Commonwealth v Verwayen

170 CLR 394
95 ALR 321

(Judgment by: TOOHEY J)

Between: COMMONWEALTH
And: VERWAYEN

Court:
High Court of Australia

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:
TOOHEY J

The circumstances giving rise to this appeal appear in the judgments of other members of the Court.

A convenient starting point is the application by the Commonwealth on 22 May 1986 to amend its defence. At that time Mr Verwayen's action had been on foot since 2 November 1984; it was commenced well out of time. The Commonwealth's advisers had told Mr Verwayen's advisers on 25 January 1985 that the Commonwealth proposed to admit liability for his claim and to waive the Statute of Limitations. The relevant legislative provision was s 5(6) of the Limitation of Actions Act 1958 (Vic.) ("the Limitation Act"). Consistent with that advice, on 14 March 1985 the Commonwealth had filed its defence to the action, admitting all the allegations in the statement of claim, save those relating to injuries, loss and damage. It contained no plea that the action was statute barred. It was therefore the not uncommon situation of a defendant admitting liability for the plaintiff's claim but putting in issue the damages to which the plaintiff is entitled.

Thereafter, during 1985, the Australian Government Solicitor had joined on five occasions in applications made on behalf of Mr Verwayen for an expedited hearing of the only issue arising on the pleadings, namely, the assessment of damages, and had signed a Certificate of Readiness for assessment of damages only. As late as 27 November 1985, the Minister Assisting the Minister for Defence had written to Mr Verwayen, repeating that "the Commonwealth has admitted negligence and is not pressing the statutory limitation period as a defence". At the same time the Minister expressed regret at "any delay in the finalization of claims". Six months later and only a short time before the assessment of damages was due to take place, the Commonwealth applied for leave to amend its defence.

Against opposition from Mr Verwayen's counsel, leave was granted to amend the defence so as to deny negligence and to plead, not only s 5 of the Limitation Act, but also that by reason of the fact that the Voyager and the Melbourne were, at the relevant time, engaged in combat exercises, no duty of care was owed to Mr Verwayen. I propose to deal first with the limitations issue. Now, it is true that different considerations apply in deciding whether a party should have leave to amend a pleading and in determining the issue raised by the amended pleading, if leave be given. Leave to amend is ordinarily given, even at a late stage, so long as this can be done "without injustice to the other party", to use the words of Bowen L.J. in Cropper v. Smith (1884) 26 ChD 700, at p 710. See also Shannon v. Lee Chun (1912) 15 CLR 257 , at p 261; Clough v. Frog (1974) 48 ALJR 481; 4 ALR 615 ; Atkinson v. Fitzwalter (1987) 1 WLR 201 , at pp 204-205; [1987] 1 All ER 483 , at pp 485-486. Nevertheless, it is appropriate to refuse leave to amend, if to grant leave may bring about an injustice to the other party which cannot be compensated by an adjournment or by an order for costs. This is not something that can always be determined by reference to the merits of the matter sought to be raised by the proposed amendment: Strauss v. Douglas Aircraft Co. (1968) 404 F 2d 1152, at p 1155. There may be cases, and a plea of limitations is a good example, where the amendment proposed has every prospect of success but comes at such a late stage that, if it be granted and the point raised succeeds, an award of costs in favour of the other party is no adequate compensation. It may be, for instance, that it is then too late for the plaintiff to bring action against anyone else: see Joint Coal Board v. Adelaide S.S. Co. Ltd. (1964) NSWR 1126. It should not be lightly assumed that the "healing medicine" (the words are those of Bowen L.J. in Cropper v. Smith, at p 711) of costs is always a sufficient cure for the disadvantage to the other side. Lord Griffiths made the point in Ketteman v. Hansel Properties (1987) AC 189, when he said at p 220:

"Whether an amendment should be granted is a matter for the discretion of the trial judge and he should be guided in the exercise of the discretion by his assessment of where justice lies. Many and diverse factors will bear upon the exercise of this discretion. I do not think it possible to enumerate them all or wise to attempt to do so. But justice cannot always be measured in terms of money and in my view a judge is entitled to weigh in the balance the strain the litigation imposes on litigants, particularly if they are personal litigants rather than business corporations, the anxieties occasioned by facing new issues, the raising of false hopes, and the legitimate expectation that the trial will determine the issues one way or the other."

