Chappel v Hart

GAUDRON J. The facts may be simply stated. Mrs Hart underwent surgery at the hands of Dr Chappel without warning as to the possible consequences should her oesophagus be perforated and infection set in. That is what happened and, in consequence, Mrs Hart suffered damage to her laryngeal nerves, paralysis of her right vocal cord and voice loss.

The condition for which Dr Chappel operated on Mrs Hart is one which is relentlessly progressive. Thus, Mrs Hart would inevitably have required surgery of the kind performed. And the surgery would have been subject to the risk which eventuated - although not necessarily in the same degree - no matter when or by whom it was undertaken. Had Mrs Hart been aware of that risk, she would not have had surgery when she did. And she would have taken steps to have it performed by "the most experienced [surgeon] with a record and a reputation in the field."

Mrs Hart commenced proceedings against Dr Chappel in the Supreme Court of New South Wales, claiming damages for the injuries which she sustained. She recovered a verdict in the sum of $172,500.61. Included in the verdict was an amount of $30,000 for general damages. Dr Chappel appealed to the Court of Appeal of the Supreme Court of New South Wales and Mrs Hart cross-appealed, complaining that the amount awarded for general damages was inadequate. Both the appeal and cross-appeal were dismissed. Dr Chappel now appeals to this Court.

The primary contention made on behalf of Dr Chappel is that there was no causal connection between his failure to give adequate warning of the risks involved in the surgery and the damage suffered by Mrs Hart. The contention was made in a context in which it is clear that the surgery was performed with skill and care and the infection which set in and led to the injuries which Mrs Hart sustained was a random event which might occur no matter when or by whom the surgery was performed. It was put that, as surgery was inevitable and carried the risk which, in fact, eventuated, "[t]here was no loss of any 'real and valuable chance', nor ... any substantial prospects of the risk being diminished or avoided" . Alternatively, it was put that the damage sustained by Mrs Hart resulted from the random risk which, in fact, eventuated and her "voluntary willingness to undertake that risk".

The primary argument for Dr Chappel proceeds on the basis that the damage sustained by Mrs Hart was not physical injury but the loss of a chance to have surgery performed by somebody else at some other time. And as the risk which eventuated was ever present, no matter when or by whom the surgery might be performed, Mrs Hart did not, according to the argument, lose a chance of any value. Although the nature of the risk was the same, the evidence does not bear out the proposition that the degree of that risk was the same regardless of the experience of the surgeon involved. That issue can, for the moment, be put to one side, because, clearly, the damage sustained by Mrs Hart was not the loss of a chance - valuable or otherwise - but the physical injury which she, in fact, sustained.

The argument that the damage sustained by Mrs Hart was simply the loss of a chance must be considered in a context concerned with the assignment of legal responsibility. In that context, philosophical and scientific notions are put aside and causation is approached as a question of fact to be answered "by applying common sense to the facts of [the] particular case." That is so both for the question whether a particular act or omission caused any damage at all and for the question whether some particular damage resulted from the act or omission in question .

Questions of causation are not answered in a legal vacuum. Rather, they are answered in the legal framework in which they arise. For present purposes, that framework is the law of negligence. And in that framework, it is important to bear in mind that that body of law operates, if it operates at all, to assign a duty to take reasonable steps to prevent a foreseeable risk of harm of the kind in issue .

It was not disputed in this Court that Dr Chappel was under a duty to inform Mrs Hart of the possible consequences in the event of the perforation of her oesophagus and subsequent infection, including the possibility of damage to her voice. The duty was called into existence because of the foreseeability of that very risk . The duty was not performed and the risk eventuated. Subject to a further question in the case of a duty to provide information, that is often the beginning and the end of the inquiry whether breach of duty materially caused or contributed to the harm suffered. As Dixon J pointed out in Betts v Whittingslowe , albeit in relation to a statutory duty, "breach of duty coupled with an accident of the kind that might thereby be caused is enough to justify an inference, in the absence of any sufficient reason to the contrary, that in fact the accident did occur owing to the act or omission amounting to the breach" .

Where there is a duty to inform it is, of course, necessary for a plaintiff to give evidence as to what would or would not have happened if the information in question had been provided . If that evidence is to the effect that the injured person would have acted to avoid or minimise the risk of injury, it is to apply sophistry rather than common sense to say that, although the risk of physical injury which came about called the duty of care into existence, breach of that duty did not cause or contribute to that injury, but simply resulted in the loss of an opportunity to pursue a different course of action.

The matter can be put another way. If the foreseeable risk to Mrs Hart was the loss of an opportunity to undergo surgery at the hands of a more experienced surgeon, the duty would have been a duty to inform her that there were more experienced surgeons practising in the field. Because the risk was a risk of physical injury, the duty was to inform her of that risk. And that particular duty was imposed because, in point of legal principle, it was sufficient, in the ordinary course of events, to avert the risk of physical injury which called it into existence . And the physical injury having occurred, breach of the duty is treated as materially causing or contributing to that injury unless there is "sufficient reason to the contrary" .

The second argument with respect to causation is that there is "sufficient reason to the contrary" to preclude a finding that Dr Chappel's failure to inform Mrs Hart of the risks involved was causally related to the injuries which she sustained. More precisely, it was argued that, even if he had adequately informed her of those risks, it would not have averted the harm suffered. There are two aspects to that argument. The first is that, as surgery was inevitable and the risk which eventuated was inherent in that surgery, Mrs Hart did not, in fact, suffer any damage. The second aspect asserts that the harm resulted from the "random risk" of infection, which, in fact, eventuated, and Mrs Hart's "voluntary willingness to undertake that risk".

The first aspect of the argument must be rejected. It assumes that the degree of risk - as distinct from the nature of the risk - was the same regardless of the experience of the surgeon concerned. That is a matter to which it will be necessary to return. For the moment, however, it can be put to one side. There is a more fundamental flaw. The argument proceeds on the erroneous footing that the damage sustained by Mrs Hart was simply exposure to risk, not the harm which eventuated. And to say that Mrs Hart would inevitably have been exposed to risk of the harm which she suffered is not to say that she would inevitably have suffered that harm.

The second aspect of the argument, which asserts that the harm suffered by Mrs Hart resulted from the "random risk" of infection which eventuated and her "voluntary willingness to undertake that risk", must also be rejected. It may be that, at some stage, Mrs Hart would have voluntarily undertaken whatever risk was involved in the surgery then necessary for her condition. However, it cannot be said that that or any other risk was voluntarily undertaken when Dr Chappel operated but nothing presently turns on that point. The second aspect of the argument must be rejected because it treats the infection which occurred as a supervening event breaking the chain of causation which would otherwise begin with Dr Chappel's failure to inform Mrs Hart of the possible consequences in the event of perforation and subsequent infection. It is contrary to common sense to treat part of the very risk which called the duty into existence as a supervening event breaking the chain of causation beginning with the breach of that duty.

The question whether the infection which set in following perforation of Mrs Hart's oesophagus broke the chain of causation can also be answered by asking what would or would not have happened if Dr Chappel had provided her with adequate information as to the risk involved . If he had, Mrs Hart would not then have undergone surgery and would not then have suffered the injuries which she did or their consequences. Thus, Dr Chappel's "breach was 'still operating', or, continued to be causally significant when [those injuries were sustained]."

The arguments advanced on behalf of Dr Chappel with respect to causation cannot succeed. Accordingly, it is necessary to turn to the question of damages. As already indicated, damages fell to be assessed, as the courts below recognised, on the basis that Mrs Hart suffered physical injury, not merely the loss of a chance to undergo surgery at the hands of some other surgeon at some other time. However, it was argued for Dr Chappel that, even on that basis, Mrs Hart suffered no damage at all, or, at most, only nominal damage.

It is well settled that an award of damages must take account of the probability that some or all of the damage suffered by the plaintiff would have occurred in any event . In this context, it was put on behalf of Dr Chappel that, "[e]ven with the benefit of a more experienced surgeon, the increased chance of avoiding the injury by reason of the surgeon's greater experience and skill ... would be minimal with the result that there was no damage; or alternatively the damages are nominal". That argument would also seem to be infected with the notion that the damage suffered by Mrs Hart was the loss of a chance rather than the physical injury which she sustained. Whether or not that is so, the argument is premised on an assumption which is made possible only by an ambiguity inherent in the proposition that surgery would involve the very risk which, in fact, eventuated, no matter when or by whom it was performed.

It is not in doubt that a risk of perforation and infection was and is inherent in surgery of the kind performed on Mrs Hart. In that sense, the risk of injury was the same, no matter when or by whom the surgery was performed. However, that is not to say that the likelihood of that risk eventuating was the same. This was recognised by Donovan AJ, at first instance, his Honour stating that "[t]here [was] no evidence that the risk in the sense of its being likely to occur as it did would be the same". Moreover, Professor Benjamin gave evidence from which it might be inferred that the risk of perforation, without which the injury sustained by Mrs Hart could not have occurred, diminished with the skill and experience of the surgeon concerned. And that inference was drawn by the Court of Appeal.

In the Court of Appeal, Handley JA (with whom Mahoney P and Cohen AJA agreed on this point) found that "[w]hile perforations could occur ... without negligence, superior skill and experience could reduce [that] risk". His Honour added, that, on the evidence, Mrs Hart was likely to "have retained the best and most experienced surgeon available" had she been fully informed of the risks involved and concluded that "the risk ... in the actual and hypothetical situations was not the same". That conclusion was clearly open.

Once it is accepted, as in my view it must be, that the risk of injury would have been less if, as Mrs Hart deposed, she had retained the services of the most experienced surgeon in the field, the argument that, at best, Mrs Hart was entitled to nominal damages must be rejected. Rather, Mrs Hart is entitled to damages for the injuries suffered. In the calculation of those damages, however, the question arises whether there was a probability that Mrs Hart would have suffered harm of the kind that eventuated no matter when or by whom surgery was performed. If that was a probability, allowance should have been made for it . However, neither the trial judge nor the Court of Appeal adverted to the question.

The evidence was that the harm suffered by Mrs Hart is extremely rare and cannot occur unless the oesophagus is perforated and infection sets in. The risk was described in evidence as "random". Apparently, no surgery of the kind performed on Mrs Hart can be described as completely free of the risk of harm of the kind that, in fact, eventuated. However, the uncontroverted evidence that it is both rare and random precludes the risk being described as other than speculative. That being so, there is no basis for a finding that it was, in any degree, probable that Mrs Hart would, in any event, have suffered harm of the kind she in fact suffered. There is, thus, no basis for any reduction of the damages awarded at first instance.

The appeal should be dismissed with costs.

McHUGH J. The question in this appeal is whether a doctor who performed an operation with reasonable care is nevertheless liable for an accidental injury occurring in the course of the operation. The question has to be determined in the context that the doctor, in breach of his duty, failed to warn his patient that such an injury could occur and that the patient, if warned, would have had the operation carried out by "the most experienced person with a record and a reputation in the field".

Proof of a cause of action in negligence or contract requires the plaintiff to prove that the breach of duty by the defendant caused the particular damage that the plaintiff suffered. In civil cases, causation theory operates on the hypothesis that the defendant has breached a duty owed to the plaintiff and that the plaintiff has suffered injury; but causation theory insists that the plaintiff prove that the injury is relevantly connected to the breach of duty. The existence of the relevant causal connection is determined according to common sense ideas and not according to philosophical or scientific theories of causation . The reason for this distinction was pointed out by Mason CJ in March v Stramare (E & MH) Pty Ltd : "In philosophy and science, the concept of causation has been developed in the context of explaining phenomena by reference to the relationship between conditions and occurrences. In law, on the other hand, problems of causation arise in the context of ascertaining or apportioning legal responsibility for a given occurrence."

In March this Court specifically rejected the "but for" test as the exclusive test of factual causation. Instead the Court preferred the same common sense view of causation which it had expressed in its decision in Fitzgerald v Penn . There, the Court said that the question is to be determined by asking "whether a particular act or omission ... can fairly and properly be considered a cause of the accident" . As a natural consequence of the rejection of the "but for" test as the sole determinant of causation, the Court has refused to regard the concept of remoteness of damage as the appropriate mechanism for determining the extent to which policy considerations should limit the consequences of causation-in-fact . Consequently, value judgments and policy as well as our "experience of the 'constant conjunction' or 'regular sequence' of pairs of events in nature" are regarded as central to the common law's conception of causation.

