2. Set aside the judgment and orders made by the Full Court of

(b) the judgment and orders of Debelle J. be set aside to

the extent that they rejected the plaintiff's claim

(c) the case be remitted to the Supreme Court for the

assessment of damages in respect of the plaintiff's

accordance with the reasons for judgment of this Court.

4. The respondent pay the appellant's costs of this appeal and in

the Full Court.

DEANE, DAWSON, TOOHEY AND GAUDRON JJ. The background facts and the issues involved in this appeal are set out in the judgment of McHugh J.

2. A plaintiff in an action in negligence is not entitled to recover damages for loss of earning capacity unless he or she establishes that two distinct but related requirements are satisfied. The first of those requirements is the predictable one that the plaintiff's earning capacity has in fact been diminished by reason of the negligence-caused injury. The second requirement is also predictable once it is appreciated that damages for loss of earning capacity constitute a head of damages for economic loss awarded in addition to general damages for pain, suffering and loss of enjoyment of life. It is that "the diminution of ... earning capacity is or may be productive of financial loss" ((1) Graham v. Baker [1961] HCA 48; (1961) 106 CLR 340 at 347, per Dixon CJ, Kitto and Taylor JJ).

3. It was submitted on behalf of the present appellant ("the plaintiff") that, in a case such as the present, the approach that damages for loss of earning capacity are recoverable only if, and to the extent that, the diminution of earning capacity is or may be productive of financial loss is inconsistent with the reasoning of the majority of the Court in Griffiths v. Kerkemeyera ((2) [1977] HCA 45; (1977) 139 CLR 161), as that reasoning was explained in the majority judgments in Van Gervan v. Fentona ((3) [1992] HCA 54; (1992) 175 CLR 327). That is not so. No doubt, the considerations which supported the conclusion in those and other cases to the effect that compensation for the increased needs of a plaintiff should not be reduced to take account of the extent that those needs have been or will be satisfied by gratuitous services are applicable to preclude the reduction of damages for loss or impairment of earning capacity by reason of the financial or other support provided by relatives or friends to reduce the deprivations of unemployment. Those considerations are not, however, applicable to entitle a plaintiff to be compensated for diminution in earning capacity as a distinct additional head of economic loss in circumstances where the diminution has had and will have no adverse effect on actual earnings and will be productive of no economic loss.

4. Nor are we persuaded that, in dealing with the plaintiff's claim as presented at first instance in the present case, the learned trial judge (Debelle J.) lost sight of the fact that in an action in negligence in this country an injured plaintiff recovers damages for loss or impairment of earning capacity as distinct from the direct recovery of past or future lost lost earnings ((4) Arthur Robinson (Grafton) Pty. Ltd. v. Carter [1968] HCA 9; (1968) 122 CLR 649 at 658; O'Brien v. McKean [1968] HCA 58; (1968) 118 CLR 540 at 546). It is true that, at a critical part of his judgment, his Honour stated that the plaintiff had "not discharged the onus of proof that he lost his earning capacity for four and one half years by reason of injuries sustained in the accident" and that there are a number of other statements in his judgment to the like effect. Those statements must, however, be read in a context where his Honour apparently saw the plaintiff's claim for damages for loss of earning capacity as being pressed only in relation to the financial loss actually sustained by him as a result of the premature termination of his appointment as Professor of Philosophy at Flinders University. Thus, he summarized that aspect of the plaintiff's case as being a claim for "loss of salary and long service leave entitlements for the four and a half year period from 30 June 1988 to 10 December 1992 which he had lost in consequence of his early retirement". His Honour correctly identified the principles applicable to the claim as so formulated:

"The plaintiff can only recover damages for the loss of four

and a half years salary if he can prove that his earning

capacity has been diminished ... Although the defendant

must take the plaintiff as it finds him, including his

commitment to high standards and to academic excellence, the

plaintiff nevertheless has this burden to discharge. He

will be entitled to recover only if he can prove a

diminution in his earning capacity which has been or may be

productive of financial loss. If, notwithstanding any

impairment, his contract of employment and his right to draw

a salary continue, his impairment has not resulted in any

financial loss: see Graham v Baker ((5) (1961) 106 CLR at

347. per Dixon CJ Kitto and Taylor JJ." (emphasis added)

In the context of that statement of applicable principle, it seems to us that, on balance, the subsequent reference to loss of earning capacity "for four and one half years" and other statements to similar effect should be seen as "shorthand" references to the dual requirements of loss of earning capacity and resulting financial loss which his Honour identified.

5. The last sentence of the above extract from the learned trial judge's judgment reflects the following statement in the joint judgment in Graham v. Baker ((6) ibid. at 347):

"And if, notwithstanding such impairment, both his contract

of employment and his right to ordinary wages continue, how

can it be said that his impairment has resulted in any loss

so far as his earning capacity is concerned?"

That statement was not directly applicable to the plaintiff's claim for loss or impairment of earning capacity in circumstances where his contract of employment and right to draw a salary had been prematurely terminated by voluntary retirement. Its relevance to the present case is that the evidence indicated, and his Honour found, that, if the plaintiff had not elected to retire, the contract of employment and the right to draw a salary would have continued throughout the remaining four and one half years until he reached the mandatory retirement age of sixty-five with the consequence that he would have suffered no financial loss in relation to that period((7) It was not suggested that the plaintiff would, but for the accident, have earned additional income over and above the salary and other emoluments of his professorial appointment.). In those circumstances, the plaintiff's claim to recover damages for lost earning capacity quantified by reference to the actual loss of University salary and other benefits during that particular four and a half year period rested upon the proposition that, notwithstanding that it flowed from his own decision to accept voluntary retirement, the premature termination of the plaintiff's University appointment had been relevantly "caused" by an accident-related loss of earning capacity. If that proposition were made good, the plaintiff would be prima facie entitled to recover, as damages for loss of earning capacity, the difference between what he would have earned during those four and one half years in his University appointment and any amount which, notwithstanding his reduced earning capacity, he had earned or could reasonably be held capable of earning during that period ((8) The prima facie entitlement would, as the trial judge recognized, be subject to any adjustment to prevent double compensation.).