Therefore, while it is true that to grant the amendment sought by the Commonwealth would not preclude Mr Verwayen, by some such plea as waiver or estoppel, from meeting the matters raised by the amendment, the delay in seeking to plead limitations, particularly in the context of the many assurances given to Mr Verwayen by the Commonwealth that no such plea would be raised against his claim and that liability would be admitted, provided compelling reasons for refusing leave to amend. However, there was no appeal from the order giving the Commonwealth leave to amend its defence and the matter must be approached on the basis that the amended defence has been properly entered. To that defence Mr Verwayen pleaded by way of reply that the Commonwealth "agreed to and did waive any such defence". That plea went to the defence that no duty of care was, in the circumstances, owed to Mr Verwayen and also to the defence based on Limitation Act. In addition, the reply asserted that the Commonwealth was "estopped" from relying upon either defence. However, the reply did not advert to the general denial of negligence, a matter that is referred to again at the end of these reasons.

Putting to one side for a moment the terminology of waiver or estoppel, it is necessary to ask - what is Mr Verwayen's complaint by reason of the introduction of the new matters into the Commonwealth's defence? One thing is clear, he cannot complain that, by reason of anything done or not done by the Commonwealth, he was induced to delay bringing his action until after the relevant limitation period had expired. As already mentioned, the limitation period had expired well before there was any communication between Mr Verwayen or his advisers on the one hand and the Commonwealth or its advisers on the other. Nor can he complain that he was induced to bring proceedings when he did because of any assurance given by the Commonwealth. It is true that two months earlier his solicitors had written to the Department of Defence on his behalf and on behalf of four other survivors from the Voyager, noting that in the case of another survivor the Commonwealth had "waived the Statute of Limitations" and had "admitted liability". But, by 2 November 1984, no reply had been received to that letter, other than an acknowledgement of receipt and a statement that the request to "treat these men the same" had been referred to the Australian Government Solicitor in Victoria, who would advise the solicitors of the Commonwealth's position.

On the other hand, it is equally clear that the conduct of the Commonwealth, beginning with its advice of 25 January 1985 and continuing until the letter of 27 November 1985, written by the Minister Assisting the Minister for Defence, and thereafter, induced Mr Verwayen to continue with his action in the belief that liability for his claim was admitted and ultimately to prepare for an assessment of damages, expected to be held during the Wangaratta sittings of the Supreme Court of Victoria to begin on 3 June 1986. When Mr Verwayen brought his action, it was open to the Commonwealth to include in its defence those matters which it later included, pursuant to the order of 29 May 1986, made on the application of 22 May 1986. It did not do so. Neverthless it might, at any time before May 1986, have sought leave to amend its defence to raise those matters. It did not do so. Furthermore, it engaged in correspondence with Mr Verwayen's solicitors that left in no doubt that it did not propose to rely upon any defence of limitations or indeed upon any defence at all, other than to put Mr Verwayen to proof of damage.

It is against that background that it is necessary to characterize the conduct of the Commonwealth. And this is no mere matter of semantics, for the consequences for the parties may depend very much upon the character to be attributed to that conduct.

The Commonwealth conceded that there had been a promise on its part not to rely upon the Limitation Act. But, it was argued, that could not give rise to a promissory estoppel because the object of promissory estoppel is to avoid detriment, not to secure performance of the promise made. The only possible detriment acknowledged by the Commonwealth to have been suffered by Mr Verwayen was that involved in preparing the action for trial; such detriment, it was said, could be met by an appropriate order for costs. I return to this aspect later in these reasons.

Any discussion of waiver or estoppel often begins with Craine v. Colonial Mutual Fire Insurance Co. Ltd. (1920) 28 CLR 305 , in which Isaacs J. delivered the judgment of the Court. At p 326, Isaacs J. identified waiver as "an intentional act with knowledge", borrowing the language of Lord Chelmsford L.C. in Earl of Darnley v. Proprietors, andc. of London, Chatham, and Dover Railway (1867) LR 2 HL 43 at p 57. Waiver, said his Honour at p 326:

"looks, however, chiefly to the conduct and position of the person who is said to have waived, in order to see whether he has 'approbated' so as to prevent him from 'reprobating' - in English terms, whether he has elected to get some advantage to which he would not otherwise have been entitled, so as to deny to him a later election to the contrary".