The rejection of the "but for" test as the sole determinant of causation means that the plaintiff in this case cannot succeed merely because she would not have suffered injury but for the defendant's failure to warn her of the risk of injury. However, his failure to warn her of the risk was one of the events that in combination with others led to the perforation of her oesophagus and damage to the right recurrent laryngeal nerve. Without that failure, the injury would not have occurred when it did and, statistically, the chance of it occurring during an operation on another occasion was very small. Moreover, that failure was the very breach of duty which the plaintiff alleges caused her injury. The defendant's failure to warn, therefore, must be regarded as a cause of the plaintiff's injury unless either common sense or legal policy requires the conclusion that, for the purposes of this action, the failure is not to be regarded as a cause of the plaintiff's injury.

Underlying the rejection of the "but for" test as the determinant of legal causation is the instinctive belief that a person should not be liable for every wrongful act or omission which is a necessary condition of the occurrence of the injury that befell the plaintiff. As Mason CJ emphasised in March , causation for legal purposes is concerned with allocating responsibility for harm or damage that has occurred. So the mere fact that injury would not have occurred but for the defendant's act or omission is often not enough to establish a causal connection for legal purposes. Thus, in Leask Timber and Hardware Pty Ltd v Thorne , members of this Court accepted that the driving of a crane by an uncertificated driver was not causally related to the death of the plaintiff's husband, notwithstanding that driving a crane without a certificate was a breach of the law and that the death would not have occurred but for that breach. Windeyer J said : "Possession of a certificate means that the driver has satisfied an inspector that he can drive a crane competently, and is a trustworthy person. If, however, he fails to exercise the competence he has and drives a crane improperly, unskilfully and negligently, it will not avail him or his employer that an inspector had certified that he was capable of doing so properly and skilfully; nor is it material that an inspector thought he was trustworthy if trust in him should prove misplaced. On the other hand, a person might have skill and competence but no certificate. If he drives a crane carefully, skilfully and competently then he is not liable in negligence for the consequences of an accident that occurs without fault on his part. That is how the matter would stand in an action for negligence." Similarly, in The Empire Jamaica Willmer J held that the act of the owners of a ship in sending it to sea with a master who had no certificate, contrary to a local Ordinance, was not a legal cause of a collision occurring on the voyage, notwithstanding that the master was guilty of negligent navigating.

Before the defendant will be held responsible for the plaintiff's injury, the plaintiff must prove that the defendant's conduct materially contributed to the plaintiff suffering that injury . In the absence of a statute or undertaking to the contrary, therefore, it would seem logical to hold a person causally liable for a wrongful act or omission only when it increases the risk of injury to another person. If a wrongful act or omission results in an increased risk of injury to the plaintiff and that risk eventuates, the defendant's conduct has materially contributed to the injury that the plaintiff suffers whether or not other factors also contributed to that injury occurring. If, however, the defendant's conduct does not increase the risk of injury to the plaintiff, the defendant cannot be said to have materially contributed to the injury suffered by the plaintiff. That being so, whether the claim is in contract or tort, the fact that the risk eventuated at a particular time or place by reason of the conduct of the defendant does not itself materially contribute to the plaintiff's injury unless the fact of that particular time or place increased the risk of the injury occurring.

In principle, therefore, if the act or omission of the defendant has done no more than expose the plaintiff to a class of risk to which the plaintiff would have been exposed irrespective of the defendant's act or omission, the law of torts should not require the defendant to pay damages. Similarly, if the defendant has done no more than expose the plaintiff to a risk for which the defendant has not undertaken responsibility and to which the plaintiff was always exposed, the law of contract should not require the defendant to pay damages for injury arising from that risk even if it follows upon a breach of contract. No principle of the law of contract or tort or of risk allocation requires the defendant to be liable for those risks of an activity or course of conduct that cannot be avoided or reduced by the exercise of reasonable care unless statute, contract or a duty otherwise imposed by law has made the defendant responsible for those risks.

For these reasons, in Carslogie Steamship Co Ltd v Royal Norwegian Government , where a vessel was delayed so that damage caused by the defendant's negligence could be repaired, the House of Lords had no difficulty in concluding that further damage to the vessel as the result of a severe storm after it resumed its voyage was not causally connected with that negligence. The House so concluded, notwithstanding that the further damage probably would not have occurred but for the delay. No doubt the House would have reached a different conclusion if the delay had increased the risk that the vessel might suffer damage from severe storms. Increased risk as the result of breach of duty was the reason that, in Monarch Steamship Co Ltd v Karlshamns Oljefabriker (A/B) , the House of Lords held the defendant liable for the cost of transhipment arising from the outbreak of war. The House held that the defendant's breach of duty had resulted in delay which had increased the chance that the cargo would have to be delivered after the outbreak of war.

Cases such as Carslogie and Monarch were concerned with damage following negligent acts. But logically the same principles must apply to the wrongful omissions as well as the wrongful acts of the defendant. Thus, if the defendant negligently fails to warn the plaintiff that a particular route is subject to landslides, no causal connection will exist between the failure to warn and subsequent injury from a landslide if every other available route carried the same degree of risk of injury from a landslide. In such a case, the injury suffered is simply an inherent risk in the course of action pursued by the plaintiff. Although the negligence of the defendant has resulted in the plaintiff being in the place where and at the time when the landslide occurred, that negligence is to be regarded as merely one of the set of conditions that combined to produce the injury. Because the negligent failure of the defendant to give a warning did not increase the risk of injury to the plaintiff, the defendant should not incur liability for the plaintiff's injury.

On the other hand, if there were alternative routes involving a lesser risk of landslide and the plaintiff would probably have taken one of them, if given a warning, the defendant's failure to warn would be causally connected with the plaintiff's injury. That is because the failure to warn deflected the plaintiff from taking a safer course and increased the chance that he or she would suffer injury. By doing so, the defendant has materially contributed to the occurrence of that injury. The case is a fortiori if the plaintiff, on being warned, would have abandoned the journey.

Furthermore, a defendant is not causally liable, and therefore legally responsible, for wrongful acts or omissions if those acts or omissions would not have caused the plaintiff to alter his or her course of action. Australian law has adopted a subjective theory of causation in determining whether the failure to warn would have avoided the injury suffered . The inquiry as to what the plaintiff would have done if warned is necessarily hypothetical. But if the evidence suggests that the acts or omissions of the defendant would have made no difference to the plaintiff's course of action, the defendant has not caused the harm which the plaintiff has suffered.

Moreover, even when the defendant's wrongful act or omission has exposed the plaintiff to a risk to which the plaintiff would not have been exposed but for that act or omission, the correct conclusion may nevertheless be that no causal connection exists between the negligence and the injury suffered. Thus, in Central of Georgia Railway Co v Price , a railway company was held not liable for injury sustained as the result of a lamp exploding in a hotel where the plaintiff had to stay as the result of the company negligently taking her beyond her destination . The risk of such an event occurring in that hotel on that particular night was so insignificant and therefore so abnormal as to be fairly described as a coincidence, rather than an event causally connected to the defendant's negligence .

The foregoing observations lead me to the following conclusions concerning whether a causal connection exists between a defendant's failure to warn of a risk of injury and the subsequent suffering of injury by the plaintiff as a result of the risk eventuating: (1) a causal connection will exist between the failure and the injury if it is probable that the plaintiff would have acted on the warning and desisted from pursuing the type of activity or course of conduct involved ; (2) no causal connection will exist if the plaintiff would have persisted with the same course of action in comparable circumstances even if a warning had been given ; (3) no causal connection will exist if every alternative means of achieving the plaintiff's goal gave rise to an equal or greater probability of the same risk of injury and the plaintiff would probably have attempted to achieve that goal notwithstanding the warning; (4) no causal connection will exist where the plaintiff suffered injury at some other place or some other time unless the change of place or time increased the risk of injury; (5) no causal connection will exist if the eventuation of the risk is so statistically improbable as not to be fairly attributable to the defendant's omission; (6) the onus of proving that the failure to warn was causally connected with the plaintiff's harm lies on the plaintiff. However, once the plaintiff proves that the defendant breached a duty to warn of a risk and that the risk eventuated and caused harm to the plaintiff, the plaintiff has made out a prima facie case of causal connection. An evidentiary onus then rests on the defendant to point to other evidence suggesting that no causal connection exists. Examples of such evidence are: evidence which indicates that the plaintiff would not have acted on the warning because of lack of choice or personal inclination; evidence that no alternative course of action would have eliminated or reduced the risk of injury. Once the defendant points to such evidence, the onus lies on the plaintiff to prove that in all the circumstances a causal connection existed between the failure to warn and the injury suffered by the plaintiff.

Upon the unusual facts of the present case - they are set out in detail in other judgments - the defendant in my opinion can escape liability only if the proper conclusion is that the plaintiff did not prove that the defendant's failure to warn resulted in her consenting to a procedure that involved a higher risk of injury than would have been the case if the procedure had been carried out by another surgeon.

In evidence Professor Benjamin said that any perforation of the oesophagus could result in mediastinitis. However, he said that it was "very rare indeed" for a perforation to "be complicated by what we call mediastinitis". In a report, tendered in evidence, Dr Lewkovitz said: "Perforation of the oesophagus is a recognised but uncommon complication of examination of the oesophagus with rigid endscope as was carried out in this instance. That the oesophagus was indeed perforated may be regarded as a complication rather than a negligent act unless it can be shown that the rigid endscope was introduced into Mrs Hart's throat in a non-conventional manner, or without due care being exercised. From the history this cannot be ascertained."

When Professor Benjamin was asked what was the incidence of perforation of the oesophagus during this kind of procedure, he said that "depending upon the experience and care with which the surgery is done, it could occur as often as one in twenty or thirty or forty operations, but it is usually just an escape of a few bubbles of air and the patient is asymptomatic." The learned trial judge found that even where mediastinitis occurred "the likelihood is that the problems would clear up" and that the risk of damage to the recurrent laryngeal nerve as the result of the mediastinitis was "less".

The outcome of this case in my opinion depends primarily upon the effect of this evidence of Professor Benjamin and Dr Lewkovitz and the above findings of the trial judge. That evidence and those findings must be read, however, with the evidence of the plaintiff, which his Honour accepted, that, if warned of the risk, she would have made further inquiries and "would have wanted the most experienced person with a record and a reputation in the field" to have performed the operation. They must also be read with the evidence that the plaintiff's condition was "relentlessly progressive" and that surgery would provide the "only relief" possible for the condition. On the evidence, the plaintiff would have undergone the procedure in the future even if she had been given a warning. Indeed in June 1985, the plaintiff once again submitted to the procedure even though the procedure performed by the defendant had reduced the severity of her symptoms.

It is clear from the evidence that mediastinitis is not an inevitable result of the perforation of the oesophagus. For practical purposes, the occurrence of mediastinitis is the result of the random chance of bacteria being present in the oesophagus when the perforation occurs. Given the principles of causation to which I have referred, the existence of a causal connection between the occurrence of mediastinitis and the defendant's failure to warn depends upon whether the plaintiff has proved that the failure to warn required her to assume a risk of mediastinitis occurring that was greater than the risk of it occurring if she had been warned. That depends in the first place on whether the effect of Professor Benjamin's evidence is that there are other surgeons who could perform the procedure with less risk of a perforation than the defendant.

I do not think that it is possible to read Professor Benjamin's evidence as asserting that either he or other unidentified surgeons could perform the procedure with greater care or more skill than the defendant ordinarily performed it. In the Court of Appeal, Handley JA read one of Professor Benjamin's answers as meaning that he had performed between one hundred and one hundred and fifty operations without a perforation of the oesophagus. If that interpretation had been correct, it would have provided an evidentiary foundation for the argument, if not the conclusion, that the defendant's failure to warn had denied her the alternative of having the procedure performed with a reduced risk of perforation of her oesophagus. However, the plaintiff did not really dispute that his Honour misunderstood Professor Benjamin's answer and that the effect of the Professor's evidence was merely that he had carried out that number of operations without the onset of mediastinitis. Indeed, it is possible to read one of Professor's Benjamin's answers as indicating that perforations have occurred on a number of occasions when he or a team of surgeons of which he was a member has carried out the procedure .

Nothing in the evidence suggested that there was available to the plaintiff the services of a surgeon of such skill that he or she would never perforate the oesophagus while performing this procedure. Nor did the evidence suggest that either Professor Benjamin or any other surgeon was so superior in skill to the defendant that an operation by that person carried with it a statistically significant lesser risk of perforation than an operation by the defendant. Professor Benjamin was no doubt a pre-eminent surgeon in this field and had performed the operation on many more occasions than the defendant. It is also true that risk of perforation will vary depending upon the degree of care taken on a particular occasion. But the evidence did not suggest, let alone prove, that an operation by the defendant carried with it a statistically significant greater risk of perforation than that of any other qualified surgeon. There is not even a suggestion that the defendant had perforated the oesophagus in any previous operation. The evidence was all one way that perforation of the oesophagus was an inherent risk of the procedure. That is to say, it was an injury that could occur even when reasonable skill and care were exercised. The fact that it happened on this occasion says nothing about whether an operation by the defendant carried with it a statistically significant greater risk of injury.