6. For the purposes of the law of negligence, the question whether the requisite causal connection exists between a particular breach of duty and particular loss or damage is essentially one of fact to be resolved, on the probabilities, as a matter of common sense and experience ((9) See Fitzgerald v. Penn [1954] HCA 74; (1954) 91 CLR 268 at 277-278; March v. Stramare (E and M.H.) Pty. Ltd. [1991] HCA 12; (1991) 171 CLR 506 at 515, 522-523; Bennett v. Minister of Community Welfare [1992] HCA 27; (1992) 176 CLR 408 at 412-413, 418-419, 428.). And that remains so in a case such as the present where the question of the existence of the requisite causal connection is complicated by the intervention of some act or decision of the plaintiff or a third party which constitutes a more immediate cause of the loss or damage. In such a case, the "but for" test, while retaining an important role as a negative criterion which will commonly (but not alwaysalwaysa) exclude causation exclude causation if not satisfied, is inadequate as a comprehensive positive test ((10) See, e.g., March v. Stramare (E and M.H.) Pty. Ltd. (1991) 171 CLR at 515-519, 522-524). If, in such a case, it can be seen that the necessary causal connection would exist if the intervening act or decision be disregarded, the question of causation may often be conveniently expressed in terms of whether the intrusion of that act or decision has had the effect of breaking the chain of causation which would otherwise have existed between the breach of duty and the particular loss or damage. The ultimate question must, however, always be whether, notwithstanding the intervention of the subsequent decision, the defendant's wrongful act or omission is, as between the plaintiff and the defendant and as a matter of common sense and experience, properly to be seen as having caused the relevant loss or damage. Indeed, in some cases, it may be potentially misleading to pose the question of causation in terms of whether an intervening act or decision has interrupted or broken a chain of causation which would otherwise have existed. An example of such a case is where the negligent act or omission was itself a direct or indirect contributing cause of the intervening act or decision. It will be seen that, on the plaintiff's evidence, the present was such a case.

7. Nor can the question of causation of damage in a negligence action be automatically answered by classification of operative causes as "pre-eminent" or "subsidiary". Regardless of such classification, two or more distinct causes, without any one of which the particular damage would not have been sustained, can each satisfy the law of negligence's common sense test of causation ((11) See, e.g., March v. Stramare (E and M.H.) Pty. Ltd. [1991] HCA 12; (1991) 171 CLR 506). This can be most obviously so in a case where a "subsidiary" cause operates both directly as a cause of the particular damage and indirectly as a contributing component of a "pre-eminent" cause. As will also be seen, the findings of the learned trial judge left open the likelihood that the present was such a case.

8. In the present case, the trial judge made a number of specific findings of fact on the basis of which he concluded that the premature termination of the plaintiff's University appointment had not been caused by the injuries which the plaintiff received in the relevant accident. He found that the plaintiff was, at the time of his premature retirement, suffering from chronic pain arising from those injuries:

"The pain was at a low level and did not usually require

pain relief medication. However, if the plaintiff

over-exerted himself physically, the pain could become quite

intense and require medication. The pain was at a

sufficient level to cause him to wake at least once on most

nights. This left the plaintiff feeling tired. He did not

have the intellectual energy which, he believed, was

required to discharge his duties to his complete

satisfaction."

At earlier stages of his judgment, his Honour had found that when the plaintiff "tires, he begins to limp and this in turn aggravates the back injury causing further pain" and that the injuries to the plaintiff's right acetabulum and right hip were so severe that they were likely to result in osteoarthritis in the right hip joint with the consequence that, "towards the end of 1995 or 1996", it would be necessary that the plaintiff undergo hip replacement surgery.

9. His Honour's further findings included the following: that, notwithstanding the effects of the accident, the plaintiff remained "a competent teacher" with "the teaching and administrative skills to retain the position as head of the Discipline of Philosophy"; that "there is no evidence that the plaintiff was not performing his duties to the satisfaction of the University"; that the plaintiff "could have retained his position as head of the discipline of Philosophy until he retired at the age of 65 years on 10 December 1992"; that the "pre-eminent reason why the plaintiff decided to retire early was to be able to devote as much time as possible to research and creative philosophy untroubled by the requirements of University life and in particular the administrative and teaching duties required of him"; and, that a subsidiary reason for the plaintiff's early retirement was that, as a result of his own high standards, he believed he should retire for the reason that "he was not performing at the level he thought desirable". In the context of the other findings which he made, his Honour appears to have treated his finding of "pre-eminent reason" as effectively defeating the plaintiff's claim for damages for loss of earning capacity.