His Honour then contrasted the notion of estoppel by conduct, saying, at p 327:

"First of all, the law of estoppel looks chiefly at the situation of the person relying on the estoppel; next, as a consequence of the first, the knowledge of the person sought to be estopped is immaterial; thirdly, as a further consequence, it is not essential that the person sought to be estopped should have acted with any intention to deceive; fourthly, conduct, short of positive acts, is sufficient."

However, the terms are not used consistently and Lord Wright commented in Smyth and Co. v. Bailey and Co. [1940] 3 All ER 60 , at p 70: "The word 'waiver' is a vague term used in many senses." See also The "Kanchenjunga" (1990) 1 Lloyd's Rep 391, per Lord Goff of Chieveley at pp 397-398. 'Estoppel' is likewise susceptible of several meanings. Indeed, in his foreword to Ewart, Waiver Distributed, (1917), Roscoe Pound observed, at p vi:

"Having previously looked into the case of that much-enduring word 'estoppel' Mr. Ewart now takes up another slippery word worn smooth with overuse and shows us 'waiver' as a pseudo-conception."

Nevertheless, usage has sanctioned waiver as apt to signify "the legal grounds on which a person is precluded from asserting one legal right when he is entitled to alternative rights inconsistent with each other" and "the legal grounds on which a person is precluded from raising a particular defence to a claim against him": Mason J. in Sargent v. A.S.L. Developments Ltd. (1974) 131 CLR 634 , at p 655. While it has been said that the loss of a right in the circumstances postulated is "better categorised as 'election' rather than as 'waiver'" (Lord Diplock in Kammins Co. v. Zenith Investments (1971) AC 850, at p 883), waiver is an appropriate term to describe the loss of a defence otherwise available to a defendant. In that sense it has been said to arise when a defendant "either agrees with the claimant not to raise that particular defence or so conducts himself as to be estopped from raising it" (Lord Diplock, at p 883) though, again, the terminology is blurred.

Waiver, as a notion applicable both to the loss of a right and the loss of a defence, has a respectable lineage. The maxim Quilibet potest renunciare juri pro se introducto (a person may renounce a right introduced for his benefit), referred to in Wilson v. McIntosh (1894) AC 129, at p 133, is mentioned by Coke in Beawfage's Case 10 Co Rep 99b, at p 101a, as "the old rule" (77 ER 1076, at p 1080). In Bovill v. Wood (1813) 2 M and S 23, at p 25 (105 ER 291, at p 292) Bayley J. said, in relation to an argument that all contracting parties to an attorney's bill should have been joined in an action for assumpsit , notwithstanding that one had been made bankrupt:

"There is no doubt but that the action ought to be brought against Dodgson jointly with the other defendants; he was not discharged absolutely, but only in such way as the Legislature has prescribed; and he was not bound to take the benefit of it, quivis renunciare potest juri pro se introducto."

Again, in Bonner v. Wilkinson (1822) 5 B and Ald 682 at p 686 (106 ER 1340 at p 1341) Abbott C.J. observed:

"It is certainly true that a party cannot, by his own private instrument, defeat the object of an Act of Parliament, but he may thereby waive a provision intended for his own benefit.

There is another maxim Omnes licentiam habere his quae pro se indulta sunt renunciare (everyone has liberty to renounce those things which are granted for his benefit) which is also of considerable age: see Rumsey v. The North-Eastern Rly. Co. (1863) 14 CB (NS) 641 at p 649 (143 ER 596 at p 600). Like most Latin maxims, these two, which are discussed in Broom's Legal Maxims, 10th ed. (1939), pp 477-478, assert rather than explain. The observation of Windeyer J. in Smith v. Jenkins (1970) 119 CLR 397 at p 410, (though in a different context) holds good:

"The intrusion of this Latin maxim into learned commentary, and also into judgments, has caused a confusion which would not have occurred if the writers had condescended to translation and had not taken the maxim into territory where it does not belong."