The plaintiff's claim must fail. This follows from her failure to prove that there was open to her an alternative course of action which would have reduced the inherent chance of a perforation and consequent onset of mediastinitis and damage to the recurrent laryngeal nerve. The highest that her case can be put is that the defendant's failure to warn her resulted in her having the procedure at an earlier date and no doubt at a different place with a different surgeon than would have been the case if the defendant had carried out his duty and warned her. On the evidence, the carrying out of the procedure by the defendant on the day and at the place did not increase the risk of injury involved in the procedure. That being so, the defendant's failure to warn did not materially contribute to the plaintiff's injury. Her claim that a causal connection existed between that failure and her injury must be rejected.

On the view that I take of the case, it is of no relevance that, if she had been warned, another surgeon would have performed the procedure and that the chance of her suffering damage to the laryngeal nerve in that procedure was very remote. Perforation of the oesophagus with consequential mediastinitis and inflammation resulting in damage to the laryngeal nerve is such a rare event that it is close to a certainty that the plaintiff would have avoided mediastinitis and consequential damage to the laryngeal nerve if another surgeon had performed the procedure. Perforation of the oesophagus can and does occur in carrying out the procedure even though the surgeon exercises reasonable skill and care. When it does occur, it will lead to mediastinitis only if bacteria is present in the oesophagus. According to the evidence of Professor Benjamin, it is "very rare" for a perforation to be complicated by mediastinitis. Even then, as the learned trial judge found "the likelihood is that the problems would clear up". It seems almost certain, therefore, that if the plaintiff had been warned and had had the operation performed by another surgeon she would have avoided damage to her laryngeal nerve.

However, it is also close to a certainty that neither mediastinitis nor damage to the laryngeal nerve would have occurred if the defendant had performed the operation on some other day or even at some different hour on that day. He was not as experienced a surgeon as Professor Benjamin but he had performed the operation successfully on previous occasions. If reasonable care is exercised, there is only a remote possibility that damage to a laryngeal nerve resulting from mediastinitis will lead to paralysis of the vocal cords, as happened with the plaintiff, irrespective of which surgeon performs the procedure. Moreover, given the plaintiff's abandonment of any claim that the defendant had performed the operation negligently, he must be taken to have exercised reasonable skill and care on this occasion. His performance on this occasion was differentiated from that of others only by the eventuation of a risk that is inherent in the procedure whoever performs it.

To hold the defendant liable on the basis that if the plaintiff had been given a warning of the risk of mediastinitis occurring she would have avoided that condition is simply to apply the "but for" test, a course which March rejects. If, as the result of the defendant warning the plaintiff about the risk of perforation, the plaintiff had sought out another surgeon who had operated and accidentally perforated the plaintiff's oesophagus with consequent mediastinitis, only the most faithful adherents to the "but for" test would argue that the defendant's warning had caused the perforation and mediastinitis. To so argue would seem an affront to common sense. Similarly, with great respect to the learned judges in the courts below, it seems contrary to common sense to conclude that the defendant's failure to warn caused or materially contributed to him perforating the plaintiff's oesophagus on this occasion. From a common sense point of view, the cause of the perforation and the consequent mediastinitis was the examination of the oesophagus with a rigid endscope, an examination which carried with it an inherent risk of perforation.

The attractiveness of the proposition that the defendant's failure to warn caused or materially contributed to the plaintiff's perforation and mediastinitis derives, I think, from the language in which the proposition is expressed. Authorities on writing recognise that using a noun instead of a verb to express action (nominalisation) and omitting an actor from a sentence are fertile sources of imprecise communication . The use of a nominalisation and the omission of an actor can also conceal reasoning errors. The question: "Did the defendant's failure to warn cause or materially contribute to the perforation of the oesophagus" is more readily answered in the affirmative than the question: "Did the defendant's failure to warn cause or materially contribute to him perforating the defendant's oesophagus?"

The first question uses a noun (perforation) instead of the verb (perforate) and expresses no action. Because the perforation follows the failure to warn and the question identifies no action or actor, that question implicitly suggests a connection between the failure to warn and the perforation. But it is merely a temporal or sequential connection between the omission and the injury. When analysed, therefore, the posing of the first question can be of little, if any, assistance in determining whether the defendant by failing to warn of the risk of injury materially contributed to him perforating the oesophagus of the plaintiff.

The second question focuses on the defendant and makes his actions central to the inquiry. Its very statement suggests a negative answer. His omission to warn had nothing to do with him perforating the oesophagus on that particular day, except as one of many events that combined to place him in the theatre that day operating on the plaintiff. For the purpose of legal causation theory, his omission to warn was no more causative of the perforation than were his medical qualifications, no more causative of the plaintiff's injury than the lack of a crane driving certificate was causative of the deceased's injury in Leask Timber .

It follows that the learned judges of the Supreme Court and the Court of Appeal erred in finding that there was a causal connection between the defendant's failure to warn and the plaintiff's injury.

The plaintiff also sought to rely on an alternative case that she lost the chance of having the procedure performed without a perforation occurring. However, this is not a case concerned with "loss of a chance" as that phrase is understood in the many cases that have come before the courts since Chaplin v Hicks authoritatively decided that a loss of a chance or opportunity was compensable in damages. No part of the relationship between the plaintiff and the defendant involved her being given the opportunity to seek a higher standard of care or better treatment from another surgeon or an opportunity to have the procedure carried out without perforation of the oesophagus . Her relationship with the defendant gave her a legal right to have her condition examined, diagnosed and treated with reasonable care and skill by the defendant and to be informed and advised by him of any material risk inherent in the proposed procedure. But nothing in that relationship required the defendant to provide opportunities of the kind to which I have just referred. The damage that the plaintiff suffered was physical injury, not loss of a chance or opportunity. That being so, her claim stands or falls according to whether the physical injury that she suffered was causally connected for legal purposes with the defendant's failure to warn.

The appeal must be allowed. GUMMOW J. The facts

The appellant, Dr Chappel, is an ear, nose and throat specialist. On 10 June 1983, the respondent, Mrs Hart, underwent surgery at the hands of Dr Chappel for the removal of a pharyngeal pouch in her oesophagus. During that procedure, her oesophagus was perforated and there ensued an infection known as mediastinitis. This was caused by bacteria present in the oesophagus escaping through the perforation into the mediastinum which is part of the chest cavity. While Mrs Hart appears by November 1984 to have recovered from the perforated oesophagus and mediastinitis, the infection damaged the laryngeal nerve and led to a paralysis of the right vocal cord. This affected the performance by Mrs Hart of her duties in a senior position in the New South Wales Department of School Education. In 1985 she was retired from that position on medical grounds.

The surgical procedure was "elective" for the respondent in June 1983, although at a later stage the position would have been reached where it could no longer sensibly be deferred. The evidence did not indicate with any precision when Mrs Hart's condition would have reached that stage.

Mrs Hart sued Dr Chappel in the Supreme Court of New South Wales. She pleaded her action in contract and in the tort of negligence. She did not allege that the operation had been performed negligently. Rather, Mrs Hart alleged that on 20 April 1983 she had consulted Dr Chappel for advice concerning medical problems relating to her throat and that, after the receipt of his advice to undergo a surgical procedure, she engaged Dr Chappel to carry out that procedure. The trial judge made findings to that effect.

Mrs Hart pleaded that her agreement with Dr Chappel contained an implied term that he would warn her of all risks associated with the procedure, that he had failed to warn her of those risks and that he caused or allowed to be caused her injuries. Mrs Hart also pleaded that Dr Chappel had been negligent in advising her in relation to the procedure by failing to warn her of any risks associated with its performance. In particular, she alleged that Dr Chappel, before obtaining her consent to the carrying out of the procedure, had failed to warn her of the risks of sustaining the injuries which she in fact sustained. Mrs Hart further alleged that, in consequence of this negligence and breach of contract, she had sustained a perforated oesophagus and consequent paralysis of the right vocal cord. Finally, she contended that she had incurred out-of-pocket expenses and sustained economic loss by reason of her compulsory retirement in 1985.

The trial was conducted by a judge sitting without a jury. There was a verdict for the respondent and judgment in the sum of $172,500.61. The New South Wales Court of Appeal (Mahoney P, Handley JA and Cohen AJA) dismissed an appeal by Dr Chappel and a cross-appeal by Mrs Hart in which she challenged as inadequate that component of the verdict which was an award of $30,000 for general damages. Dr Chappel appeals to this Court.

This Court decided in Rogers v Whitaker that a medical practitioner has a duty to warn a patient of a material risk inherent in a proposed procedure or treatment and that : "a risk is material if, in the circumstances of the particular case, a reasonable person in the patient's position, if warned of the risk, would be likely to attach significance to it or if the medical practitioner is or should reasonably be aware that the particular patient, if warned of the risk, would be likely to attach significance to it". Mrs Whitaker sued successfully in tort, for negligence. In this Court, there was no challenge to the holding that Mrs Whitaker would not have undergone the surgery had she been advised of the risk of sympathetic ophthalmia. That finding was treated as one going to the establishment of causation .

In the present case, the primary judge held that Dr Chappel had breached the duty to warn of a material risk, which is formulated in the above passage in Rogers v Whitaker and which, in the circumstances of the case, Dr Chappel had owed to his patient. Further, the injury resulting from the sequence of the perforation, the infection (mediastinitis), the damage to the laryngeal nerve and the paralysis of the right vocal cord had been, on the evidence, reasonably foreseeable. In this case, unlike Rogers v Whitaker itself, the outcome of the appeal turns upon questions of causation and the measure of damages. Damage was the gist of her action in negligence. Breach of the contract entitled Mrs Hart to a verdict and at least nominal damages, but recovery beyond that would have required her to establish an element of causation analogous to that in tort .

The trial judge found that Dr Chappel had reassured Mrs Hart that what was proposed was "a common operation" and that, whilst he mentioned the risk of perforation as a recognised complication, he did not warn her of the risk of injury to the laryngeal nerve and the consequent risk of partial or total loss of her voice. His Honour accepted Mrs Hart's evidence that if she had been warned of this risk she would not have undergone surgery when she did. It followed that "but for" the failure of Dr Chappel to warn Mrs Hart she would not have undergone surgery on 10 June 1983 and she would not have sustained the injuries which then ensued from the surgery undergone at that particular time.

However, Dr Chappel challenges the finding against him with respect to causation. In particular, he bases this challenge on the finding by the trial judge that, even if Mrs Hart had elected to defer surgery after receipt of an adequate warning of the risk, sooner or later she would have had the operation. This would have carried the risk of the complication which in fact had come to pass after the surgery performed on 10 June 1983. Mrs Hart's injuries were a random event which could have resulted whenever the surgery was performed. In ground 3A of the Amended Notice of Appeal , Dr Chappel contends that damages should have been assessed "as a loss of a chance rather than as the physical injuries which, in fact, the [r]espondent suffered".

The analysis of the issues in this Court was confused by a failure properly to distinguish those factors which are relevant to the issue of causation and those that are to be considered in the assessment of damages. Causation

In Australia, it is settled by the decision of this Court in March v Stramare (E & MH) Pty Ltd that the legal concept of causation differs from philosophical and scientific notions of causation. Mason CJ said : "In philosophy and science, the concept of causation has been developed in the context of explaining phenomena by reference to the relationship between conditions and occurrences. In law, on the other hand, problems of causation arise in the context of ascertaining or apportioning legal responsibility for a given occurrence." Mason CJ (with whom Toohey J and Gaudron J agreed) also held that, generally speaking, a sufficient causal connection is established if it appears that the plaintiff would not have sustained the injuries complained of had the defendant not been negligent . However, the "but for" test is not a comprehensive and exclusive criterion, and the results which are yielded by its application properly may be tempered by the making of value judgments and the infusion of policy considerations . So, it may be "unjust" to hold a defendant legally responsible for an injury which, though it may be traced back to the wrongful conduct of the defendant, was the immediate result of unreasonable action on the part of the plaintiff .

In Environment Agency (formerly National Rivers Authority) v Empress Car Co (Abertillery) Ltd , the leading judgment in the House of Lords was given by Lord Hoffmann. His Lordship stressed that whilst "the notion of causation should not be overcomplicated", it should not "be oversimplified" . He went on to emphasise that (a) the legal issue is not what caused the result complained of, but did the defendant cause it , and (b) "common sense" answers to questions of causation will differ according to the purpose for which the question is asked and the rule by which responsibility is being attributed . In particular, "one cannot give a common sense answer to a question of causation for the purpose of attributing responsibility under some rule without knowing the purpose and scope of the rule" .