10. The above express primary findings of fact made by the learned trial judge did not, however, suffice to provide an answer to the question whether, for relevant purposes, the premature termination of the plaintiff's University appointment had been caused by the injuries which the plaintiff had sustained in the relevant accident. Clearly enough, the chronic and sometimes intense pain, the nightly disturbance of sleep and the associated loss of "intellectual energy" which his Honour found resulted from the plaintiff's injuries had caused some diminution in his earning capacity. The plaintiff's evidence, if accepted, supported a conclusion that, accepting the findings of the trial judge, those effects of his injuries were a causative influence in the premature termination of his employment in two distinct ways. First, they were a contributing cause of his decision to retire in that their effect was to decrease his ability to discharge the duties of his appointment at what he saw as an acceptable standard. That is to say, they were a contributing cause of what his Honour found to have been a subsidiary "reason for the plaintiff's early retirement". Secondly, and more importantly for present purposes, they were also a contributing cause of what his Honour found to have been the "pre-eminent" reason why the plaintiff decided to retire early", namely, a desire "to devote as much time as possible to research and creative philosophy untroubled by the requirements of University life and in particular the administration and teaching duties required of him". As we read the whole of the plaintiff's evidence, including a letter of 26 May 1987 upon which both the trial judge and the members of the Full Court placed particular reliance, the plaintiff's case was to the effect that the pain, fatigue and loss of intellectual energy caused by the accident had led to a situation where the discharge of his administrative and teaching duties absorbed much more of his time and capacity than would otherwise have been the case and precluded him from doing the research and creative work which he would otherwise have been able to do: as he wrote in that letter, teaching "uses up all the energy I have available. Non-teaching periods of the year are mostly consumed by teaching - either preparation or recovery." That being so, the trial judge's findings that the "pre-eminent reason why the plaintiff decided to retire early was to be able to devote as much time as possible to research and creative philosophy" and that a subsidiary reason for early retirement was that the plaintiff believed he should retire for the reason that "he was not performing at the level he thought desirable" did not suffice to answer any question of causation in the present case. Rather, those findings gave rise to the question whether, taking account of any part played by the plaintiff's injuries as a contributing cause of each of those reasons, it should be concluded that the plaintiff's premature retirement was, for relevant purposes, caused by the accident in the sense that it was a product of the loss of his earning capacity resulting from the injuries which he had received in it. His Honour did not expressly deal with that question or with the combined significance of those different causative effects.

11. Nor did the learned trial judge expressly consider the subsidiary question whether the plaintiff's premature retirement would have occurred "but for" the plaintiff's accident-caused injuries. If, in the context of the plaintiff's express evidence that he believed that "if it had not been for the accident and the consequent pain" he "would still be teaching at Flinders", that subsidiary "but for" question were answered in the negative, that negative answer would be a factor supporting, but not compelling, the conclusion that the premature termination of the plaintiff's appointment had been caused by the accident.

12. We have carefully considered Debelle J.'s judgment for the purpose of determining whether it is possible to discern in it either an implicit finding that the plaintiff's premature retirement would have occurred in any event or an implicit finding that the plaintiff's accident-caused diminution in capacity had not been a contributing cause of "the pre-eminent reason", as well as the subsidiary reason, of his decision to retire prematurely. While the matter is far from clear, it appears that his Honour did not direct his attention to either of those matters and that it is impossible to discern in his judgment any implicit findings in relation to either of them. Examination of the judgment of Millhouse J. (with which Mohr and Bollen JJ. concurred) indicates that the Full Court also failed to advert to those questions. The explanation of the trial judge's and Full Court's failure in that regard would seem to be that their Honours mistakenly considered that the trial judge's finding that the pre-eminent reason for the plaintiff's retirement was his desire to devote as much time as possible to research and creative philosophy automatically defeated the plaintiff's claim for damages for loss of earning capacity. The result is that, subject to the effect of a finding by the trial judge that the plaintiff's decision to retire was not "reasonable", to which reference is made below, the learned trial judge and the Full Court failed to deal adequately with that aspect of the plaintiff's case.

13. Specific mention should be made of the significance of the trial judge's finding that, if the plaintiff had not decided to resign his appointment, he would have remained in the employ of the University for the four and a half years that remained until his retirement on reaching sixty-five. The members of the Full Court treated the plaintiff's failure to prove that he "could not have continued in his position as Professor of Philosophy and head of the Discipline until he reached the normal retiring age" as, of itself, requiring rejection of the plaintiff's claim for damages for loss of earning capacity. The basis of that approach would seem to be an assumption that the fact that an employer would not have dismissed an injured plaintiff who had not voluntarily retired, automatically means that the termination of that particular employment by such retirement cannot be seen as the product of an accident-caused loss of earning capacity. Such an assumption is mistaken. The necessary causation between a defendant's negligence and the termination of a plaintiff's employment, in the sense that the termination of the employment is the product of an accident-caused loss of earning capacity, can exist notwithstanding the fact that the immediate trigger of the termination of the employment was the plaintiff's own decision to retire prematurely. If, for example, it appears that a plaintiff's decision to retire prematurely would not have been made were it not for the fact that the effect of accident-caused injuries is that continuation in employment would subject him or her to constant pain and serious risk of further injury, it may well be that common sense dictates the conclusion that the plaintiff's decision to retire prematurely was a natural step in a chain of causation which suffices to designate, for the purposes of the law of negligence, the termination of the employment as a product of those injuries.