The point of referring to the maxims and to the early decisions is to illustrate the means by which the courts have grappled with the notion to which the label of waiver has been attached and to show that a combination of the two maxims is reflected in the general proposition that a person may waive both a substantive right and an available defence, in the sense that he may agree or choose not to rely upon the one or the other. The maxims embody concepts and that is their value. They are not definitions and, of course, they involve translations. There are translations of the maxims in Jowitt's Dictionary of English Law, 2nd ed. (1977), pp 1283, 1488. Broom's translation of the second maxim reads: "every man may renounce a benefit or waive a privilege which the law has conferred upon him": at p 478. Nothing in the maxims limits the terms "introduced", "granted" or "conferred" to a benefit arising in some formal way, for example by statute.

The idea that waiver involves a renunciation of a right or benefit comes through in the judgment of the Earl of Selborne L.C. in Great Eastern Railway Co. v. Goldsmid (1884) 9 App Cas 927. Speaking of a benefit granted to the City of London by a charter of 1st Edward III, his Lordship said, at pp 936-937:

"It (the benefit) is a jus introductum for the particular benefit of the city of London, and it falls within the general principle of law, 'Unusquisque potest renunciare juri pro se introducto;' a principle not only of ancient but also of modern application, applicable even where Acts of Parliament have been passed of a much more public character. In such cases, when the rights given have been only private rights, unless there has been also in the Act of Parliament a clause excluding a power of contract, it has been held that by contract or by voluntary renunciation such rights, as far as they are personal rights, may be parted with and renounced."

At times other expressions have crept into judgments. Thus, in Wright v. John Bagnall and Sons Limited [1900] 2 QB 240 , where agreement had been reached between employer and worker that there was a statutory liability on the former to pay compensation for an accident to the latter, the headnote speaks of evidence upon which a judge or arbitrator "may properly find that the employer is estopped from setting up the defence that the request for arbitration was not filed within six months of the accident". Yet the judgment of Collins L.J., with whom the other members of the Court of Appeal agreed, speaks, at p 244, of the respondent being "debarred" from raising the point that the statutory limitation applied. No member of the Court of Appeal uses the expression "estopped" or indeed "waived". See also Lubovsky v. Snelling (1944) KB 44.

The question is not merely one of terminology; however the law has to work with words. It is important, though, that labels are used with clarity and consistency. It may well be, as suggested in Spencer Bower and Turner, The Law Relating to Estoppel by Representation, 3rd ed. (1977), at p 318:

"Most of the cases which use the term 'waiver' (using the term in a sense bringing it within the ambit of this chapter) use it in the sense of 'election' as it is used here, and as different on ultimate analysis from estoppel proper."

On one view, that is merely to replace one legal label with another. But waiver, seen as a form of election, does have characteristics that distinguish it from estoppel, though there may be occasions when the distinction is not an easy one to draw. The distinction needs further consideration in these reasons. What it is necessary to do is to give to waiver, in any sense relevant for the disposition of this appeal, the aspect that it should properly bear.

Much of the confusion is due to the fact that, as Edward L. Rubin, in a learned article entitled "Toward a General Theory of Waiver", (1981) 28 UCLA Law Review 478, observes: "Waivers appear in almost every area of law and in connection with almost every type of legal right": at p 478. Rubin comments at p 479:

"Precisely the same term with precisely the same effect is used in both criminal and civil law, yet no theory of waiver bridges these two areas."

Equally, it may be said, the term is applied to events which occur within the adjudicative process and events which occur prior thereto. In seeking for a definition of waiver, the author points to the difficulties arising from the emphasis in the familiar definition of waiver on "an intentional relinquishment ... of a known right or privilege": Johnson v. Zerbst (1938) 304 US 458, at p 464. The same emphasis is to be found in Craine, at p 327. But it has not seemed possible, whether in this country or elsewhere in the common law world, to reconcile the requirement of intention with the notion that waiver may be inferred from conduct. It may be noted that in Craine, at p 326, Isaacs J. spoke of "intentional" as "such as either expressly or by imputation of law indicates intention ...". See also Gresson P in Auckland Harbour Board v. Kaihe (1962) NZLR 68, at p 88. Because of the difficulty in effecting that reconciliation, it is preferable to speak of waiver in the broad way in which it was referred to by Mason J. in Sargent, though it is still necessary to identify the "legal grounds" on which a person is precluded from asserting a right or from raising a particular defence.