Lord Hoffmann illustrated these points by the following discussion of Stansbie v Troman : "A decorator working alone in a house went out to buy wallpaper and left the front door unlocked. He was held liable for the loss caused by a thief who entered while he was away. For the purpose of attributing liability to the thief (eg in a prosecution for theft) the loss was caused by his deliberate act and no one would have said that it was caused by the door being left open. But for the purpose of attributing liability to the decorator, the loss was caused by his negligence because his duty was to take reasonable care to guard against thieves entering." His Lordship concluded : "Before answering questions about causation, it is therefore first necessary to identify the scope of the relevant rule. This is not a question of common sense fact; it is a question of law. In Stansbie v Troman the law imposed a duty which included having to take precautions against burglars. Therefore breach of that duty caused the loss of the property stolen."

The nature and purpose of a duty with the content established in Rogers v Whitaker concern the right of the patient to know of material risks which are involved in undergoing or forgoing certain treatment. This, in turn, arises from the patient's right to decide for himself or herself whether or not to submit to the treatment in question . That choice "is, in reality, meaningless unless it is made on the basis of relevant information and advice" .

In the present appeal, not only was the damage which Mrs Hart suffered reasonably foreseeable, but the fact that the relevant conjunction of circumstances could occur should have been the subject of any adequate warning and the reason for giving it. It is true that in some cases of a failure to warn by a medical practitioner an application of the "but for" test without qualification could lead to absurd or unjust results. Such would have been the situation if, for example, instead of suffering damage to her laryngeal nerve, Mrs Hart had been injured through the misapplication of anaesthetic. Whilst it would still be open to conclude that, but for Dr Chappel's failure to warn her of the possibility of damage to her voice, she would not have opted for the operation at that time and would not have been injured by the anaesthetic, the law would not conclude that the failure to warn of the risk of injury to the laryngeal nerve caused the injury resulting from the anaesthetic.

The present appeal is significantly different from the situation described. In Mrs Hart's case, the very risk of which she should have been warned materialised. In his written submissions filed by leave after the hearing of the appeal, Dr Chappel conceded that, if the surgery had been performed at a different time, then "[i]n all likelihood" Mrs Hart "would not have suffered the random chance of injury". In addition, the particular risk involved had been the subject of a specific inquiry by Mrs Hart of the medical practitioner who then was engaged by her to perform the surgery. She was a person for whom the potential consequences of damage to her voice were more significant than the "statistical" risk. Those additional factors combined with the satisfaction of the "but for" test were sufficient to establish causation in this case.

Here, the injury to Mrs Hart occurred within an area of foreseeable risk. In the absence of evidence that the breach had no effect or that the injury would have occurred even if Dr Chappel had warned her of the risk of injury to the laryngeal nerve and of the consequent risk of partial or total voice loss, the breach of duty will be taken to have caused the injury .

In those circumstances the task of Dr Chappel was to demonstrate some good reason for denying to Mrs Hart recovery in respect of injuries which she would not have sustained at his hands but for his failure adequately to advise her. Dr Chappel founds his case upon the circumstance that injuries of the nature which were sustained by his patient may be caused without negligent performance of the procedure. He joins to that consideration three matters. The first is the circumstance that sooner or later (though it does not appear whether this would have been before Mrs Hart's retirement in August 1985 or, indeed, at any particular time) Mrs Hart would have been obliged to submit to the procedure. The second is the finding by the trial judge that at some future time Mrs Hart would in fact have done so, even after being made adequately aware of the risk. The third is that this later operation would have carried the same risk of injury. Thus, it was said to follow that Mrs Hart had lost no "real and valuable chance ... of the risk [of injury] being diminished or avoided". In support of that conclusion, reliance was placed upon passages in Sellars v Adelaide Petroleum NL which deal with lost opportunities or chances to acquire benefits. However, as is emphasised later in these reasons, Mrs Hart did not sue to recover the value of an opportunity or chance lost to her by the act or omission of Dr Chappel.

In this way the submissions for Dr Chappel tended to divert attention from the central issue, namely whether there was adequate reason in logic or policy for refusing to regard the "but for" test as the cause of the injuries sustained by Mrs Hart, by the allurement of further cogitation upon the subject of "loss of a chance".

Once the criterion for assessment of the adequacy of causation has been determined as a matter of law, the question whether the plaintiff has suffered some damage and therefore has a complete cause of action in tort is normally established by evidence which satisfies the civil standard of proof . If causation is not established in this way, then the plaintiff will fail and recover nothing .

The difficulties which this standard of proof may present to plaintiffs in certain types of litigation have attracted attention in recent times. In Snell v Farrell , Sopinka J, who gave the judgment of the Supreme Court of Canada, referred with approval to the treatment of the subject by Professor Fleming. That scholar had written : "This traditional approach has come increasingly under challenge in dealing with non-traumatic injuries such as man-made diseases linked to dust, deafness, dermatitis, asbestosis, or linked to chemical products like Thalidomide, DES, and Agent Orange. Another group of cases involves medical procedures depriving patients of a chance of survival or cure. It is often difficult to prove medical causation by 'particularistic' evidence, that is direct, anecdotal, non-statistical evidence from the mouth of witnesses."

The result of the application of the traditional criterion of proof may be to deny plaintiffs any recovery in tort. There has been discussion of alternatives to denial of recovery in obedience to the "more probable than not" civil standard of proof . Writing in 1989, Professor Fleming said of these alternatives : "One is to lower the conventional standard and accept exposure to the risk of injury instead of actual injury as a compensable event. Another is to limit liability in an amount proportionate to the risk created by each individual agent. Both of these modifications have gained reluctant and by no means universal acceptance by Anglo-American courts."

In Snell v Farrell , Sopinka J referred to material suggesting that in the United States the loosening of the criteria for recovery in medical malpractice suits had been followed by the withdrawal of some major insurers from the market . Subsequently, in Laferrière v Lawson , the Supreme Court of Canada held that it had not been proved on the balance of probabilities that the failure in 1971 of the defendant to inform his patient that the growth removed by him was cancerous had caused her death in 1978. The evidence was that the patient's chances of survival would not have been greater had she been informed in 1971 of the diagnosis. The Court also held that the theory of liability for loss of a chance was not to be adopted in such a case .

The present appeal does not involve any consideration of whether such means should be adopted to assist recovery by plaintiffs in certain cases. Mrs Hart did not plead that she contracted for the benefit of a chance of avoiding physical harm or damage. She alleged an obligation to warn her of all risks associated with the procedure, and the failure to discharge that obligation. Nor does Mrs Hart submit in tort that the deprivation of the chance of a full recovery should be accepted as the equivalent of or substitute for her physical injury and damage.

To the contrary, it is Dr Chappel who seeks (in ground 3A of the Amended Notice of Appeal) to intrude considerations of risk and chance with the objective of denying recovery to Mrs Hart. I have set out earlier in these reasons the steps by which Dr Chappel seeks to achieve that result. However, this is not a case in which Mrs Hart seeks damages for the loss of an opportunity or chance to acquire or receive a benefit with a value to be ascertained by reference to the degree of probabilities or possibilities. As is explained in Sellars v Adelaide Petroleum NL , in Australia this generally is what is involved in the "loss of a chance" cases. Similarly, in Athey v Leonati the Supreme Court of Canada observed : "The [loss of chance] doctrine suggests that plaintiffs may be compensated where their only loss is the loss of a chance at a favourable opportunity or of a chance of avoiding a detrimental event." Rather, Mrs Hart claimed damages for the injuries she sustained. To make good her case and to obtain the award of damages she recovered, Mrs Hart was not required to negative the proposition that any later treatment would have been attended with the same or a greater degree of risk.

This is not a case such as Hotson v East Berkshire Area Health Authority . There, the facts precluded the adoption of the plaintiff's hypothesis that he would have escaped disability to his hip joint but for the negligence of the defendant in failing to diagnose a fracture and to treat it promptly.

In the present case, the chain of causation appears from the historical facts found to have intervened between the negligent omission of Dr Chappel and the injuries sustained by Mrs Hart . There was no difficulty in demonstrating what would have happened if Dr Chappel had given Mrs Hart the warning required by Rogers v Whitaker before the surgical procedure on 10 June 1983 in which her oesophagus was perforated, leading to the development of mediastinitis and the paralysis of her right vocal cord. Mrs Hart would not have undergone that procedure at the hands of Dr Chappel. She would have wanted "the most experienced person with a record and reputation in the field", such as Professor Benjamin.

Professor Benjamin, a pre-eminent specialist in throat surgery, was called by Mrs Hart. He was asked questions concerned with the incidence of perforation of the oesophagus, followed by the mediastinum infection and then by injury to the laryngeal nerve. It indicates that, had Mrs Hart undergone the same surgical procedure in other circumstances, the cumulative risks which produced her injuries were so unlikely to recur as to border upon the speculative. In chief, the matter was dealt with as follows: "Q. To your knowledge what was the incidence of perforation of the oesophagus in this type of procedure? A. Well, I think it is higher than most surgeons would recognise. We have done studies simply by taking an x-ray of every patient who has this operation within an hour of the operation and I would think that, depending upon the experience and care with which the surgery is done, it could occur as often as one in twenty or thirty or forty operations, but it is usually just an escape of a few bubbles of air and the patient is asymptomatic. It is very rare indeed for that to then be complicated by what we call mediastinitis. That is a very severe infection. Q. If there is a full perforation, does mediastinitis always follow? A. That depends on what a full perforation is. If there is any perforation mediastinitis can follow. Q. What would be the process, if it be the case, [whereby] the mediastinitis would compromise the laryngeal nerve? A. There you have me guessing again. It is an abscess, pus formation in a smaller or larger quantity. If there is a delicate nerve in the area one must presume it could undergo some form of damage." (emphasis added) In cross-examination there was the following exchange: "Q. Just, finally, you have given evidence as to this complication of perforation being one in twenty or forty; that is, the complication of perforation at all. Of course, that statistic applies to any perforation whatsoever, most of which, as I understand it, don't lead to any complication of any significance at all. Is that right? A. That's correct. Q. Can you give us, likewise, a statistic of this complication arising that does lead to mediastinitis? You say it is very rare. Are you able to translate that in similar terms to the way you have expressed yourself apropos the statistic of one in twenty to forty, or not? A. Could I give my own experience? Q. Yes? A. I believe, not having counted it, that I have performed between one hundred and 150 operations and have not had a patient with that complication. Nevertheless, I think every practising ear nose and throat surgeon , whether he does this operation or not, is aware of the possibility of perforation and mediastinitis that may follow the operation. " (emphasis added)

The reference by Professor Benjamin to experience and care in the particular case underlines the significance of several observations by Gonthier J in delivering the majority judgment in the Supreme Court of Canada in Laferrière v Lawson . His Lordship said that he was not prepared to conclude that "particular medical conditions should be treated for purposes of causation as the equivalent of diffuse elements of pure chance, analogous to the non-specific factors of fate or fortune which influence the outcome of a lottery". He had earlier identified loss of chance cases where the damage can only be understood in probabilistic terms as those where there was no factual context in which to evaluate the likely result other than the realm of "pure statistical chance", so that "the pool of factual evidence regarding the various eventualities in the particular case is dry". Gonthier J concluded : "I can certainly see no reason to extend such an artificial form of analysis to the medical context where faults of omission or commission must be considered alongside other identifiable causal factors in determining that which has produced the particular result in the form of sickness or death. As far as possible, the court must consider the question of responsibility with the particular facts of the case in mind, as they relate concretely to the fault, causation and actual damage alleged in the case."

In the present case, the obtaining of adequate advice as to the risks involved was a central concern of Mrs Hart in seeking and agreeing to undergo the surgical procedure in question. It would, in the circumstances of the case, be unjust to absolve the medical practitioner from legal responsibility for her injuries by allowing decisive weight to hypothetical and problematic considerations of what could have happened to Mrs Hart at the hands of some other practitioner at some unspecified later date and in conditions of great variability. Assessment of damages

Once the liability of Dr Chappel was established in contract and in tort, as was properly done, there may have been a question when assessing Mrs Hart's loss of what, if any, reductions arising from the uncertainty of future events properly were to be taken into account . The principles applicable were laid down in Malec v J C Hutton Pty Ltd . In that case, the plaintiff was entitled to be compensated for the near certainty that, as a result of the defendant's negligence between 1975 and 1977, he would suffer from a psychiatric condition and be unemployable for the rest of his life . However, the majority in the Queensland Full Court had found that it was "likely" that, independently of the defendant's negligence, as a result of the plaintiff's unemployability, he would have developed a similar neurotic condition. This Court held that the Full Court had erred in refusing to award damages for economic loss suffered after May 1982. The case was returned to the Supreme Court to determine if the damages otherwise recoverable should be reduced to provide for the chance that, independently of the negligence of the defendant, the plaintiff would have been placed in a similar position by May 1982. A chance expressed in terms of probability as "say less than 1 per cent" would properly be disregarded as speculative .