14. The learned trial judge did not fall into the error of treating his finding that the plaintiff could have continued in his University appointment as automatically defeating his claim for damages for loss of earning capacity. Instead, his Honour embarked upon the task of "determining whether the plaintiff should have remained in his employment". His Honour expressed the view that, in determining that question, "the proper course is to apply a test that asks whether, in the light of the medical condition of the plaintiff and the fact that his post remained available to him and the other factors affecting him, the plaintiff's decision to retire early was reasonable". One can point to some support in the decided cases for his Honour's approach in that regardaWith due regard ((12) Glavonjic v. Foster [1979] VicRp 52; (1979) VR 536 at 539-540; Karabotsos v. Plastex Industries Pty. Ltd. [1981] VicRp 65; (1981) VR 675 at 677-679, 683; Hisgrove v. Hoffman (1981) 29 SASR 1 at 8.). With due respect, however, the question whether a sixty year old man who has sustained permanently incapacitating injuries of the kind sustained by the plaintiff "should" continue in his employment or is "acting reasonably" in accepting premature retirement was not the appropriate one. As has been mentioned, the evidence established and the trial judge found, that the injuries sustained by the plaintiff caused him chronic and sometimes intense pain and that the combined effect of the pain and consequent loss of sleep was to reduce the plaintiff's intellectual energy to an extent which prevented him from discharging the duties of his office to the standards which he considered appropriate. As has also been mentioned, the plaintiff's evidence supported a finding that the effects of the accident were that his routine administrative and teaching work used up all his energy with the consequence that he was incapable of doing the research and creative work which he desired to carry out and which, one would think, would be and should be expected of the holder of the Chair of Philosophy in a major university. In these circumstances, the relevant question was not whether the plaintiff "should" have continued in his University post or whether his decision to retire was not "reasonable" but whether, in the context of what was reasonable between the plaintiff and the defendant in determining the defendant's liability in damages, the premature termination of the plaintiff's employment was the product of the plaintiff's loss of earning capacity notwithstanding that it was brought about by his own decision to accept voluntary retirement.

15. Counsel for the plaintiff further submitted that, even if the conclusion had been properly reached that the premature termination of the plaintiff's University appointment had not been relevantly caused by the accident, that was not necessarily the end of the plaintiff's claim for damages for loss of earning capacity in relation to the four and a half year period until the plaintiff reached sixty-five years. There is obvious force in that submission. A plaintiff is not precluded from recovering damages for loss of earning capacity merely by reason of the fact that he or she voluntarily left employment which was unsuitable or in which he or she was unhappy. The continued availability of such employment will, of course, be relevant to the question of the existence and extent of any loss of earning capacity and a finding that a plaintiff's termination of employment was not the product of an accident-caused loss of earning capacity will necessarily preclude the calculation of damages on the basis that it was. Such a finding does not, however, mean that damages cannot be recovered for loss of earning capacity in relation to the period subsequent to the termination of that employment if there is in fact an accident-caused loss of earning capacity which has been or will be productive of financial loss during that subsequent period. Counsel for the plaintiff also submitted that both the learned trial judge and the Full Court had failed to deal at all with the plaintiff's claim for loss of earning capacity in so far as it related to the period after the date on which the plaintiff would have retired in any event on reaching the age of sixty-five years. Not all the material in evidence has been reproduced for the purposes of the appeal to this Court. The material before us does, however, indicate an absence of evidence addressed to the question of what, if any, financial loss had been or would be sustained by the plaintiff after retirement. As we have indicated, the learned trial judge apparently saw the plaintiff's claim to loss of earning capacity as being confined to financial loss sustained by reason of the premature termination of his University appointment. The Court was, however, informed by counsel for the plaintiff, without dissent on the part of counsel for the respondent, that the claim for damages for loss of earning capacity had never in fact been so confined.

16. It follows from the foregoing that it is necessary to reconsider the plaintiff's claim for damages for loss of earning capacity both in relation to financial loss allegedly sustained by reason of the premature termination of his University appointment and in relation to any additional financial loss that has been or will be sustained in retirement.

17. There remains for consideration the question whether the case should be remitted to the Supreme Court for reconsideration of the plaintiff's claim for damages for loss of earning capacity or whether this Court should itself undertake the task of determining that claim. The preferable course is that adopted by McHugh J., namely, that this Court determine whether the premature termination of the plaintiff's University appointment was the product of the plaintiff's relevant loss of earning capacity but otherwise remit the case to a judge of the Supreme Court for the assessment of damages. We agree with McHugh J. that the question whether the premature termination of the plaintiff's University appointment was the product of the plaintiff's diminution of earning capacity should be answered in the affirmative. Subject to one qualification, we also agree with his Honour's reasons for that answer in all the circumstances of this case. The qualification is that, as indicated, any question of reasonableness should be framed in terms of what is reasonable as between the plaintiff and the defendant in the context of assessing damages for negligence rather than as a question of whether the plaintiff acted reasonably or unreasonably in resigning his post. However, that qualification is not important for the purposes of the present case since we consider that, for the reasons which McHugh J. gives, the learned trial judge's finding that the plaintiff's decision to retire was not reasonable cannot stand and that any question of reasonableness should be answered favourably to the plaintiff.

18. The appeal should be allowed. The judgment and orders of the Full Court of the Supreme Court of South Australia should be set aside. In lieu thereof, it should be ordered that the appeal to that court be allowed, that the judgment and orders of Debelle J. be set aside to the extent that they rejected the plaintiff's claim for damages for loss of earning capacity and that the case be remitted to the Supreme Court for the assessment of damages by the learned trial judge or another judge (or Master) of that court in respect to that claim in accordance with the reasons for judgment of this Court. It will be a matter for the judge of the Supreme Court dealing with the matter to determine whether, if an application to that effect is made, the parties should be allowed to lead further evidence. It should further be ordered that the respondent pay the plaintiff's costs both in this Court and in the Full Court.

