in Legione v. Hateley (1983) 152 CLR 406 , has undermined the idea that voluntary promises cannot be enforced in the absence of consideration. What is more, promissory estoppel has an extensive area of operation now that it is acknowledged that the doctrine is not confined to pre-existing contractual relationships: see Waltons Stores, at pp 399-406. Furthermore, the acceptance of the doctrine of promissory estoppel has been accompanied by a recognition that the distinction between present and future fact is unsatisfactory and produces arbitrary results instead of serving any useful purpose: Moorgate Ltd. v. Twitchings (1976) QB 225 at p 242; Waltons Stores, at pp 398-399, 450-451, 452; Foran v. Wight (1989) 64 ALJR 1 at pp 12, 22-23; 88 ALR 413 at pp 430-431, 448-449. Indeed, the difference between the majority and Lord St Leonards in Jorden v. Money was a striking illustration of the arbitrary nature of the distinction.

In conformity with the fundamental purpose of all estoppels to afford protection against the detriment which would flow from a party's change of position if the assumption that led to it were deserted, these developments have brought a greater underlying unity to the various categories of estoppel. Indeed, the consistent trend in the modern decisions points inexorably towards the emergence of one overarching doctrine of estoppel rather than a series of independent rules: see Waltons Stores, at pp 403-404, 447-451; Foran v. Wight, at pp 12, 22-23; pp 430-431, 448-449 of ALR; Collin v. Holden (1989) VR 510 at pp 515-516; Taylors Fashions Ltd. v. Liverpool Trustees Co. (1982) QB 133 at p 153; Amalgamated Property Co. v. Texas Bank (1982) QB 84 at p 122; Attorney-General of Hong Kong v. Humphreys Estate (1987) AC 114 at pp 123-124.

One obstacle to the existence of a single overarching doctrine is a suggested difference in the nature of estoppel by conduct on the one hand and equitable estoppel (including promissory estoppel) on the other and in the character of the protection which they respectively provide. Traditionally, estoppel by conduct has been classified as a rule of evidence, available where there is a cause of action, to prevent a person from denying what he previously represented, and has not itself constituted a cause of action: Grundt, at p 658; Low v. Bouverie (1891) 3 Ch 82 at pp 101, 105. Being an evidentiary principle, estoppel by conduct achieved, and could only achieve, the object of avoiding the detriment which would be suffered by another in the event of departure from the assumed state of affairs by holding the party estopped to that state of affairs. The rights of the parties were ascertained and declared by reference to that state of affairs. On the other hand, equity was more flexible. Equity was concerned, not to make good the assumption, but to do what was necessary to prevent the suffering of detriment. To do more would sit uncomfortably with a general principle whose underlying foundation was the concept of unconscionability. So, in Waltons Stores, a majority of this Court concluded that equitable estoppel entitled a party only to that relief which was necessary to prevent unconscionable conduct and to do justice between the parties. Mason C.J. and Wilson J. referred (at p 404) to the statement of Scarman L.J. in Crabb v. Arun District Council (1976) Ch 179 at p 198, that the court should determine what was "the minimum equity to do justice to the plaintiff". We went on to state (at p 405):

"Holding the representor to his representation is merely one way of doing justice between the parties."

Similarly, Brennan J. said (at p 419):

"The element which both attracts the jurisdiction of a court of equity and shapes the remedy to be given is unconscionable conduct on the part of the person bound by the equity, and the remedy required to satisfy an equity varies according to the circumstances of the case. As Robert Goff J. said in Amalgamated Property Co. v. Texas Bank (1982) QB 84 at p 103: 'Of all doctrines, equitable estoppel is surely one of the most flexible.' ... However, in moulding its decree, the court, as a court of conscience, goes no further than is necessary to prevent unconscionable conduct."

It follows that, as a matter of principle and authority, equitable estoppel will permit a court to do what is required in order to avoid detriment to the party who has relied on the assumption induced by the party estopped, but no more. In appropriate cases, that will require that the party estopped be held to the assumption created, even if that means the effective enforcement of a voluntary promise. To that extent there is an overlap between equitable estoppel generally and estoppel by conduct in its traditional form. But since the function of equitable estoppel has expanded and it has become recognized that an assumption as to future fact may ground an estoppel by conduct at common law as well as in equity, it is anomalous and potentially unjust to allow the two doctrines to inhabit the same territory yet produce different results. Moreover, as I have already indicated, the fact that estoppel by conduct has expanded beyond its evidentiary function into a substantive doctrine means that there is no longer any justification for insisting on the making good of assumptions in every case.

In any event, there is a very strong case for saying that equity had discarded earlier the notion that the purpose of the rules of estoppel by conduct was to make good the relevant assumption. As Professor Finn points out in his essay "Equitable Estoppel" in Finn (ed.), Essays in Equity, (1985), at p 68, "the language of expectations (was) forsaken entirely for that of 'equities'" in Crabb v. Arun District Council. Lord Denning M.R. had qualified the language of expectations in E.R. Ives Investment Ltd. v. High [1967] 2 QB 379 at pp 394-395, by stating that the "court will not allow (the) expectation to be defeated when it would be inequitable so to do" (emphasis added). That qualification, made repeatedly in cases which can be traced back to the Privy Council's statement in Plimmer v. Mayor, and, of Wellington (1884) 9 App Cas 699 at p 714, that "the Court must look at the circumstances in each case to decide in what way the equity can be satisfied", has transformed the basis of the equitable principles of estoppel: see Finn, at pp 62-71 .