In the present case, it would have been for Dr Chappel to show that Mrs Hart's damages were to be reduced to reflect the possibility, being more than a speculation, that independently of his negligence Mrs Hart would have sustained at some later date the injuries of which she complained. That was not the way in which the case for the appellant was presented. Rather, the attempt was to show a lack of causation and to deny any liability. The submissions by Dr Chappel in a large measure attempt to turn speculative matters, which are relevant, if at all, upon the assessment of damages, to account by disrupting the principles governing causation. In this Court, as in the Court of Appeal, Dr Chappel seeks an order setting aside the verdict for Mrs Hart and its replacement by a judgment in his favour.

In any event, by her Notice of Contention Mrs Hart submits that Dr Chappel would have failed in any attempt at trial to obtain a Malec discount and I agree. The evidence of Professor Benjamin, which is set out earlier in these reasons, and the observations in the Supreme Court of Canada to which I have referred, indicate the serious difficulty that would have arisen in this case in passing from the speculative to the ascertainment of a degree of probability. That consideration serves also to emphasise the strength of Mrs Hart's case on causation. Conclusion

The appeal should be dismissed with costs.

KIRBY J. This is yet another appeal concerned with the difficult topic of causation. Causation: a complex and controversial problem

Establishing a causal connection between an alleged wrongdoer's conduct or default and the harm complained of is a pre-condition to the legal liability to pay damages. But, as Professor Dieter Giesen has observed, establishing a causal connection between medical negligence and the damage alleged is often the most difficult task for a plaintiff in medical malpractice litigation (as, indeed, in other negligence actions) . Judges in common law countries can take only the smallest comfort from the fact that determining what caused an injury, for the purposes of legal liability, is also regarded as a most difficult task by the courts of civil law countries . Like courts of the common law, those courts have searched for principles to provide a "filter to eliminate those consequences of the defendant's conduct for which he [or she] should not be held liable" . The search sets one on a path of reasoning which is inescapably "complex, difficult and controversial" . The outcome is a branch of the law which is "highly discretionary and unpredictable" . Needless to say, this causes dissatisfaction to litigants, anguish for their advisers, uncertainty for judges, agitation amongst commentators and friction between healthcare professionals and their legal counterparts .

There are no easy solutions to these problems. This is apparent from the many cases concerned with causation in the context of medical negligence coming before final and other courts of appeal in England , Canada , the United States of America and Australia . It is further illustrated by the division of opinions in this case: Gaudron J and Gummow J favouring the dismissal of the appeal; McHugh J and Hayne J being in favour of allowing it. I agree with the remarks of my colleagues that the case is a difficult one involving an unusual chain of events. But, it is not unique. Other cases exist which bear certain similarities . Whilst avoiding the dangers of endless theoretical argument and acknowledging the disputability of a result depending upon the drawing of lines which fix the outer perimeter of legal liability , this Court must endeavour to give guidance in this case as to the approach to be taken when problems of this kind arise in the future, as surely they will. A patient is not warned and suffers damage

By the time this appeal from the New South Wales Court of Appeal reaches its conclusion in this Court more than fifteen years will have passed since the surgery on 10 June 1983 which gives rise to it. Mrs Beryl Hart (the respondent) underwent an operation performed by Dr Clive Chappel (the appellant). He was, and is, a medical practitioner and an ear, nose and throat specialist. The purpose of the operation, from Mrs Hart's point of view, was to relieve a long period of difficulty she had experienced in swallowing, eating and digestion, as well as with soreness of the throat. Radiological examination revealed pharyngeal diverticular and associated narrowing of the adjacent oesophagus. Dr Chappel suspected the presence of a pharyngeal pouch in which food could become caught. He proposed a procedure known as a Dohlman's operation. The hospital records state that Dr Chappel reported: "dilated pharyngeal pouch/oesophageal wall to find with difficulty, very thick, nasogastric tube placed, contents acidic on litmus testing. Operation performed with oesophageal dilation and Dohlman's endoscopic division of pharyngeal pouch."

Unfortunately, the operative procedure perforated Mrs Hart's oesophagus. This set in train the escape of an infection (mediastinitis) which, in turn, compromised one of her laryngeal nerves. This, in its turn, severely affected her voice. It resulted in her premature retirement from a position as Principal Education Officer. Mrs Hart sued Dr Chappel for negligence and breach of contract. At the trial in the Supreme Court of New South Wales, Donovan AJ upheld her claim. He awarded her $172,500.61 damages . Dr Chappel and Mrs Hart both appealed to the Court of Appeal. He contended that no damages should have been awarded. She argued that the damages were inadequate. The Court of Appeal dismissed both appeals. Dr Chappel now appeals to this Court. Although Mrs Hart filed a notice of contention, supporting the judgment on the assumption that her entitlement to damages was (as Dr Chappel belatedly contended) to be assessed as a case of loss of a chance of a successful operation, no cross-appeal was filed by her. In this way the issues were reduced to Mrs Hart's entitlement to damages and, if so, how the damages should be calculated. Common ground

The issues in the appeal were even further refined before this Court: 1. Mrs Hart's claim against Dr Chappel was limited to a complaint that he had failed to warn her adequately, or at all, of the dangers involved in the operation: specifically, that there was a danger that her voice could be compromised by the complications which, in fact, occurred. A claim that Dr Chappel had conducted the operation negligently, although initially pleaded, was not supported by evidence and was abandoned at the trial. 2. Although originally strongly contested, Dr Chappel (for the purpose of the appeal) accepted (as the primary judge had found), that when asking about the risks prior to the operation, Mrs Hart had said to him words to the effect: "I don't want to wind up like Neville Wran". This remark was taken to be an allusion to a contemporaneous problem which, following operation, the then Premier of New South Wales (Mr N K Wran) had experienced with his voice which had only been partly restored by a teflon injection to his vocal cords. After the subject operation, Mrs Hart came under the care of Professor B N Benjamin. In treating the damaged laryngeal nerve to allow her improved use of the vocal cords he actually injected teflon. However, this procedure left Mrs Hart's voice weak and affected, much as Mr Wran's voice had been. Dr Chappel fought this appeal on the footing that he had failed properly to respond to his patient's inquiry. To that extent he was in breach of his duty to provide information to his patient which this Court's decision in Rogers v Whitaker required him to give. 3. The aetiology of the damage to Mrs Hart's laryngeal nerve was not in doubt. It required the coincidence of three events: (1) the operative tear to the oesophagus; (2) an escape of bacteria from the oesophagus; and (3) consequential impingement of the resulting infection upon the nearby right vocal cord causing paralysis and damage. Each of these preconditions was accepted to be very rare. A tear could occur (according to Professor Benjamin's evidence) once in every 20, 30 or 40 operations. Usually, it resulted in nothing more than the "escape of a few bubbles of air". The complication of mediastinitis that occurred in this case was "very rare indeed". It had not occurred in the 100 to 150 operations performed by Professor Benjamin. However, it was a recognised possibility. Once a patient asked a question about that possibility, he or she was entitled to have an accurate and candid answer so that the patient could make an informed decision about the surgery. For Mrs Hart, the consequences were important and they were large. 4. The condition which originally took Mrs Hart to Dr Chappel was "relentlessly progressive". Surgery was the "only relief" for it. Without surgery there would not only be soreness and difficulty in swallowing but the danger that food might become caught in the throat needing emergency attention. It was therefore accepted that, even if Mrs Hart had been warned of the danger of damage to her voice, she would eventually have undergone an operation on her throat. In any such operation the slight risk would exist of the kind that followed Dr Chappel's procedure. Mrs Hart did not dispute this. Dr Chappel conceded that, if the surgery had in fact been postponed and carried out at a different time, "[i]n all likelihood [Mrs Hart] would not have suffered the random chance of injury" to her vocal cord. This represented nothing more than acceptance that such injury was an extremely rare occurrence. It was not even mentioned in some clinical textbooks. 5. Mrs Hart swore that if she had been told by Dr Chappel of the risks to her voice she would not have gone ahead with the operation by him. She would have sought further advice. She would have wanted the operation performed by the most experienced person available. Professor Benjamin was posited as such a person. The evidence showed that he had performed many more operations of this kind than Dr Chappel had. The primary judge accepted that Mrs Hart was a witness of truth. Her claim must therefore be assessed on the footing that, with the warning that the law required Dr Chappel to give her, she would not have gone ahead with the operation when she did. She would thus not in fact have suffered the damage which ensued.

Dr Chappel contended that, in the foregoing facts, Mrs Hart was not entitled to recovery. The random chance of complications could just as easily have struck during an operation at a later time and place and conducted by a different surgeon. In the absence of proof of negligence in the performance of the operation, his accepted failure to warn Mrs Hart had not caused her damage. Mrs Hart, armed with the decisions below, contended that she had established sufficient facts to demonstrate a causal connection and to retain her damages. Causation: general legal propositions