McHUGH J. The main issue in this appeal is whether the plaintiff/appellant, who retired from his employment at Flinders University in South Australia four and a half years before reaching the normal University retirement age of 65 years, is entitled to recover the equivalent of four and a half years salary as compensation for loss of earning capacity. The plaintiff claims that his retirement was caused by injuries that resulted from the defendant's negligence. A further and related issue is whether the plaintiff is entitled to compensation for loss of earning capacity in relation to the period after age 65. The factual background

2. The plaintiff was involved in a motor vehicle accident on New Year's Eve 1983. He sustained serious injuries and still suffers continuing pain that is a source of constant discomfort. At the time of the accident, the plaintiff was aged 56. He was the Foundation Professor of Philosophy at Flinders University. His duties included lecturing, research, participation in seminars, supervision of postgraduate students and routine administrative duties. Before the accident, he had been very active in the intellectual life of the University and in community affairs. He was very fit and enjoyed a wide variety of physical pursuits. His injuries now restrict his capacity to pursue those activities. The plaintiff resumed his university duties in March 1984, but on 30 June 1988 he retired in accordance with the University's early retirement scheme. He was then aged 60. He was not due to retire until 10 December 1992, at the age of 65.

3. The action was tried in the Supreme Court of South Australia by Debelle J. who made the following relevant findings of fact:

1. By 1987, the plaintiff had made a good recovery from his

injuries although he was still suffering from chronic

pain at a low level. The pain caused him to wake at

least once on most nights which left him feeling tired.

While the plaintiff usually did not require medication

for pain relief, if he over-exerted himself the pain

could become quite intense and require medication.

2. The pre-eminent reason for the plaintiff's retirement was

his desire to devote as much time as possible to research

and creative philosophy untroubled by the requirements of

university life, particularly his teaching and

administrative duties. However, another reason for his

retirement was his belief that, as he was not performing

at the level he thought desirable, he should resign. But

this was not the main reason for his retirement.

3. The plaintiff was still a competent teacher who had the

teaching and administrative skills necessary to retain

his position as head of the Discipline of Philosophy. He

could have retained that position until he retired at the

age of 65 years on 10 December 1992. There was no

evidence that the University believed that either his

teaching or his administrative skills were

unsatisfactory.

4. Debelle J. said that the plaintiff could only recover damages for the loss of four and a half years of salary if he could prove that his earning capacity had been diminished and that the diminution had been or may be productive of financial loss ((13 Graham v. Baker [1961] HCA 48; (1961) 106 CLR 340). His Honour did not accept that the plaintiff's earning capacity was diminished. His Honour also said that, by deciding to retire early, the plaintiff had not only failed to mitigate his loss but had caused it to be increased. The learned judge said:

"In a case such as this, the question whether the plaintiff

has taken reasonable steps to mitigate or avoid his loss

turns into a question of whether the plaintiff has proved

that he has lost his earning capacity."

"In determining whether the plaintiff should have remained

in his employment, it is not appropriate to apply a wholly

objective test. Nor is it appropriate simply to apply a

subjective test and accept the plaintiff's reasons. I think

the proper course is to apply a test that asks whether, in

the light of the medical condition of the plaintiff and the

fact that his post remained available to him and the other

factors affecting him, the plaintiff's decision to retire

early was reasonable."

His Honour held that the plaintiff's decision to retire was not reasonable and that he had not proved that he had lost his earning capacity for four and a half years by reason of injuries sustained in the accident.

5. Debelle J. awarded the plaintiff damages of $77,517 which did not include any component for loss of earning capacity. However, his Honour calculated that the plaintiff would have been entitled to further damages of approximately $155,000 if the evidence had proved a loss of earning capacity.

6. The plaintiff appealed to the Full Court of the Supreme Court of South Australia. The only ground of appeal was the trial judge's refusal to award any damages for lost earning capacity. Millhouse J., with whose reasons Mohr and Bollen JJ. agreed, identified the issue on the appeal as "whether the appellant retired because of the injuries caused in the accident or for other reasons". His Honour held that the trial judge's decision on this point was correct. He said:

"Had the appellant continued in his position until December

1992 as he apparently could have done then he would have

suffered no loss of earning capacity. The decision to

retire early was his and not that of the University: the

University merely accepted the appellant's application for

early retirement: it did not initiate the retirement."

7. In Australia, a plaintiff is compensated for loss of earning capacity, not loss of earnings ((14) See Luntz, Assessment of Damages for Personal Injury and Death, 3rd ed. (1990) at 224 esp. fn.3.). In practice, there is usually little difference in result irrespective of whether the damages are assessed by reference to loss of earning capacity or by reference to loss of earnings. That is because "an injured plaintiff recovers not merely because his earning capacity has been diminished but because the diminution of his earning capacity is or may be productive of financial loss" ((15) Graham (1961) 106 CLR at 347). Nevertheless, there is a difference between the two approaches, and the loss of earning capacity principle more accurately compensates a plaintiff for the effect of an accident on the plaintiff's ability to earn income. Earning capacity is an intangible asset. Its value depends on what it is capable of producing. Earnings are evidence of the value of earning capacity but they are not synonymous with its value. When loss of earnings rather than loss of capacity to earn is the criterion, the natural tendency is to compare the plaintiff's pre-accident and post-accident earnings. This sometimes means that no attention is paid to that part of the plaintiff's capacity to earn that was not exploited before the accident. Further, there is a tendency to assume that if pre-accident and post-accident incomes are comparable, no loss has occurred.

8. In this case, the learned judges in the Supreme Court wrongly applied a loss of earnings test. Thus, Debelle J. at first instance said:

"the plaintiff has not discharged the onus of proof that he

lost his earning capacity for four and one half years by

reason of injuries sustained in the accident".