Necessarily, questions have to be answered. In what circumstances does waiver operate? In particular, when is a defendant precluded from relying upon a defence such as a plea of limitations? To attempt an answer to the general question first. A defendant may be precluded from asserting a right or from raising a defence, as a matter of contract. In Tropical Traders Ltd. v. Goonan (1964) 111 CLR 41 , where this Court was considering the consequences of repeated acquiescence by one party to a contract in non-observances by the other party of stipulations as to time, Kitto J. commented, at p 52, that "it may not matter whether the result is described as a promissory estoppel or a waiver or a variation of the contract by mutual, though tacit, consent". Yet, variation of contract does stand in a category of its own since, as the law presently stands, consideration is necessary to support the altered bargain. In many cases where waiver is sought to be argued, consideration is not present; the situation may be an entirely non-contractual one, as in Vakauta v. Kelly (1989) 167 CLR 568 , where the question was whether a party to litigation, with knowledge of the right to take objection on the ground of bias on the part of the trial judge, lost that right through failure to take objection at the appropriate time .

There may be much to be said for the suggestion made by Wootten J. in Wilson v. Kingsgate Mining Industries Pty. Ltd. (1973) 2 NSWLR 713, at p 730, and noted in Cheshire and Fifoot's Law of Contract, 5th Aust. ed. (1988), pp 628-629, "that the time is at hand when these various approaches to the same or similar problems should be rationalized under one subsuming principle". The learned authors offer, as a subsuming principle, the statement by Dixon J. in Thompson v. Palmer (1933) 49 CLR 507 , at p 547 (though said in regard to estoppel in pais), that the law will not allow "an unjust departure by one person from an assumption adopted by another ... which ... would operate to that other's detriment".

Whether or not the time is ripe for a subsuming principle, the shape that this case has taken since its inception makes the formulation here of such a principle inappropriate. More importantly, such a formulation has sweeping implications, not only for waiver and estoppel, but for other areas of the law as well, not least the doctrine of consideration. In the circumstances of the present case, the efficacy of the Commonwealth's limitations defence must fall to be determined within the language of waiver or estoppel.

In my view, waiver, by that name, has a role to play. And it is a role which involves no confusion with variation of contract or promissory estoppel. It may be seen as a form of election between inconsistent rights, in the former of the categories mentioned by Mason J. in Sargent, at p 655. Election implies that a choice must be made between two rights which are mutually exclusive. "Obviously there can be no election, choosing one course to the exclusion of another, when in fact there is only one course to take, or where the two courses are such that the adoption of one of them does not necessarily indicate a final intention to abandon the other": Spencer Bower and Turner, at p 342. But, in the second of those categories, "the legal grounds on which a person is precluded from raising a particular defence ...", an election is involved, only in the sense that a defendant may choose to take a jurisdictional point, rely upon an irregularity in the proceedings, plead a particular defence, or take some other step in the adjudicative process, or he may choose not to do so. But he may not take up "two inconsistent positions", the language used in Craine, at p 326. It may be, therefore, that "election" is best reserved for the former of Mason J.'s categories and "waiver" for the latter. The distinction is noted by Lord Wright in Smyth and Co. v. Bailey and Co., at p 70.

Halsbury's Laws of England, 4th ed. (1976), vol 16, par.1471, part of which reflects the judgment of Lord Hailsham of St Marylebone L.C. in Banning v. Wright (1972) 1 WLR 972 , at p 979; [1972] 2 All ER 987 , at p 998, offers the following definition which comes close to this notion; nevertheless, it is open to the objection that it tends to blur different concepts:

"Waiver is the abandonment of a right in such a way that the other party is entitled to plead the abandonment by way of confession and avoidance if the right is thereafter asserted, and is either express or implied from conduct. It may sometimes resemble a form of election, and sometimes be based on ordinary principles of estoppel, although, unlike estoppel, waiver must always be an intentional act with knowledge."