To answer the problem presented by the appeal, it is useful to collect a number of propositions, established by authority, relevant to a case such as the present: 1 . A practical question : The starting point is to remember the purpose for which causation is being explored. It is a legal purpose for the assignment of liability to one person to pay damages to another. It is not to engage in philosophical or scientific debate, still less casuistry . As Windeyer J explained in The National Insurance Co of New Zealand Ltd v Espagne : "Philosophy and science seek the explanation of phenomena and look to relationships and concurrences. Law is not concerned rerum cognoscere causas , but with attributing responsibility to persons." The law allocates responsibility by a process which at once determines the entitlement of the particular plaintiff and sets the standards of conduct that may be expected of other persons in positions analogous to the defendant. The law's concern is entirely practical . "In the varied web of affairs, the law" said Lord Wright, "must abstract some consequences as relevant, not perhaps on the grounds of pure logic but simply for practical reasons" . Where a breach of duty and loss are proved, it is natural enough for a court to feel reluctant to send the person harmed (in this case a patient) away empty handed . However, such reluctance must be overcome where legal principle requires it. It must be so not only out of fairness to the defendant but also because, otherwise, a false standard of liability will be fixed which may have undesirable professional and social consequences . 2. A commonsense approach : Causation is essentially a question of fact . It is to be resolved as a matter of commonsense . This means that there is usually a large element of intuition in deciding such questions which may be insusceptible to detailed and analytical justification. As Dixon CJ, Fullagar and Kitto JJ remarked in Fitzgerald v Penn "it is all ultimately a matter of common sense" and "[i]n truth the conception in question [ie causation] is not susceptible of reduction to a satisfactory formula". Similarly, in Alphacell Ltd v Woodward , Lord Salmon observed that causation is "essentially a practical question of fact which can best be answered by ordinary common sense rather than by abstract metaphysical theory." Yet, a losing party has a right to know why it has lost and should not have its objections brushed aside with a reference to "commonsense", at best an uncertain guide involving "subjective, unexpressed and undefined extra-legal values" varying from one decision-maker to another. Nevertheless, despite its obvious defects, the commonsense test has been embraced by this Court as a reminder that a "robust and pragmatic approach" to such questions is the one most congenial to the common law. 3. The "but for" consideration : If, but for the negligent act or omission, the actual damage suffered by a plaintiff would not have occurred, it will often be possible, as a practical matter, to conclude the issue of causation in the plaintiff's favour. Similarly, where the damage would probably have happened anyway, it will often be possible to conclude that the act or omission was not the cause for legal purposes . In this sense, the "but for" test, so qualified, remains a relevant criterion for determining whether the breach of duty demonstrated is a cause of the plaintiff's damage . However, it is not the exclusive test. Nor is it sufficient on its own to demonstrate the causal link for legal purposes . It is a mistake to read this Court's cautionary words about the "but for" test as an expulsion of that notion from consideration where the question of causation is in contest. On the contrary, a sufficient causal connection will, generally speaking, be established if it appears that the plaintiff would not have suffered the damage complained of but for the defendant's breach of duty. The Court has simply added the warning that it is necessary to temper the results thereby produced with "value judgments" and "policy considerations". This qualification has been expressed lest a party, shown to have been in breach of duty, is forever thereafter to be liable for every misfortune that follows in time whatever the breach demonstrated and however irrelevant it may appear to the damage which ensued. As Windeyer J observed in Faulkner v Keffalinos : "But for the first accident, the [plaintiff] might still have been employed by the [defendants], and therefore not where he was when the second accident happened: but lawyers must eschew this kind of 'but for' or sine qua non reasoning about cause and consequence." In an attempt to assist decision-makers in the task of drawing lines and in the assignment of legal responsibility, various phrases have been proffered by generations of judges to mark out a legally relevant cause (such as "proximate cause", "legal cause", "true cause", "effective cause", "substantial cause", "direct cause", "foreseeable cause" or "cause in fact") . These phrases, whilst well-intentioned, beg the question that is to be answered. They also carry dangers of their own . So does the attempt to convert the inquiry, as McHugh J has suggested, from the passive to the active voice as if this will solve the quandary of causation. That quandary remains, however it is expressed in verbal formulation. 4. The plaintiff's legal onus : It is elementary to say that it is a pre-condition to recovery of damages for an established breach of a legal duty that the onus is upon the plaintiff to prove that the breach alleged was the cause of the damage shown. It is important to keep separate the questions of liability and the calculation of damages. Where, as in this case, a plaintiff relies on a claim in contract, proof of breach of that contract will entitle the plaintiff to nominal damages at least . For recovery of compensation beyond nominal damages in contract, the plaintiff must prove that the breach was the cause of the damage. This is as true of a claim based on the tort of negligence as of one framed in contract . In this sense, the legal burden of proving causation is, and remains throughout the proceedings, upon the plaintiff. It is not an insubstantial burden. In some medical contexts it has even been described as Herculean . In cases similar to the present, it has been characterised as "the most formidable obstacle confronting health care consumers" . The reasons include the imprecision of, and uncertainty about, some medical conditions; the progressive nature of others; the complexity of modern medical practice and technology; and the fact that some mistakes, serious enough in themselves, have no untoward results which can properly be attributed to them. In the present case, Dr Chappel argued that he fell into the last stated class of exemption. The recognised difficulties of causation for plaintiffs in medical negligence cases have occasionally given rise to legal devices designed to lighten their burdens . Some of these will be mentioned below. 5. Displacing apparent causation : In certain circumstances, the appearance that there is a causal connection between the breach and the damage, arising from the application of the "but for" test and the proximity of the happening of the damage, has been displaced by a demonstration that: (a) The happening of the damage was purely coincidental and had no more than a time connection with the breach ; (b) The damage was inevitable and would probably have occurred even without the breach, for example by the natural progression of an undetected, undiagnosed or unrevealed condition , or because the condition presented a life threatening emergency which demanded instant responses without time for the usual warnings and consents ; (c) The event was logically irrelevant to the actual damage which occurred ; (d) The event was the immediate result of unreasonable action on the part of the plaintiff ; or (e) The event was ineffective as a cause of the damage, given that the event which occurred would probably have occurred in the same way even had the breach not happened . 6. Reinforcing the duty to warn : In judging the performance of a health care or other professional, the law does not require perfection. It recognises the variability of professional skills. Even an expert, acting at the highest standards of the profession, may turn in a less than perfect performance on a particular day. However, the requirement to warn patients about the risks of medical procedures is an important one conducive to respect for the integrity of the patient and better health care. In Australia, it is rigorous legal obligation . Its rigour was not challenged in this appeal. It must be accepted that, by establishing the requirement to warn patients of a risk to which they would be likely to attach significance, or of which they should reasonably be aware, the law intends that its obligations be carefully observed. Breaches must be treated seriously. Because in some cases the failure to warn would have no, or no relevant, consequences, proof of a breach will not of itself be sufficient to establish an entitlement to damages for every harm that thereafter occurs to the patient. To reason in such a way would involve the logical fallacy of post hoc ergo propter hoc . The plaintiff's legal obligation to show the causal connection remains throughout the proceedings. 7. Accepting subjective intentions : In considering the suggested consequences of a failure on the part of a medical practitioner to advise a patient about the risks of a particular procedure, courts in Australia have adopted a "subjective" approach which has regard to what the particular patient's response would have been had proper information been given . A contrary ("objective") approach, having regard to the response of a reasonable person in the patient's situation, was not urged in this case, although it has found favour in Canada and the United States of America . The subjective criterion involves the danger of the "malleability of the recollection" even of an upright witness . Once a disaster has occurred, it would be rare, at least where litigation has commenced, that a patient would not be persuaded, in his or her own mind, that a failure to warn had significant consequences for undertaking the medical procedure at all (where it was elective) or for postponing it and getting a more experienced surgeon (as in this case). Yet, these dangers should not be over-stated. Tribunals of fact can be trusted to reject absurd, self-interested assertions. Where such a conclusion is reached the case will rarely come before an appellate court. The present appeal must be approached on the footing accepted by the primary judge. This was that, if she had been warned, Mrs Hart would not have had the operation, not have suffered the physical injuries which then ensued and would have sought a more experienced surgeon when the time for operation eventually came. 8. Shifting the evidentiary onus : One means of alleviating the burden cast by law on a plaintiff to establish a causal relationship between the breach and the damage concerns the evidentiary onus. Australian law has not embraced the theory that the legal onus of proof shifts during a trial . Nevertheless, the realistic appreciation of the imprecision and uncertainty of causation in many cases - including those involving alleged medical negligence - has driven courts in this country, as in England, to accept that the evidentiary onus may shift during the hearing. Once a plaintiff demonstrates that a breach of duty has occurred which is closely followed by damage, a prima facie causal connection will have been established . It is then for the defendant to show, by evidence and argument, that the patient should not recover damages. In McGhee v National Coal Board , a Scottish appeal, Lord Wilberforce explained why this was so. Although Lord Wilberforce's statement in McGhee has proved controversial in England , it has received support in this Court . Its principle has also been accepted by international experts such as Professor Giesen. I find Lord Wilberforce's exposition compelling : "[T]he question remains whether a pursuer must necessarily fail if, after he has shown a breach of duty, involving an increase of risk of disease, he cannot positively prove that this increase of risk caused or materially contributed to the disease while his employers cannot positively prove the contrary. In this intermediate case there is an appearance of logic in the view that the pursuer, on whom the onus lies, should fail - a logic which dictated the judgments below. The question is whether we should be satisfied, in factual situations like the present, with this logical approach. In my opinion, there are further considerations of importance. First, it is a sound principle that where a person has, by breach of a duty of care, created a risk, and injury occurs within the area of that risk, the loss should be borne by him unless he shows that it had some other cause. Secondly, from the evidential point of view, one may ask, why should a man who is able to show that his employer should have taken certain precautions, because without them there is a risk, or an added risk, of injury or disease, and who in fact sustains exactly that injury or disease, have to assume the burden of proving more: namely, that it was the addition to the risk, caused by the breach of duty, which caused or materially contributed to the injury? In many cases ... this is impossible to prove, just because honest medical opinion cannot segregate the causes of an illness between compound causes. And if one asks which of the parties, the workman or the employers, should suffer from this inherent evidential difficulty, the answer as a matter of policy or justice should be that it is the creator of the risk who, ex hypothesi must be taken to have foreseen the possibility of damage, who should bear its consequences". 9. Valuing a lost chance : A further way in which, in some circumstances, the difficulties of causation for a plaintiff are alleviated is by treating the plaintiff's loss as a "loss of a chance". In cases in which this approach is permissible, it may allow evaluation of the plaintiff's loss in terms of comparing the chances of suffering harm (given the breach which has occurred) against those that would have existed (if the breach is hypothesised away). In CES v Superclinics (Aust) Pty Ltd I indicated my attraction to this approach as a more rational and just way of calculating damages caused by established medical negligence. It is clearly laid down by the authority of this Court that, in some circumstances, a plaintiff may recover the value of a loss of a chance caused by a wrongdoer's act or omission . The approach also has some judicial support in the context of medical negligence in England , Canada and the United States . A number of commentators favour this approach because of the failure of orthodox reasoning to do justice to some patients' losses and because it invites a more empirical calculation of loss, with the use of statistics which might offer outcomes that are more accurate and fair to all concerned . On the other hand, the weight of judicial opinion in England and Canada and some academic writing appears to be critical of the application of the loss of a chance theory to cases of medical negligence. In part this is because, where medical negligence is alleged, "destiny...[has] taken its course", arguably making an analysis by reference to chance inappropriate or unnecessary in the view of the critics of this approach . Alternatively, the loss of a chance calculation has been criticised on the ground that it would discard commonsense, undermine the plaintiff's onus of proving the case and submit the law to the "paralysis" of statistical abstractions. 10. Discounting damages : If it is established that damage was caused by the breach alleged, it remains to calculate the amount of compensation recoverable. It is then proper to reduce any damages which a defendant should pay for the harm it has caused to a proper proportion actually attributable to its breach . If, independently of the breach on the part of a defendant, the evidence shows that the plaintiff would have suffered loss, the damages may be reduced by reference to the estimate of the chances that this would have occurred. If those chances are less than one percent, this Court has held that they may properly be disregarded as speculative . Dr Chappel argued that, even if he had given the requisite warning to Mrs Hart, and she had postponed the procedure and later undergone an operation by a more experienced surgeon, there was still the same random chance that she would have suffered the complications that occurred; neither more nor less. Mrs Hart argued that the true comparison was between the loss that had in fact occurred to her and the concededly small risk that such loss would have happened at the postulated postponed operation. She resisted any reduction in her damages, submitting that a chance of injury in a postponed operation was minuscule, ie "speculative" in the sense described by this Court. Conclusion: causation was established

The application of the foregoing principles to the facts of this case, as now established, presents difficult puzzles upon which reasonable minds may differ; as indeed they have. The strongest arguments for Dr Chappel, as it seems to me, are those which lay emphasis upon a logical examination of the consequences which would have flowed had he not breached his duty to warn his patient. Dissecting the facts in that way affords a powerful argument which would banish from consideration the events which in fact occurred in the operation which he carried out. All that would have happened, had he given the requisite warning, would have been a change in the timing of the operation and of the identity of the surgeon. For Dr Chappel, these were irrelevant changes as the evidence showed that, whenever the operation was performed and whoever did it, the tripartite chances which had to combine to produce the misfortune which Mrs Hart suffered were extremely rare. There was thus an equivalence of unlikelihood. They were risks inherent in the procedure, not wholly avoidable even by the most skilful and experienced of surgeons. In the view which Dr Chappel urged of the case, Mrs Hart was left with nothing more than the time sequence. To burden a surgeon, in whose actual performance no fault could be found, with civil liability for randomised chance events that followed the surgery would not be reasonable. It would penalise him for chance alone. It would do nothing to establish a superior standard in the performance of the work of surgeons generally.

For a time I was attracted to Dr Chappel's arguments. Ultimately, I have concluded against them. The "commonsense" which guides courts in this area of discourse supports Mrs Hart's recovery. So does the setting of standards which uphold the importance of the legal duty that was breached here . This is the duty which all health care professionals in the position of Dr Chappel must observe: the duty of informing patients about risks, answering their questions candidly and respecting their rights, including (where they so choose) to postpone medical procedures and to go elsewhere for treatment.

In Environment Agency (formerly National Rivers Authority) v Empress Car Co (Abertillery) Ltd , Lord Hoffmann emphasised that commonsense answers to questions of causation will differ according to the purpose for which the question is asked. The answer depends upon the purpose and scope of the rule by which responsibility is being attributed. In Rogers v Whitaker , this Court decided that "a doctor has a duty to warn a patient of a material risk inherent in the proposed treatment" and that: "a risk is material if, in the circumstances of the particular case, a reasonable person in the patient's position, if warned of the risk, would be likely to attach significance to it or if the medical practitioner is or should reasonably be aware that the particular patient, if warned of the risk, would be likely to attach significance to it." These standards have fairly been described as onerous. They are. But they are the law. They are established for good reason. When not complied with (as was held to be so in this case) it should occasion no surprise that legal consequences follow. This was an unusual case where the patient was found to have made very clear her concerns. The practicalities are that, had those concerns been met as the law required, the overwhelming likelihood is that the patient would not, in fact, have been injured. So much was eventually conceded. In such circumstances, commonsense reinforces the attribution of legal liability. It is true to say that the inherent risks of injury from rare and random causes arise in every surgical procedure. A patient, duly warned about such risks, must accept them and their consequences. Mrs Hart was ready to accept any general risks of the operation of which she was warned. However, she declined to bear the risks about which she questioned the surgeon and received no adequate response. When those risks so quickly eventuated, commonsense suggests that something more than a mere coincidence or irrelevant cause has intervened. This impression is reinforced once it is accepted that Mrs Hart, if warned, would not have undergone the operation when she did.