"Had the appellant continued in his position until December

1992 as he apparently could have done then he would have

suffered no loss of earning capacity."

Both judges speak of loss of earning capacity, but they are equating that concept with loss of earnings. By equating the two concepts, the learned judges fell into error.

9. The reason that Debelle J. held that the plaintiff's earning capacity was not diminished was that his Honour concluded that the plaintiff could have continued as head of the Discipline of Philosophy and that his main reason for retiring was not connected with his accident-related injuries. With respect, I think that his Honour erred in approaching the question in that way. Whether there had been an impairment of the plaintiff's earning capacity was a matter conceptually distinct from the plaintiff's reasons for retiring. The correct question was whether, as a result of the accident, the plaintiff had been rendered less capable of earning income. In answering that question, his Honour was required to look at the plaintiff's capacity for work beyond the particular employment in which he was engaged at the time of the accident (16) Mann v. Ellbourn (1974) 8 SASR 298). That the plaintiff was able to retain his position as Professor of Philosophy was not conclusive proof that he had suffered no loss of earning capacity.

10. In my opinion, the plaintiff proved that his earning capacity had been diminished as a consequence of the defendant's negligence. The trial judge found that, at the time of his retirement, the plaintiff was still suffering from chronic, and sometimes quite intense, pain which caused him to wake at least once on most nights. His Honour found that this left the plaintiff feeling tired and without the intellectual energy that the plaintiff believed was necessary to properly discharge his duties as Professor and head of the Discipline of Philosophy at Flinders University. In a letter which the plaintiff wrote to the University in May 1987 seeking retirement on the ground of invalidity, he stated that his capacity for work had been reduced, and that because of a lack of energy he was no longer able to combine research and teaching. In the letter, he said that his non-teaching periods throughout the year were "mostly consumed by teaching - either preparation or recovery". In evidence, the plaintiff stated that at the time of his retirement he had reached the point where, because of his pain and fatigue, he could no longer continue "to be reliably present" at the University and would have to reduce his workload. These statements of the plaintiff were not rejected by his Honour. Nor is there anything in the evidence or in his Honour's judgment that provides any ground for rejecting them. The statements were supported by evidence that his peers were finding his lecturing less satisfactory than it had been before the accident.

11. Even if the plaintiff was capable of retaining his position as head of the Discipline of Philosophy until he reached the normal retiring age of 65 years, he still had a reduced capacity to do his work. The evidence showed that it was more difficult for him to do his work after the accident while in a state of constant pain and fatigue than it was before the accident. If for any reason he left the University, he would almost certainly have found it difficult to secure a comparable position elsewhere.

12. It follows that the Supreme Court erred in holding that the plaintiff had not suffered any loss of earning capacity.

13. Damages can be awarded for loss of earning capacity, however, only to the extent that that loss produces or might produce financial loss ((17) Graham (1961) 106 CLR at 347). Debelle J. purported to apply this principle. In this Court, the plaintiff contended that his Honour erred in holding that, in order to recover damages for loss of earning capacity, the plaintiff had to prove that any diminution in his earning capacity was or may have been productive of financial loss. The plaintiff argued that such an approach is inconsistent with the decisions in Griffiths v. Kerkemeyer ((18) [1977] HCA 45; (1977) 139 CLR 161) and Van Gervan v. Fenton ((19) [1992] HCA 54; (1992) 175 CLR 327). In Griffiths, the issue was whether a disabled plaintiff could recover compensation in respect of caring services required as a result of his accident-related injuries, even though those services were being gratuitously rendered by his fiancee and family. Gibbs, Stephen and Mason JJ. held that the plaintiff could recover for those services. In Van Gervan ((20) ibid. at 333), Mason CJ, Toohey J. and I (with whom Brennan and Gaudron JJ. agreed in principle) held that, in a Griffiths claim, the plaintiff is compensated for the need for services and does not have to show that the need "is or may be productive of financial loss".

14. In Beck v. Farrelly ((21) (1975) 13 SASR 17 at 23), Bray CJ saw a possible inconsistency between Graham and the principle that was subsequently laid down in Griffiths. However, I do not think that there is an inconsistency. Graham formulates a principle for compensation for a loss. Griffiths and Van Gervan formulate a principle for compensation for a need. Earning capacity is an intangible asset that has a value only to the extent that it is or could be exploited financially. Consequently, Graham held that no compensation is payable for loss of earning capacity unless the loss is or may be productive of financial loss. But the value of the need for caring services is quantifiable whether or not the plaintiff has suffered financially as a result of the need. As a matter of policy, the wrongdoer is required to compensate the plaintiff in a Griffiths v. Kerkemeyer claim even if the services have been or will be performed without payment for them by the plaintiff.

15. In Mann v. Ellbourn ((22) (1974) 8 SASR 298) the Full Court of South Australia preferred an approach that compensates the plaintiff to the extent that the loss of earning capacity has resulted in a loss of probable earnings and the loss of a chance to exploit any residual capacity that would probably not have been exploited even if the accident had not occurred. I see nothing wrong with that approach in most cases. It gives effect to the fundamental principle underlying the assessment of damages that a person is only compensated for what he or she has actually lost. But, as the judgments of the Supreme Court in this case show, it is an approach that can mislead a court in a case where the plaintiff continues in employment. It is always necessary to bear in mind as Bright J. said in Mann ((23) ibid. at 307) that "one first of all determines that there has been a loss of capacity, and then having regard to the established facts of the past and the probabilities of the future one determines the damages that flow from the loss of that capacity".