In these circumstances, it would confound principle and common sense to maintain that estoppel by conduct occupies a special field which has as its hallmark function the making good of assumptions. There is no longer any purpose to be served in recognizing an evidentiary form of estoppel operating in the same circumstances as the emergent rules of substantive estoppel. The result is that it should be accepted that there is but one doctrine of estoppel, which provides that a court of common law or equity may do what is required, but not more, to prevent a person who has relied upon an assumption as to a present, past or future state of affairs (including a legal state of affairs), which assumption the party estopped has induced him to hold, from suffering detriment in reliance upon the assumption as a result of the denial of its correctness. A central element of that doctrine is that there must be a proportionality between the remedy and the detriment which is its purpose to avoid. It would be wholly inequitable and unjust to insist upon a disproportionate making good of the relevant assumption. (See also the conclusion of Lord Denning M.R. in Amalgamated Property Co. v. Texas Bank, at p 122.)

The assumption may be one as to a legal as well as to a factual state of affairs. There is simply no reason to restrict the assumption to a factual matter as there was at the time when the rules of estoppel by conduct were evidentiary. It has already been recognized that an equitable estoppel may relate at least to a matter of mixed fact and law (see Waltons Stores, at pp 415-416, 420-421, 452; Foran v. Wight, at p 22; p 448 of ALR). Moreover, the distinction between assumptions as to fact and assumptions as to law is artificial and elusive; see the discussion of Oliver J. in Taylors Fashions Ltd. v. Liverpool Trustees Co., at pp 150-151. So it would be productive only of confusion and arid technicality to restrict the operation of the doctrine so as to exclude from its scope an assumption as to a purely legal state of affairs. It is therefore not surprising that long ago the Judicial Committee recognized that a representation as to the legal effect of an agreement can give rise to an estoppel: Sarat Chunder Dey v. Gopal Chunder Laha (1892) LR 19 Ind App 203; Calgary Milling Co. Ltd. v. American Surety Co. of New York (1919) 3 WWR 98; see also Amalgamated Property Co. v. Texas Bank, at pp 106-107.

Turning to the facts of the present case, at least in so far as the statutory defence is concerned, the difficulty facing the respondent at the outset is to establish that the required assumption was induced by the Commonwealth. What must be established is either that the Commonwealth represented that it had decided not to plead the Statute or the Groves defence and that it did not regard itself as free to change its decision (cf. Waltons Stores, at pp 422-423) or that the Commonwealth represented that it would not plead those defences.

In an ordinary case, the nature of pleadings and their susceptibility, whether by leave or otherwise, to amendment would make it most unlikely that it could be inferred from the pleadings alone that the pleader had induced another party to make an assumption that a particular matter would or would not be pleaded. The other party might reasonably be expected to appreciate that no inference can be drawn from the state of the pleadings alone at a particular time as to the future course which the pleader may decide to take. Still less would it be reasonable to assume that an implied promise not to amend the pleadings, if such a promise could be identified, would be enforceable in the absence of consideration: see Waltons Stores, at p 403.

However, in the present case the respondent is able to point to more than the mere filing and serving of the defence by the Commonwealth. There were clear indications that a deliberate and considered decision had been made whereby the limitation defence and the defence of no duty of care would not be pleaded in any of the ensuing actions brought by survivors of the collision. Those indications apparently included express representations to some claimants followed by the assessment and award of damages on the footing that no defence was pleaded. In the respondent's case, the Commonwealth had joined in making applications for an expedited hearing of the damages issue.

In all the circumstances the proper conclusion to be drawn is that the respondent had been induced by the Commonwealth's conduct to assume that the Commonwealth had made a decision not to plead the limitation defence or the Groves defence and that that decision would not be changed. The fact that the circumstances pointed to the existence of a definitive government policy which had been followed to the point of judgment on other occasions supports the conclusion that that assumption was a reasonable assumption for a person in the respondent's position to make. The relevance of this conclusion is that there is no reason to doubt the respondent's assertion that he made the assumption and continued his action against the Commonwealth in reliance on it.

The element of detriment presents more difficulty. Of course the respondent would suffer detriment in reliance on the assumption if the Commonwealth were to depart from it, at least in the sense that he would fail in his action for damages. However, the question of detriment is not as simple as such an answer would suggest, and is closely related to the other elements of the claim of estoppel.

When a person relies upon the correctness of an assumption which is subsequently denied by the party who has induced the making of the assumption, two distinct types of detriment may be caused. In a broad sense, there is the detriment which would result from the denial of the correctness of the assumption upon which the person has relied. In a narrower sense, there is the detriment which the person has suffered as a result of his reliance upon the correctness of the assumption.