In this appeal we are concerned with waiver as it exists within the adjudicative process. It is commonplace to speak of a person "waiving" a right, for instance, by submitting to the jurisdiction of a court which otherwise has no jurisdiction over him, by not insisting upon arbitration , where an agreement so provides, by not taking advantage of some irregularity in proceedings or by not pressing a particular argument that is available at trial: see, by way of example, Graham v. Ingleby (1848) 1 Ex 651 (154 ER 277); Park Gate Iron Co. v. Coates (1870) LR 5 CP 634; Ward v. Raw (1872) LR 15 Eq. 83; Fry v. Moore (1889) 23 QBD 395; Shrager v. Basil Dighton Ltd. [1924] 1 KB 274 ; Water Board v. Moustakas (1988) 62 ALJR 209; 77 ALR 193.

Waiver, in the sense used for the purposes of this appeal, may be found in the deliberate act of a defendant not to rely upon a defence available to him. That is not to say that there must be an intention to bring about the consequences of waiver; rather, the conduct from which waiver may be inferred, must be deliberate. Detriment is not an essential attribute of waiver, though it will often be found as a consequence. Within the adjudicative process at any rate, it is enough that the defendant "renounces" a defence which is available to him and which is there for his benefit.

A defence available to a defendant, whether it be on the facts or on the law, is not waived merely because the defendant does not initially plead that defence. It is commonplace for pleadings to change as an action progresses, whether by way of expansion or contraction (though usually the former). But what happened in the present case is more fundamental than a pleading point. The situation is not simply that the Commonwealth had filed a defence to the action and later sought to add a plea of limitation to that defence. The Commonwealth sought to adopt a position which had been open to it at the outset but which it deliberately chose not to make a part of the issues for adjudication. The stance of the Commonwealth from the beginning, consistent with its communications to Mr Verwayen's advisers, was that it was not relying upon the Limitation Act, indeed that it was not defending the action, save as to the amount of damages to be awarded to Mr Verwayen. It was on that footing that the plaintiff pursued his action for damages right through to the listing of the action for the assessment of damages. To uphold a limitations plea in those circumstances would be to permit the Commonwealth to rely upon a defence which it had unequivocally renounced.

Limitation Act provides that no action for damages for negligence in respect of personal injuries "shall be brought after the expiration of three years after the cause of action accrued". Such a provision has been long held to go, not to the jurisdiction of a court to entertain a claim, but to the remedy available and hence to the defences which may be pleaded. Absent a plea of limitations, in those circumstances the matter does not arise for the consideration of the court: see Ronex Properties v. John Laing (1983) QB 398 and the authorities there cited. Equally, it is well established that a provision such as s 5(6) is procedural in nature: Ketteman v. Hansel Properties, at p 219; see also Admiralty Commissioners v. Valverda (Owners) (1938) AC 173, at p 185. Such a right is capable of being waived if it is for the benefit of the party concerned: Phillips v. Martin (1890) 11 NSWLR 153; Wilson v. McIntosh; Brown v. The Queen (1986) 160 CLR 171 , at p 178. The Commonwealth renounced and thereby waived its right to rely upon the defence of limitations, a defence which was otherwise available to it and which was for its benefit.4

The passage from Halsbury, referred to earlier in these reasons, concludes:

"The waiver may be terminated by reasonable but not necessarily formal notice unless the party who benefits by the waiver cannot resume his position, or termination would cause injustice to him."5

Waiver, in the sense relevant for the purposes of this appeal, is not capable of being withdrawn. It is of the essence of waiver in this sense that the defendant has unequivocally renounced his right to rely upon the particular defence; once the defence has been so renounced (and this will be hard to establish except in a case as clear as this one), the defendant should not be permitted to rely upon it. In this context, to say that waiver may be terminated unless the other party cannot resume his position is, once again, to move into the area of estoppel. This is made evident by the authority cited by Halsbury in support of the proposition - Ajayi v. R.T. Briscoe (Nig.) Ltd. (1964) 1 WLR 1326 ; [1964] 3 All ER 556 speaks expressly of promissory estoppel. As to the causing of injustice to the other party, the authority cited is W.J. Alan and Co. v. El Nasr Export [1972] 2 QB 189 . That case concerned contracts for the sale of coffee and a question as to the currency in which payment should be made. Lord Denning M.R. spoke of waiver, but in relation to the enforcement of rights under a contract, adding, at p 213:

"It may be too late to withdraw: or it cannot be done without injustice to the other party. In that event he is bound by his waiver. He will not be allowed to revert to his strict legal rights."