Has the plaintiff suffered actual or potential financial loss as the result of his reduced earning capacity?

16. If the plaintiff had stayed at the University he would have suffered no financial loss and so the impairment of his earning capacity would not have been productive of financial loss in respect of the period up to his ultimate retirement at age 65. But once he left the University, financial loss resulted. He no longer had his salary, and, because of his reduced earning capacity, he was not able to otherwise earn what he could have earned before the accident. It is at this point that the question of the plaintiff's motivation for his early retirement is relevant. If the plaintiff's decision to retire early was caused by the legacy of the injuries which he received in the accident, then his impaired earning capacity has been productive of financial loss, and, subject to any question of mitigation of damages, he would be entitled to recover substantial damages for impairment of earning capacity up to age 65. On the other hand, if the plaintiff's decision to retire early was not causally connected to his injuries then, in respect of the period up to age 65, the plaintiff's financial loss was not caused by the defendant's negligence but by his voluntary act in retiring for reasons unconnected with his injuries. In that case the plaintiff would be entitled to nothing more than damages for the loss of the chance that, but for the accident, he might have exploited his earning capacity in some way other than as a professor of philosophy. So, the question whether the impairment of the plaintiff's earning capacity has been or may be productive of financial loss involves a question of causation. Was the negligence of the defendant, which resulted in the plaintiff's injuries and consequential loss of earning capacity, a cause of the financial loss that the plaintiff has suffered?

17. This Court has recently considered the nature of causation in March v. Stramare (E and M.H.) Pty. Ltd. ((24) [1991] HCA 12; (1991) 171 CLR 506) and Bennett v. Minister of Community Welfare ((25) [1992] HCA 27; (1992) 176 CLR 408). In those cases the Court held that, in order to determine whether something was the cause of an occurrence, the tribunal of fact must look at the matter in a common sense way. The "but for" test, applied as a negative criterion of causation, has an important role to play but it is not the exclusive test of causation.

18. In the Supreme Court, Debelle J. applied a "dominant cause" test in deciding whether the plaintiff's injuries brought about his retirement. Thus his Honour said:

"While there can be little doubt that the plaintiff was

continuing to suffer from chronic pain, I do not think that

was the pre-eminent reason for the plaintiff choosing to

take advantage of the University's early retirement scheme."

"The pre-eminent reason why the plaintiff decided to retire

early was to be able to devote as much time as possible to

research and creative philosophy untroubled by the

requirements of University life and in particular the

administrative and teaching duties required of him."

"His reasons for retiring early were mixed but the

pre-eminent reason was his desire to be able to devote as

much time as possible to research and creative philosophy."

The Full Court held that Debelle J.'s finding as to the pre-eminent reason for the plaintiff's early retirement was "entirely justified". But even if it is accepted that that conclusion is correct, the learned trial judge nevertheless erred because he considered only what was the "pre-eminent" or "main" reason for the plaintiff's retirement. If the effects of the plaintiff's injuries were, on a common sense approach, a material cause ((26) Attorney-General v. Gabell (1968) SASR 44) of his decision to retire early, even if not the pre-eminent cause, then the plaintiff's financial loss has resulted from his loss of earning capacity and was connected to the defendant's negligence.

19. The judges in the Supreme Court concluded that the plaintiff's main reason for retiring early was that he wanted to spend his available time on creative research without the added burden of teaching. But that finding is not by itself an answer to the plaintiff's case. One of the plaintiff's problems was that he was no longer able to combine research with teaching as an active academic should. Because of his pain and lack of energy, he found that his available time was taken up by his teaching duties, in preparation, teaching or recovering. Further, his teaching was not as satisfactory as it had been before the accident. He had ideas for research and felt that he was entering the most creative period of his life as a philosopher. Unfortunately, his accident-related complaints made it virtually impossible for him to pursue this research while working. Faced with the choice of neglecting his research work or pursuing it, the plaintiff chose the latter course. If the plaintiff had not been injured, he would have had the capacity to pursue his research work while carrying out his other duties. The plaintiff's early retirement was, therefore, the result of his accident-related complaints and causally connected with the defendant's negligence in the "but for" sense.

20. However, the ultimate question is whether, as a matter of common sense, the financial loss that the plaintiff has suffered was caused by the plaintiff's act in resigning his office rather than by the defendant's negligence. If the plaintiff acted unreasonably in retiring early, then the law will disregard the defendant's negligence as a cause of the plaintiff's financial loss. That negligence and its effects will be treated as background conditions of the plaintiff's loss and, as a matter of common sense, not material causes of the plaintiff's loss.

21. A further question, and one that can be conveniently considered at this point, is whether, having regard to the loss of earning capacity, the plaintiff failed to mitigate his financial loss by not continuing in employment which would have continued to give him the earnings which he was receiving before the accident. In Adams v. Ascot Iron Foundry Pty. Ltd. ((27) (1968) 72 SR(NSW) 120 at 132-133), Sugerman JA correctly pointed out that, where a question arises as to whether a plaintiff could have obtained employment that was within his post-accident capacity, the question is not really one of mitigation of damages. The plaintiff must prove that such employment is beyond his or her capacity "as part of the general burden which lies upon him of proving the extent of the damage he has suffered by reason of the injury". But here the plaintiff has proved a general impairment of earning capacity and the exact degree of the impairment is not a matter for this Court to decide. Accordingly, the question that then arises is whether the plaintiff failed in his duty to take steps that could have avoided or reduced the financial loss which he claims flowed from that loss of earning capacity.