Although no statistical or other evidence was called to demonstrate that recourse to a more experienced surgeon would necessarily have reduced the risk of the kind of injury that occurred (and while some risk was unavoidable), intuition and commonsense suggest that the higher the skill of the surgeon, the less is the risk of any perforation of the oesophagus into the mediastinum. In 100 to 150 operations of this kind, Professor Benjamin had never experienced mediastinitis. Whilst that may indeed be the result of chance and amount to good luck on his part (and on the part of his patients) intuition and commonsense suggest that the greater the skill and more frequent the performance, the less the risk of perforation. And without perforation (already a rare occurrence) the second and third events necessary to produce paralysis of the vocal cords in a patient like Mrs Hart (occurrences even more rare) would not occur. As Gaudron J points out, the nature of the risk would be the same. But the degree of risk would be diminished. This was the view taken by the Court of Appeal . It is a view which involved no error.

Once Mrs Hart showed the breach and the damage which had immediately eventuated, an evidentiary onus lay upon Dr Chappel to displace the inference of causation which thereupon arose. He failed to do so. Nor, in my view, causation being established, did he prove that Mrs Hart would have been exposed to the same, or substantially the same, possibilities of like injury if she had postponed the procedure and had it done by someone more experienced, as was her right. On the contrary, the evidence demonstrated that the chances of her receiving such injury in any other operation were minuscule. For the reasons stated those chances would probably be even smaller in the hands of a surgeon with the experience and skill of Professor Benjamin.

To the complaint that Professor Benjamin (or his equivalent) could not possibly undertake every Dohlman's operation (any more than the most skilful barrister can appear for every client) the answer comes back. This was not an ordinary patient. It was an inquisitive, persistent and anxious one who was found to have asked a particular question to which she received no proper answer. Had a proper answer been given, as the law required, it was found that she would not have undergone the operation at the hands of Dr Chappel when she did. It is virtually certain, then, that she would not have suffered mediastinitis at all. She would not have been injured. She would not have been obliged to bring her case before the courts. She therefore adequately proved causation. Dr Chappel did not displace the inferences to which her evidence gave rise. Nor was it shown that the damages to which she was entitled should be reduced on the footing that they would have occurred in any event.

As to the question of loss of a chance, Dr Chappel, by leave, added a ground of appeal to assert that Mrs Hart's damages should have been assessed in those terms . Mrs Hart resisted the amendment but, in any case, said that it mattered not . At trial, the only claim for damages, which she had asserted, was in respect of the physical injury done to her vocal cords and its sequelae. She neither pleaded, nor sought to prove, a case expressed in terms of a loss of a chance. Accordingly, no evidence was tendered as to the value of that chance. The case is therefore not one in which an entirely new perspective should be adopted at such a late stage. One day loss of a chance in this area of discourse will return to this Court. However, this case must be approached on the footing that the loss suffered by Mrs Hart was that claimed: physical injury and its consequence - nothing more. Order

The appeal should be dismissed with costs.

HAYNE J. On 10 June 1983, the appellant, an ear nose and throat specialist, performed an operation on the respondent. The respondent, who was then employed by the New South Wales Department of Education as a Principal Education Officer - Library Services, had first been referred to the appellant in April that year because she had a persistent sore throat and was experiencing difficulty swallowing. In May 1983, the appellant diagnosed a pharyngeal pouch in the oesophagus - a pouch in the oesophagus in which food could be caught. The appellant recommended to the respondent that she have surgery and explained to her that the procedure could be undertaken through the mouth or through the neck. She chose the former method and the procedure (a Dohlman's operation using a rigid endoscope) was carried out at Mona Vale Hospital on 10 June 1983.

The respondent's recovery did not proceed as would be expected and she was transferred to Royal North Shore Hospital. There, she came under the care of other doctors. She was told that her oesophagus had been perforated in the operation. Infection set in and it became clear that one of her laryngeal nerves (the right recurrent laryngeal nerve) was damaged. This affected her voice despite the steps that were taken by another consultant ear nose and throat specialist (Professor Benjamin) to treat the damaged nerve and allow her proper use of her vocal cords. As a result of this treatment (first an injection of a gel foam paste and later a teflon injection of the vocal cords) her voice improved but she was left with a weak and husky voice, the use of which tired her.

She continued to experience difficulty in swallowing and in February 1985 had to have a grape which had lodged in her throat removed under general anaesthetic. In June 1985, Professor Benjamin undertook another procedure to treat the pharyngeal pouch. This procedure resolved that problem but left her voice as it had been after the first operation.

The respondent felt that the problems with her voice prevented her performing her work properly and in August 1985 she retired, having been assessed as medically unfit to continue in her employment.

In 1989 she began an action against the appellant in the Supreme Court of New South Wales claiming damages. The action was framed in breach of contract and in negligence. She alleged that the appellant had failed to warn her of the risks associated with the procedure and had performed the procedure negligently and that in consequence of that negligence or breach of contract had sustained injuries, namely a perforated oesophagus with "consequent division of one of the laryngeal nerves and paralysis of the right vocal cord" and had suffered loss and damage being out of pocket expenses for medical and hospital expenses and economic loss because she was unable to continue her work.

The primary judge (Donovan AJ) gave judgment for the plaintiff for damages including $30,000 general damages, $72,581.96 for past economic loss and $35,000 for the past and future economic loss of not being able to take engagements as a consultant in the years after she would have been bound to retire on account of age. An appeal against the judgment, and a cross appeal about assessment of damages, were both dismissed by the Court of Appeal. The appellant now appeals by special leave.

At trial, the respondent's claim that the appellant conducted the original procedure negligently was not pressed and no evidence was led in support of that claim. The respondent limited her claim to an allegation that the appellant had failed to warn her of the risks of the operation. The primary judge found (and it is not now disputed) that in the course of the consultation with the appellant in May 1983 the respondent told him that "I don't want to wind up like Neville Wran" - a comment which the respondent described in her evidence as being a "throw-away line" but which was taken by the primary judge to be a clear indication by the respondent to the appellant of concern for the safety of her voice. It was found that the appellant did warn the respondent that there was a risk of perforating the oesophagus in the course of the procedure but he did not warn her that the operation posed a risk to her voice. The primary judge found (and again this is not now disputed) that in the light of her reference to Neville Wran the risk to the respondent's voice was a material risk of which she should have been warned . The risk was slight but if it was realised, the consequences for the respondent were large.

The central question debated on the hearing of the appeal to this Court was the question of causation. Did the appellant's failure to warn of the risk to the respondent's voice cause all or any part of the loss she claimed?

There are several features of the case which it will be necessary to bear steadily in mind. 1. The appellant's breach of duty was his failure to warn the respondent of the risk to her voice; he performed the procedure concerned without negligence. In particular, to perforate the oesophagus in the course of the procedure, while not intended, was not negligent. 2. The respondent's condition was one which could be treated only by surgery and her condition was one the symptoms of which, it was accepted, were correctly described by Professor Benjamin as being "relentlessly progressive". 3. Although perforation of the oesophagus in the course of a procedure like that performed by the appellant on the respondent was not uncommon (according to Professor Benjamin a perforation might occur once in every twenty or thirty or forty operations) it is very rare for that then to be complicated by infection of the mediastinum (mediastinitis). According to the Professor, the compromise of the recurrent laryngeal nerve, which the respondent suffered, may have been caused by operative trauma, intubation trauma during anaesthesia or spread of infection from the mediastinitis. In his report prepared in May 1994 and tendered in evidence at trial, Professor Benjamin said: "With respect to the paralysis of the right vocal cord and the cause, it is very difficult to say which is the most likely idiological factor. Vocal cord paralysis during or after an uncomplicated or a complicated endoscopic operation for pharyngeal pouch is very uncommon. However in view of the mediastinitis and the long term nature of the paralysis perhaps it is likely the paralysis was associated with perforation of the oesophagus and mediastinal infection." 4. Because the pharyngeal pouch could be treated only by surgery and because the respondent's symptoms would have grown worse over time, it was accepted that she would have had an operation of the kind she did have at some time. 5. The respondent swore that, if the appellant had told her of the risks to her voice, she would not have had the operation when she did but would have sought further advice because she would have wanted the operation performed by the most experienced person with a record and reputation in the field. (The primary judge accepted the respondent as a witness of truth.) There was some evidence to suggest that the better the surgeon, the less likely was there to be a perforation of the oesophagus.

The elementary proposition that a defendant is liable in negligence only if the damage suffered by the plaintiff was caused by the defendant's negligent act or omission identifies the connection between the defendant's act or omission and the plaintiff's damage as that of causation. As is said in Bennett v Minister of Community Welfare : "In the realm of negligence, causation is essentially a question of fact, to be resolved as a matter of common sense . In resolving that question, the 'but for' test, applied as a negative criterion of causation, has an important role to play but it is not a comprehensive and exclusive test of causation; value judgments and policy considerations necessarily intrude ." The resolution of that question will often find expression in an assertion of its result without any lengthy articulation of reasons. Especially would that be so in a case where policy considerations do not assume prominence in the process.

In this case, however, it is as well to try to identify the process of reasoning that is adopted.

The search for causal connection between damage and negligent act or omission requires consideration of the events that have happened and what would have happened if there had been no negligent act or omission. It is only by comparing these two sets of facts (one actual and one hypothetical) that the influence or effect of the negligent act or omission can be judged.

If the damage of which the plaintiff complains would have happened without the intervention of the negligent behaviour, it will often be possible to conclude that the negligent behaviour was not a cause of that damage. Thus, to take examples cited in Prosser and Keeton on the Law of Torts : - a failure to fence a hole in the ice plays no part in causing the death of runaway horses which could not have been halted if the fence had been there ; - a failure to have a lifeboat ready is not a cause of the death of a person who sinks without trace immediately upon falling into the ocean ; - the omission of crossing signals by an approaching train is of no significance when a car driver runs into the sixty-eighth car in the line .

If, however, the damage of which the plaintiff complains would not have happened without the intervention of the negligent behaviour, it will often be possible to conclude that the negligent behaviour was a cause of that damage. Thus, the plaintiff in Rogers v Whitaker would not have had surgery on her blind eye if she had been warned of the risk that the operation posed to her good eye. The negligent failure to warn her of that risk was held to be a cause of her damage.

The "but for" test is, however, neither a comprehensive nor exclusive test of causation . To take but one example where its application is not conclusive, it does not readily resolve the case where two causes are at work and either of them, alone, would have been sufficient to bring about the result. If two separate fires, negligently lit by separate persons, merge to destroy the plaintiff's home, and each fire would have been sufficient in itself to cause the damage, is each of the fire lighters liable ? If the "but for" test were to be applied to each defendant's conduct separately then neither would be liable. And what if neither fire, by itself, would have destroyed the plaintiff's house ?

The "but for" test is of most use as a negative test. If it is not satisfied, it is unlikely that there is the necessary causal connection. But showing that "but for" the defendant's conduct, the plaintiff would not have suffered damage does not demonstrate the required degree of connection between the defendant's act or omission and the plaintiff's damage. The application of a "but for" test does not identify what might be called the "quality" of the causal connection. No doubt it is with this in mind, that the cases and literature use many different epithets to describe the kind of causation that is necessary - "proximate cause", "legal cause" and so on - as opposed to "causation in fact" . (No doubt also, those epithets will sometimes reflect the value judgments or policy considerations mentioned in cases like March v Stramare (E & M H) Pty Ltd .)

The importance of examining the nature of the connection between the negligent conduct and the damage can be demonstrated in this way. If the respondent had not been operated on when she was, but had had her operation on another day, the chances are that she would not have suffered the damage to her laryngeal nerve that she did. There may have been no perforation of the oesophagus, there may have been no infection, there may have been no damage to the nerve. The whole tenor of the evidence given at the trial was that if it was the infection that led to paralysis of the laryngeal nerve (and this was the explanation favoured by Professor Benjamin in his written report) infection was such a rare event that it was unlikely (indeed very unlikely) that it would have happened if the operation had been performed on another day. Of course, the respondent did suffer a perforated oesophagus, she did suffer an infection, she did suffer paralysis of the laryngeal nerve. But if she had not attended the hospital on that day, the probabilities are that none of this would have happened. And if the appellant had told her of the risk to her voice, she would not have had the operation when she did. But precisely the same argument would be open if, instead of suffering damage to her voice, as she has, the operating theatre in which her procedure was performed had been struck by lightning, or a runaway truck, and she had been injured. But for the negligent failure to warn she would not have been in harm's way.

No doubt the case of the lightning strike or the runaway truck invite consideration of novus actus interveniens and whether, although "the earlier wrongful act or omission may have amounted to an essential condition of the occurrence of the ultimate harm, it was not the true cause or a true cause of that harm" . But that is no more than a particular example of the general proposition that the tort of negligence requires a particular kind of causal relationship between the negligent act or omission of the defendant and the damage suffered by the plaintiff.