This is waiver in the first of the two senses mentioned by Mason J. in Sargent. It is the second of those senses with which this appeal is concerned. It follows that I disagree with the view expressed in Kerrison v. Martin and Heyward (1975) VR 401, at p 405, that "any such unilateral waiver without consideration can be terminated at any time", at least if it is thought to apply to waiver as I have sought to identify it in the present case. Because waiver, in this sense, involves unequivocal renunciation or abandonment of a defence, it may occur at any stage of the adjudicative process. In the ordinary course, proof that there has been such a renunciation or abandonment will be the harder to establish, the earlier the stage reached in that process. But that is an evidentiary problem; it does not mean that a particular stage in the adjudication process must be reached before waiver of a defence may occur.6

What I have said regarding the limitations plea applies equally to the defence that the Commonwealth was engaged in combat exercises. There is no statutory provision involved but, in the sense in which waiver is used in these reasons, the Commonwealth had available to it, by way of answer to Mr Verwayen's claim, a contention that it owed no duty of care to him in the circumstances. The distinction drawn by Windeyer J. in Australian Iron and Steel Ltd. v. Hoogland (1962) a condition of the remedy rather than an element in the right" was drawn in the context of a limitations defence and is not necessarily apposite in all situations in which waiver arises. For the reasons already given, the Commonwealth must be taken to have renounced its right to rely upon a defence that, in the circumstances of this case, it owed no duty of care to Mr Verwayen. There is no operative difference between this defence and the limitations plea. In one case the Commonwealth said to Mr Verwayen: "There is available to us an argument that, in the circumstances prevailing at the time you were injured, we owed you no duty of care. But we abandon any such argument." In the other case the Commonwealth said: "There is available to us an argument that any claim you might otherwise have is statute barred. But we abandon any such argument." To abandon the combat exercise defence offended no statutory provision, whether for the benefit of the public or otherwise, and no rule of public policy. In all the circumstances the Commonwealth must be held to have renounced that particular answer to Mr Verwayen's claim.

The conclusions I have reached make it unnecessary to deal with the argument based on estoppel. But on the present state of the authorities, the consequence of any promissory estoppel is that the court should enforce the promise only as a means of avoiding detriment and to the extent necessary to achieve that end: see Waltons Stores (Interstate) Ltd. v. Maher (1988) 164 CLR 387 , at pp 404-405, 423-427. In Mr Verwayen's case, that end may be achieved by compensating him for what he has lost by way of costs in pursuing the action. If there are other factors to be taken into account (and before O'Bryan J. counsel for the respondent contended that there were such factors), they have not been established in the proceedings that have taken place. Whether Mr Verwayen should now be given the opportunity to supplement the material relevant to detriment is a matter I need not pursue for he has succeeded in making good his claim of waiver.

Something must be said, however, about the consequences of these conclusions for this appeal, having regard to the present state of the pleadings. By its amended defence the Commonwealth denied negligence and pleaded Limitation Act and that, by reason of combat exercises in which the Voyager and the Melbourne were engaged at the relevant time, no duty of care was owed to Mr Verwayen. The reply to that defence asserted, inter alia, that the Commonwealth had waived or was estopped from raising the limitations defence and the combat exercise defence. The reply did not assert that the Commonwealth had waived or was estopped from raising its general denial of negligence. Those pleadings stand though possible consequences were raised by this Court during the hearing of the appeal. A conclusion that the Commonwealth waived the limitations defence and the combat exercise defence does not directly touch the general denial of negligence.

Having regard to the issues raised by the appeal, it is appropriate that the appeal be dismissed. It may well be that the Commonwealth will agree to the action proceeding to trial for the assessment of damages only, since the combat exercise defence was the only basis upon which the absence of a duty of care was sought to be raised and, once that issue has been removed from the action, negligence may no longer be contested. But that is a matter to which Mr Verwayen's advisers must direct their attention and, if necessary, seek an amendment to the reply to assert that the Commonwealth is precluded from denying negligence. In all the circumstances, it is hard to see any basis on which such an application may be refused. Although the denial of negligence presently remains in the defence, the effect of this judgment is that the Commonwealth is held to its admission of liability. The only matter on which the Commonwealth should now be heard is as to the amount of damages to which Mr Verwayen is entitled .

The appeal should be dismissed.