22. It does not seem to me to matter in this case whether the plaintiff's act of early retirement is attacked as a voluntary act that was a novus actus interveniens or as a failure to mitigate loss. His retirement would be a novus actus interveniens or a failure to mitigate loss only if it was unreasonable in all the circumstances. If the retirement is treated as going to mitigation, the onus is on the defendant to prove that the plaintiff failed to mitigate his loss ((28) Watts v. Rake [1960] HCA 58; (1960) 108 CLR 158 at 159). If it is treated as going to causation, the onus is also on the defendant. In Adams ((29) (1968) 72 SR(NSW) at 139), Walsh JA said, correctly in my opinion, that, if a person has the post-accident capacity to do a particular job but a question as to the reasonableness of a refusal to do it arises, the onus is on the defendant to show that the refusal was unreasonable. Walsh JA gave as an example the case where the plaintiff is offered work at a remote place. Similarly, where a plaintiff with impaired earning capacity resigns from a position that he or she is capable of retaining and sustains a financial loss, the onus is on the defendant to prove that the resignation was unreasonable. Whatever approach is followed in this case, therefore, the onus is on the defendant to prove that the plaintiff acted unreasonably in retiring early.

23. In my opinion, it was not unreasonable for the plaintiff to retire early. Admittedly, the plaintiff's security of tenure would have made it difficult for the University to dismiss him even if his performance was below that which could reasonably be expected of a person in his position. Moreover, there was no evidence that the University was dissatisfied with his general performance as a member of staff, although the plaintiff's colleague, Dr O'Hair, expressed the opinion that the plaintiff was working less effectively and that his lecturing was less satisfactory than previously. The plaintiff himself said:

"I could still have been occupying a desk at whatever the

going rate for professors may be. I think I would

increasingly have had to endure the impatience and scorn of

my immediate colleagues but I think that I could have gone

on fronting up and occupied a desk as a tenured professor.

It would have been very difficult for the University to have

got rid of me had I done nothing whatsoever."

But to concentrate on these matters is to miss the vital point. The plaintiff's complaints of pain and fatigue, his decreasing confidence in his own abilities, his belief that he was no longer teaching as well as he was before the accident and his inability to find time for research combine to make a strong case for concluding that his early retirement was not unreasonable. A defendant cannot reasonably require a plaintiff to remain in employment for the purpose of reducing the damages that the defendant would otherwise have to pay if to do so would interfere with the plaintiff's reasonable enjoyment of life. The doctrine of mitigation of loss was not intended to turn injured plaintiffs into economic slaves.

24. Before the accident the plaintiff could have pursued his research and writing interests as well as his teaching duties. By reason of the accident he was unable to do so. He chose to take advantage of the early retirement scheme to pursue his research and writing. As a result, he is financially worse off although, by reason of his pension and early lump sum payments, not to the extent of his annual salary. I see nothing unreasonable in his early retirement even when the defendant's interests are taken into account. Moreover, the defendant is entitled to have credited against the loss of salary any benefit to the plaintiff arising from the accelerated payments of his pension and lump sum entitlements.

25. No doubt it is true that the plaintiff's act of early retirement created a financial loss where none had been present or was likely to arise. But if the defendant's negligence caused the conditions which produced the loss, that negligence is a material cause of the financial loss unless the plaintiff's act of retirement is to be treated as "breaking the causal chain" or was a breach of his duty to mitigate his loss. Because the act of retirement was not unreasonable and the reasons for it were the result of his injuries, the plaintiff's loss was causally related to the defendant's negligence. When the matter is looked at as one of mitigation of damage, the plaintiff's act of early retirement was not unreasonable and did not result in a failure to mitigate his damage.

26. In my opinion the plaintiff was entitled to damages for his loss of earning capacity up to the age of 65. The starting point for calculating that loss is four and a half years loss of salary. In case he was wrong in concluding that the plaintiff had suffered no loss of earning capacity, Debelle J. performed the calculation. After taking account of the plaintiff's lump sum retirement payout and his pension entitlements, his Honour concluded:

"In the result, the plaintiff is entitled to $29,980 for

past economic loss, $82,407 for future economic loss and

$51,944 for the loss of the higher pension. If these

amounts are awarded, there should be a deduction from the

claim for past and future (non-) economic loss for the

amount included in each to represent compensation for not

being able to pursue his academic interests or to discharge

his academic duties to his complete satisfaction. An amount

of $5,000 should be deducted from each of the awards for

past and future non-economic loss."

27. His Honour's finding may be viewed as the prima facie measure of the plaintiff's economic loss for the period up to age 65. It is subject to any finding which the court assessing the plaintiff's damages makes in relation to the question of what alternative forms of employment, if any, were reasonably available to the plaintiff after his early retirement.

28. A further question arises as to whether the Supreme Court erred in failing to award damages for loss of earning capacity for the period beyond age 65. No mention of this issue is made in the judgments of the Supreme Court. There is no reason why the plaintiff's impairment of earning capacity should be treated as ceasing at age 65. However, the financial loss which flows from that impairment in respect of the period beyond age 65 is of a different measure from the loss for the period up to age 65. It will be necessary to look at all the circumstances, including the intentions of the plaintiff, to determine what, if any, post-retirement remunerative activities the plaintiff would probably have engaged in if the accident had not occurred and for how long he would have continued engaging in those activities. Once that has been done, the court must determine the extent to which the plaintiff's capacity to engage in these activities has been impaired and the loss which flows therefrom.

29. The appeal should be allowed. The matter should be remitted to the Supreme Court for assessment of damages for loss of earning capacity.