The cases concerning estoppel by conduct, at least at common law, were in one respect concerned with the broader concept of detriment. So, Dixon J. said in Grundt, at p 674, that "the real detriment or harm from which the law seeks to give protection is that which would flow from the change of position if the assumption were deserted that led to it". This makes it clear that the detriment must flow from the reliance upon the assumption (see Foran v. Wight, at p 12; p 431 of ALR), but goes further and suggests that the relief granted by virtue of the estoppel (in that context, the making good of the assumption) corresponds with the detriment which would be suffered were the assumption to be deserted. However, that further suggestion was dictated by the then existing confines of the rules of estoppel by conduct; Dixon J. was plainly stating that a person's "change of position" could not be allowed to "operate as a detriment": at p 674. His Honour's exposition is now instructive as an indication that the detriment against which the law protects is that which flows from reliance upon the deserted assumption, even though at that time the evidentiary rule operated to hold the representor to the assumption created.

In the same way, cases of equitable estoppel have been concerned to grant relief where detriment would be suffered if the assumed state of affairs upon which reliance had been placed was held not to exist. But, as we have seen, the relief which equity grants is by no means necessarily to be measured by the extent of that detriment. So, while detriment in the broader sense is required in order to found an estoppel (and it would be strange to grant relief if such detriment were absent), the law provides a remedy which will often be closer in scope to the detriment suffered in the narrower sense.

It remains only to determine what relief is appropriate to satisfy the estoppel which the respondent has successfully raised in this case. When a court approaches the task of ascertaining the minimum relief necessary to "do justice" between the parties, it is not correct to make an assessment of the moral rectitude of the actions of the parties in a manner divorced from a consideration of the legal consequences and attributes of those actions. Thus it must be borne in mind that a voluntary promise is generally not enforceable and that pleadings are susceptible of amendment. The breaking of a promise, without more, is morally reprehensible, but not unconscionable in the sense that equity will necessarily prevent its occurrence or remedy the consequent loss. In the same way, with estoppel, something more than a broken promise is required.

Each case is one of degree. Reliance upon an assumption for an extended period may give rise to an estoppel justifying a court in requiring that the assumption be made good. The same result may follow from substantial and irreversible detriment suffered in reliance upon the assumption or from detriment which cannot satisfactorily be compensated or remedied. In the present case the detriment suffered by the respondent in reliance on the assumption induced by the Commonwealth appears to be of a more limited nature. The procedure adopted for the determination of this case in the Supreme Court means that we have no finding or evidence of the detriment, flowing from his reliance upon the assumption, which the respondent would suffer from the Commonwealth's pleading of the limitation defence or the Groves defence. It must be assumed, however, that that detriment would include significant expense and inconvenience. However, as far as the respondent's emotional condition is concerned, it is sheer speculation to suggest that his reliance on the Commonwealth's actions after commencement of the action caused any deterioration of that condition. Evidence of detriment must be affirmatively demonstrated; this is not a case involving the exercise of judicial discretion: cf. Murray v. Munro (1906) 3 CLR 788 at p 796; Ketteman v. Hansel Properties, at p 220.

The question then is whether an order for costs is a sufficient recompense for the respondent in respect of the detriment suffered by him. An order for costs has traditionally been regarded as a sufficient adjustment to meet prejudice in terms of expense and inconvenience occasioned by the pleading of new defences and I am not persuaded that principle or circumstance call for any different answer in the present case. There is no material before the Court to justify a conclusion that the respondent commenced his action on the basis of any express or implied representation on the part of the Commonwealth. The respondent's solicitors inquired of the Secretary of the Department of Defence in September 1984 whether he would agree to waive the Statute of Limitations and admit liability. The relevant Minister replied on 29 October 1984, stating that the matter had been referred for consideration to the Australian Government Solicitor. The statement of claim was issued on 2 November 1984. It can hardly be said that these circumstances establish the existence of an assumption on the part of the respondent that liability was not in issue. Indeed, the Minister's letter alerted the respondent's solicitors to the fact that, although the defences had not been pleaded when the question had previously arisen, the Commonwealth still saw fit to refer the question for legal advice on this occasion. If anything, this suggests the absence of a definite and unambiguous policy at the time when the respondent commenced his action.

To hold the Commonwealth to its representations, thereby depriving it of defences which were available to it by statute or the general law, would be a disproportionate response to the detriment suffered by the respondent in reliance upon the assumption that the defences would not be pleaded. True it is that the representations reflected a deliberate policy decision made by government at ministerial level at least. That circumstance gave the representations the quality of apparent reliability and went to the issue of reliance. But the apparent reliability of the representations does not enlarge the nature or scope of the detriment which the respondent has suffered in reliance on the representations following the denial of the assumption generated by them. Likewise, the fact that the Commonwealth is the party against whom an estoppel is pleaded is not in this case a point of distinction. It was not argued that any special rule of estoppel applies to assumptions induced by government, either so as to expand or so as to contract the field of operation of the doctrine.

In the result I conclude that the respondent's case of waiver and estoppel has not been made out and that the Full Court of the Supreme Court was wrong in holding otherwise. The appeal should be allowed.