Being able to say that the damage would not have happened but for the negligent act or omission is not enough. As Windeyer J said in Faulkner v Keffalinos (where the plaintiff had been injured in a car accident, and then injured in a separate car accident before the trial) : "The consequences that flow from the second accident cannot I think be regarded as caused, in any relevant sense, by the defendants' tort. I realise that philosophers and casuists may see these as indirect consequences. But for the first accident, the respondent might still have been employed by the appellants, and therefore not where he was when the second accident happened: but lawyers must eschew this kind of 'but for' or sine qua non reasoning about cause and consequence."

In my view, the only connection between the failure to warn and the harm the respondent has suffered is that but for the failure to warn she would not have been in harm's way. The appellant's conduct did not affect whether there would be pathogens present in the respondent's oesophagus when the procedure was carried out; his conduct did not affect whether the pathogens that were present would, in all the circumstances, produce the infection which they did; his conduct did not affect whether that infection would damage the laryngeal nerve as it did. Of course, he manipulated the instrument which perforated the oesophagus but he did so without negligence.

I should mention the recent decision of the House of Lords in Environment Agency (formerly National Rivers Authority) v Empress Car Co (Abertillery) Ltd . Lord Hoffmann, who gave the leading speech, said that : "... common sense answers to questions of causation will differ according to the purpose for which the question is asked" and that "... one cannot give a common sense answer to a question of causation for the purpose of attributing responsibility under some rule without knowing the purpose and scope of that rule". So much may be accepted. But consideration of the purposes of asking about causation in a case like the present should not be permitted to obscure the fact that the search is for a relationship between the negligent act or omission of one party and the damage which the other party alleges has been sustained.

The law of negligence may be seen as directed to several purposes but purposes of compensating the injured and promoting reasonable conduct are prominent among them. In this particular area of negligent advice by a medical practitioner it is important to bear in mind "the paramount consideration that a person is entitled to make his own decisions about his life" .

With these purposes in mind, it may be suggested that a sufficient causal relationship is established by showing that the subject-matter of the negligent conduct - a failure to warn of risk to the voice - is the very subject-matter of the damage. But that connection is not enough. If it were enough, it would follow that if the operating theatre had been struck by lightning and the respondent had suffered damage to the laryngeal nerve (because of the resulting power surge affecting the diathermy equipment being used in the operation) the appellant would be liable but that he would not if the power surge caused burns to her body. Similarly, it would mean that the appellant would be liable if the respondent's voice were damaged as a result of an infection stemming from some failure of the hospital to sterilise, properly, instruments or other items used in the procedure.

No doubt the fact that what I have called the subject-matter of the negligent conduct and the subject-matter of the damage are the same is important to that intuitive process of analysis that is referred to when it is said that questions of causation are questions of fact to be resolved as a matter of commonsense. But important as this consideration is, it is not determinative.

Nor is it enough to say that a purpose of this area of the law is to promote reasonable conduct by medical practitioners and, particularly, the giving of advice necessary to enable people to make their own decisions about their lives. Enlarging the circumstances in which damages will be awarded if there has been a negligent failure by a medical practitioner to advise a patient of risks may well tend to promote the giving of fuller advice. So too may the imposition of a penalty for failing to give proper advice. But the ambit of the liability is not to be decided only according to whether enlarging that ambit will promote careful conduct. The question of causation must still be answered . What is the connection between the negligent act or omission and the damage sustained?

The difficulty in the analysis that looks only to whether the subject-matter of the negligent conduct (failure to warn of risk to voice) and the damage suffered (damage to the voice) are the same is that it does not pay sufficient heed to the comparison that the law requires between the facts of what happened and the hypothetical facts of what would have happened if there had been no negligent act or omission.

It was accepted in this case that, if the respondent had been given proper advice of the risks of the operation, she would, nevertheless, have had the operation which she did. She would have had it at a different time and may have had it performed by a different doctor but she would have had it done. Until she had the operation, she would have continued to suffer the discomforts and dangers that she was suffering when she consulted the appellant - persistent sore throat, difficulty in swallowing, a constant danger of food being caught in her throat. But the hypothetical situation that was to be considered was one in which the respondent had the operation in any event.

If she had had the operation at some later time and if she had engaged the appellant to perform it, the risk of her suffering the consequences to her voice that in fact befell her would, for all practical purposes, have been the same . If she had been given proper advice, even if she would have then deferred the operation, that would not have altered the risk that her voice would be affected (any more than it would have affected the risk that the operating theatre would be struck by lightning).

If, on being given proper advice, she would have deferred the operation, I would conclude that the respondent did suffer damage and would suffer damage because she did not defer the operation. But the damage she would suffer in those circumstances would not be the damage to her voice - it would be the loss of the period for which she would have deferred the operation and have had her voice and her job, subject nevertheless to the continuing disabilities of her untreated condition. Thus, if, because of the failure to warn, she had the operation (say) two years earlier than she otherwise would have had it and if the damage to her voice thus occurred two years earlier than it might have occurred in a later operation, she would have lost two years of employment and attendant enjoyment of life, discounted to take account of the disabilities she would have suffered during that period of two years. But the damage to her voice would not be caused by the failure to warn.

The respondent's claim focused upon the damage to her voice. The evidence that was led, and the arguments that were advanced on her behalf, were all directed to showing that the appellant's failure to warn caused the respondent the physical damage which she had suffered (the damage to the laryngeal nerve with consequent effects on her voice) and the economic consequences that were said to follow from that damage. No evidence was led to suggest that the respondent, if advised of the risks to her voice, would have deferred the operation for any significant period. She said that she would have sought "a second opinion ... perhaps several opinions" and no doubt this would have taken time but it was not suggested that she would then have put off the operation for some months let alone years. Thus no factual foundation was laid for a claim based upon delaying the operation.

It will be seen that the comparison I have drawn is between the times at which she would have confronted the risk about which she should have been warned. It is not a comparison that involves any prediction of whether that risk would have occurred if the operation had been deferred. That is because the operation has risks even if reasonable care is exercised; those risks cannot be eliminated by the exercise of reasonable care. It was not alleged in this case that the appellant performed the procedure negligently. That is, it was not alleged that the risk which the respondent faced in undergoing this operation could be eliminated if the surgeon was careful. This is not to deny that professional performance varies: that some surgeons are better than others. But the law is not concerned to do more than enforce standards of reasonable care. The respondent could ask no more than that the doctor she engaged to perform this procedure should exercise reasonable care in doing so, and the appellant did just that.

There was evidence that if she had been properly advised of the risks to her voice, the respondent would not have had the procedure performed by the appellant, but by another doctor. There was, as I have said, some evidence which suggested that the better the doctor, the less the chance of perforation of the oesophagus. That evidence was, at best, exiguous and stopped far short of identifying any sound basis for assessing what effect the surgeon's skills may have had on the unusual chain of events which happened in this case. Nevertheless, it was submitted that the evidence permitted the conclusion that the appellant's failure to give a proper warning of the risks deprived the respondent of a chance to seek better treatment, or exposed the respondent to a greater risk of injury than she faced in undergoing the procedure at the hands of the appellant.

I deal first with the contention that the failure to warn deprived the respondent of a chance to seek better treatment.

I do not think it necessary or appropriate to analyse this case as one of loss of a chance. There are several reasons why that is so.

First, the case was not put in this way at trial. The damage which the respondent alleged that she had suffered was the physical damage to her voice and the economic consequences of that damage. She did not seek to make any loss of chance case at trial.

Secondly, the chance which it is now said that the respondent lost is the chance to engage a better doctor. She said in evidence that she "would have wanted the most experienced person, with a record and reputation in the field". But it was never part of the respondent's case that the appellant should have told her to seek a better doctor; it was never suggested that there had been some negligent failure by the appellant to refer the respondent to another doctor. Moreover, it is important to bear steadily in mind that it was not said that the appellant had performed the procedure negligently. Thus it was never suggested that she was deprived of the opportunity to have the procedure performed properly - only that had she been advised of the risks to her voice she (of her own volition) would have sought out the "most experienced" practitioner in the field.

I do not consider that the appellant should be held responsible for the loss of that chance. No doubt it may be said that the failure to warn led to this result - in the sense that "but for" the negligent failure to advise, the respondent would have pursued the course that she described in her evidence - but why should the law provide for compensation for loss of that chance and what is it that she lost?

The law of negligence is intended to compensate those who are injured as a result of departures from standards of reasonable care. It is not intended to compensate those who have received reasonable care but who may not have had the best available care. To hold that the appellant's failure to warn the respondent of the risks of the operation caused her to lose the chance of the best available care would depart from that fundamental premise of the law of negligence.

Further, what is it that is lost when it is said that the respondent lost a chance of better treatment? It is said that by going to the best doctor in the field she could have reduced the chance of an adverse outcome of the operation. She could not, however, have eliminated those risks. How then is this alteration in the size of the risks to be measured and how is the loss of it to be compensated?

Leaving aside whatever may have been the difficulty of assembling evidence that bore upon the point (and those difficulties may have been very large) what kind of enquiry would have to be undertaken? Presumably the comparison to be made would be a comparison between the risks if the procedure was carried out by the appellant and the risks if the best available doctor carried it out. But how would that be measured? Any observer of skilled professionals at work knows that some are better than others but it is equally obvious that the performance of even the best is subject to variation. Is the comparison to be made a comparison with the best performer doing his or her best work? But how is that to be demonstrated? It is often enough difficult to identify what reasonable care requires; proof of what would be the best available care would be harder. And why should the law of negligence concern itself with more than what reasonable conduct would require?

Further, the risks of which we are speaking are risks that are very small. If the risk of disaster is assessed as being (say) 1 in 100 if the procedure is performed by the appellant but 1 in 200 if performed by another, what use is to be made of that data? If we are to speak in the language of loss of chance, has the respondent lost the chance of a 99.5 per cent chance of successful operation in return for a 99 per cent chance? Has she, that is, lost a 0.5 per cent chance of success? What is that worth? (The point is all the sharper if the comparison is between a 1 in 10,000 and a 1 in 20,000 chance.) Or is the relevant conclusion that the chances of disaster could have been halved?

Whichever description of the change in the risks is adopted, how does one assess the value of the chance that has been lost? It was suggested in the course of argument that it is reflected in the assessment of damages by discounting the damages otherwise allowed. But that invites attention to what are those damages that are to be discounted - is it, as the argument appeared to assume, the damages attributable to the physical consequences which the respondent suffered? That could be so only if the physical consequences which the respondent suffered were caused by the appellant's negligence.

All of these considerations point to the conclusion that the loss of chance analysis is flawed and should not be adopted. I therefore need not (and do not) express any view on the difficult questions that arise where a plaintiff claims damages for negligence, as opposed to contract, and contends that the damage suffered is the loss of a chance .

Much, if not all, of what I have said about the contention that the respondent lost a chance of better treatment applies equally to the mirror contention that she was exposed to greater risk.

I agree with McHugh J that there is insufficient evidence in this case to say, on the balance of probabilities, that the appellant's failure to warn exposed the respondent to greater risk of injury. The respondent would have had the operation at some time. The operation has risks even if performed by the most skilled surgeon available. There was very little evidence on the difference between the risk of injury actually faced by the respondent and the risk that she would have faced had the operation been performed by, say, Professor Benjamin.

I do not need to deal separately with the respondent's claim in contract. If the appellant's failure to warn the respondent of the risks of the operation was a breach of contract, for the reasons I have given earlier, I do not consider that that breach caused damage to her voice or caused her to lose a chance of better treatment or exposed her to greater risk.

I have said that the resolution of the question of causation will often be asserted without lengthy articulation of reasons. Since it is a question of fact resolved as a matter of commonsense and experience, the conclusion is often reached intuitively. The description of the steps involved in that kind of process is difficult and is apt to mislead. Articulating the reasoning will sometimes appear to give undue emphasis to particular considerations. No doubt if policy and value judgments are made, they should be identified. But the lengthy analysis which I have made should not be taken as intending to state any qualification upon the generality of the propositions recognised in cases like March v Stramare (E & M H) Pty Ltd . Causation is a question of fact to be resolved as a matter of commonsense. I have made the extended analysis which I have in order to draw out the various considerations which I consider bear upon the resolution of a difficult and unusual case, not because I consider that a trial judge should be expected (except, perhaps, in the most unusual case) to do more than record the conclusion that he or she reaches about whether the plaintiff's damage was caused by the defendant's negligence.

The respondent did not establish that she had suffered damage as a result of the appellant's negligence. The claim having been framed in breach of contract and breach having been established, she is, of course, entitled to nominal damages but, in my view, to no more. I would allow the appeal.